Skip to main content

Full text of "Valuation, its nature and laws : being an introduction to the general theory of value"

See other formats


Xibrarp  of  philosophy 

EDITED  BY  f.  H.  MUIRHEAD,  LL.D. 


THE  LIBRARY  OF  PHILOSOPHY. 


THE  LIBRARY  OF  PHILOSOPHY  is  in  the  first  instance  a 
contribution  to  the  History  of  Thought.  While  much  has  been 
done  in  England  in  tracing  the  course  of  evolution  in  nature, 
history,  religion  and  morality,  comparatively  little  has  been 
done  in  tracing  the  development  of  Thought  upon  these  and 
kindred  subjects,  and  yet  "  the  evolution  of  opinion  is  part  of 
the  whole  evolution." 

This  Library  will  deal  mainly  with  Modern  Philosophy, 
partly  because  Ancient  Philosophy  has  already  had  a  fair  share 
of  attention  in  this  country  through  the  labours  of  Grote,  Fer- 
rier,  Benn  and  others,  and  through  translations  from  Zeller ; 
partly  because  the  Library  does  not  profess  to  give  a  complete 
history  of  thought. 

By  the  co-operation  of  different  writers  in  carrying  out  this 
plan,  it  is  hoped  that  a  completeness  and  thoroughness  of  treat 
ment  otherwise  unattainable  will  be  secured.  It  is  believed, 
also,  that  from  writers  mainly  English  and  American  fuller  con 
sideration  of  English  Philosophy  than  it  has  hitherto  received 
from  the  great  German  Histories  of  Philosophy  may  be  looked 
for.  In  the  departments  of  Ethics,  Economics  and  Politics, 
for  instance,  the  contributions  of  English  writers  to  the  common 
stock  of  theoretic  discussion  have  been  especially  valuable,  and 
these  subjects  will  accordingly  have  special  prominence  in  this 
undertaking. 

Another  feature  in  the  plan  of  the  Library  is  its  arrangement 
according  to  subjects  rather  than  authors  and  dates,  enabling  the 
writers  to  follow  out  and  exhibit  in  a  way  hitherto  unattempted 
the  results  of  the  logical  development  of  particular  lines  of 
thought. 

The  historical  portion  of  the  Library  is  divided  into  two 
sections,  of  which  the  first  contains  works  upon  the  develop 
ment  of  particular  schools  of  Philosophy,  while  the  second 
exhibits  the  history  of  theory  in  particular  departments. 

To  these  have  been  added,  by  way  of  Introduction  to  the 
whole  Library,  (i)  an  English  translation  of  Erdmann's  His 
tory  of  Philosophy,  long  since  recognised  in  Germany  as  the  best ; 
(2)  translations  of  standard  foreign  works  upon  Philosophy. 

J.  H.  MUIRHEAD, 

General  Editor. 


ALREADY  PUBLISHED. 
THE  HISTORY  OF  PHILOSOPHY.     By  DR.  JOHANN  EDUARD  ERDMANN. 

English  Translation.  Edited  by  WILLISTON  S.  HOUGH,  M.Ph.,  Pro 
fessor  of  Mental  and  Moral  Philosophy  and  Logic  in  the  University  of  Min 
nesota. 

In  3  vols.,  medium  8vo,  cloth. 

Vol.  I.  Ancient  and  Mediaeval  Philosophy,  155.  ,.  Third  Edition. 
Vol.  II.  Modern  Philosophy,  155.  „  .  .  .  Third  Edition. 
Vol.  III.  Modern  Philosophy  since  Hegel,  125.  .  Third  Edition. 

THE  HISTORY  OF  ESTHETIC.     By  BERNARD  BOSANQUET,  M.A.,  LL.D.,  late  Fellow 
of  University  College,  Oxford.     IDS.  6d.  net. 

THE  DEVELOPMENT  OF  RATIONAL  THEOLOGY  since  Kant.     By  PROFESSOR  OTTO 
PFLEIDERER,  of  Berlin.     IDS.  6d.  net.  Second  Edition. 

PHILOSOPHY  AND  POLITICAL  ECONOMY  IN  SOME  OF  THEIR  HISTORICAL  RELATIONS. 
By  JAMES  BONAR,  M.A.,  LL.D.     105.  6d.  net. 

APPEARANCE  AND  REALITY.     By  F.  H.  BRADLEY,  M.A.,  Fellow  of  Merton  Col 
lege,  Oxford.     125.  net.  Third  Edition. 

NATURAL  RIGHTS.     By  DAVID  G.  RITCHIE,  Professor  of  Logic  and  Metaphysics 
in  the  University  of  St.  Andrews.     IDS.  6d.  net. 

SIGWART'S  LOGIC.     Translated  by  HELEN  DENDY.     2  vols.     215.  net. 

ANALYTIC  PSYCHOLOGY.     By  G.  F.  STOUT,  M.A.,  Fellow  of  St.  John's  College, 
Cambridge,  Wilde  Reader  in  Mental  Philosophy,  Oxford.     2  vols.     215. 
net.  Second  Edition. 

A  HISTORY  OF  ENGLISH  UTILITARIANISM.     By  ERNEST  ALBEE,  Ph.D.,  Instructor 
in  the  Sage  School  of  Philosophy,  Cornell  University.     105.  6d.  net, 

CONTEMPORARY  PSYCHOLOGY.     By  PROFESSOR  GUIDO  VILLA,  Lecturer  on  Philo 
sophy  in  the  University  of  Rome.     Authorized  Translation.     105.  6d.  net. 

THOUGHT  AND  THINGS  :  a  Study  of  the  Development  and  Meaning  of  Thought, 
or  Genetic  Logic.     By  JAMES  MARK  BALDWIN,  Ph.D.,  Hon.  D.Sc.  (Oxon.). 
LL.D.    (Glasgow),    Professor   in    the    Johns    Hopkins    University.     Vol    I, 
Functional  Logic  :    or  Genetic  Theory  of  Knowledge.      los.  6d.  net. 
Vol.  II.,  Experimental  Logic,  and  Vol.  III.,  Real  Logic  (in  preparation). 


SWAN  SONNENSCHEIN  &  CO.,  LIM.,  LONDON. 
THE  MACMILLAN  COMPANY,  NEW  YORK. 


ERDMANN'S  HISTORY  OF  PHILOSOPHY. 


NOTICES  OF  THE  PRESS. 

"  A  SPLENDID  monument  of  patient  labour,  critical  acumen  and  admirable 
methodical  treatment.  ...  It  is  not  too  much  to  predict  that,  for  the  library 
of  the  savant,  for  the  academical  student,  whose  business  it  is  to  be  primed  in 
the  wisdom  of  the  ages,  and  for  the  literary  dilettante,  who  is  nothing  if  not 
well  up  in  '  things  that  everybody  ought  to  know,1  these  volumes  will  at  once 
become  a  necessity  for  purposes,  at  least,  of  reference,  if  not  of  actual  study.  .  .  . 
We  possess  nothing  that  can  bear  any  comparison  with  it  in  point  of  complete 
ness." — Pall  Mall  Gazette. 

"  It  is  not  necessary  to  speak  of  the  great  merits  of  Erdmann's  History  of 
Philosophy.  Its  remarkable  clearness  and  comprehensiveness  are  well  known.  .  .  . 
The  translation  is  a  good,  faithful  rendering,  and  in  some  parts  even  reaches  a 
high  literary  level." — Professor  JOHN  WATSON,  in  The  Week,  of  Canada. 

"  It  is  matter  of  real  congratulation,  in  the  dearth  still  of  original  English  or 
American  work  over  the  whole  field  of  historical  philosophy,  that  by  the  side  of 
the  one  important  German  compend  of  this  generation,  the  other,  so  well  fitted 
to  serve  as  its  complement,  is  now  made  accessible  to  the  English-speaking 
student." — Mind. 

"  It  has  been  long  known,  highly  esteemed,  and  in  its  successive  editions 
has  sought  to  make  itself  more  worthy  of  the  success  it  has  justly  achieved. 
Erdmann's  work  is  excellent.  His  history  of  mediaeval  philosophy  especially 
deserves^  attention  and  praise  for  its  comparative  fulness  and  its  admirable 
scholarship.  ...  It  must  prove  a  valuable  and  much  needed  addition  to  our 
philosophical  works." — Scotsman. 

"  The  combination  of  qualities  necessary  to  produce  a  work  of  the  scope 
and  grade  of  Erdmann's  is  rare.  Industry,  accuracy,  and  a  fair  degree  of  philo 
sophic  understanding  may  give  us  a  work  like  Ueberweg's  ;  but  Erdmann's 
history,  while  in  no  way  superseding  Ueberweg's  as  a  handbook  for  general 
use,  yet  occupies  a  different  position.  Erdmann  wrote  his  book,  not  as  a  refer 
ence  book,  to  give  in  brief  compass  a  digest  of  the  writings  of  various  authors,  but 
as  a  genuine  history  of  philosophy,  tracing  in  a  genetic  way  the  development 
of  thought  in  its  treatment  of  philosophic  problems.  Its  purpose  is  to  develop 
philosophic  intelligence  rather  than  to  furnish  information.  When  we  add  that, 
to  the  successful  execution  of  this  intention,  Erdmann  unites  a  minute  and 
exhaustive  knowledge  of  philosophic  sources  at  first  hand,  equalled  over  the 
entire  field  of  philosophy  probably  by  no  other  one  man,  we  are  in  a  condition 
to  form  some  idea  of  the  value  of  the  book.  To  the  student  who  wishes,  not 
simply  a  general  idea  of  the  course  of  philosophy,  nor  a  summary  of  what  this 
and  that  man  has  said,  but  a  somewhat  detailed  knowledge  of  the  evolution 
of  thought,  and  of  what  this  and  the  other  writer  have  contributed  to  it,  Erd 
mann  is  indispensable  ;  there  is  no  substitute." — Professor  JOHN  DEWEY,  in 
The  Andover  Review. 

"  It  is  a  work  that  is  at  once  compact  enough  for  the  ordinary  student,  and 
full  enough  for  the  reader  of  literature.  ...  At  once  systematic  and  interest 
ing." — Journal  of  Education. 

"  The  translation  into  English  of  Erdmann's  History  of  Philosophy  is  an 
important  event  in  itself,  and  in  the  fact  that  it  is  the  first  instalment  of  an  under 
taking  of  great  significance  for  the  study  of  philosophy  in  this  country.  Apart, 
however,  from  its  relation  to  the  Library  to  which  it  is  to  serve  as  an  introduc 
tion,  the  translation  of  Erdmann's  History  of  Philosophy  is  something  for  which 
the  English  student  ought  to  be  thankful.  ...  A  History  of  past  endeavours, 
achievements  and  failures  cannot  but  be  of  great  use  to  the  student.  Such  a 
History,  able,  competent,  trustworthy,  we  have  now  in  our  hands,  adequately 
and  worthily  rendered  into  our  mother-tongue." — Spectator. 


Xtbrar$  of  pbilosopbp. 

EDITED  BY  J.    H.   MUIRHEAD,   LL.D.    - 


VALUATION:  ITS  NATURE  AND  LAWS 


VALUATION 

NATURE    AND    LAWS 


AN  INTRODUCTION  TO 
THE  GENERAL  THEORY  OF  VALUE 


BY 

WILBUR  MARSHALL  URBAN,  PH.  D. 

FORMERLY   CHANCELLOR   GREEN    FELLOW    IN    MENTAL   SCIENCE,    AND   HEADER    IN 

PHILOSOPHY,    PRINCETON    UNIVERSITY;    PROFKSSOR   OF    PHILOSOPHY, 

TRINITY   COLLF.UE,    HARTKORO,    CONN. 


LONDON 

SWAN    SONNENSCHEIN    &    CO.,    LIM. 

NEW   YORK:    THE    MACMILLAN    CO. 

1909 


ovvc  Vt*-'':rr* 

,.    .   RlP 

.VAk-'^*-"  * 


TO 

JAMES    MARK    BALDWIN 

AS   A    MARK    OF 
APPRECIATION    AND    RESPECT 


' 


PREFACE 

THE  "  point  of  view,"  the  unexpressed  assumptions  in  the  half- 
light  of  which  any  book,  even  the  most  "  scientific,"  is  written, 
and  which  sometimes  find  their  way  into  the  preface,  are  often 
more  informing  and  more  interesting  than  the  book  itself.  This 
should  be  eminently  true  in  the  case  of  a  book  on  Values,  for 
here,  if  anywhere,  the  writer  might,  perhaps,  be  permitted  to 
wear  his  heart  on  his  sleeve.  It  is  not  my  purpose,  however, 
to  add  this  glimmer  of  possible  human  interest  to  an  otherwise 
dull  book.  In  so  far  as  the  statement  of  such  presuppositions 
may  have  any  significance  for  the  main  developments  of  the 
book,  they  are  presented  with  due  objectivity  and  detachment 
in  the  Introduction,  and  carried — with  equal  care,  it  is  to  be 
hoped — to  some  of  their  more  obvious  conclusions  in  the  final 
chapter.  It  is  unnecessary  to  anticipate  them  here.  But  with 
regard  to  the  more  general  background — the  relation  of  such 
a  book  to  the  characteristic  assumptions  of  the  time — the  situa 
tion  is  different.  Here  a  prefatory  word  may  be  a  word  in 
season  ;  it  may  not  only  illuminate  some  of  the  dark  places,  but 
may  even  give  the  key  to  an  appreciation  if  not  to  a  complete 
understanding  of  the  entire  discussion. 

Until  recently  there  was  little  question  as  to  what  should 
be  the  suppressed  major  premise  of  any  serious  inquiry,  what 
ever  its  subject-matter  might  be.  To  science  and  the  scientific 
method  belonged  the  whole  "  choir  of  heaven  "  as  well  as  the 
"  furniture  of  earth."  To  leave  this  assumption  unquestioned 
was  felt  to  be  the  only  correct  attitude,  and  where  its  acceptance 
was  not  whole-souled,  half-hearted  imitation  took  its  place. 


viii  Preface 

For  some  time,  it  is  true,  this  assumption  has  not  been  without 
its  disquieting  effects  upon  life  and  art,  upon  morals  and  religion. 
The  sophistication  and  even  pruriency  of  thought  and  feeling 
to  which  some  of  the  expressions  of  the  scientific  spirit  have 
given  rise,  have  led  to  a  reaction  against  intellectualism  which 
if  not  widespread  is  at  least  profound.  But  it  was  not  until 
the  triumphant  march  of  science  led  to  the  soul  and  its  inmost 
values  that  this  reaction  took  definite  form.  The  realisation 
of  the  capacity  for  large  ineptitudes  no  less  than  for  small  futili 
ties,  which  an  uncriticised  application  of  scientific  method  may 
display,  has  led  to  a  questioning  of  its  most  fundamental  assump 
tions.  Vague  distrust  has  developed  into  outspoken  alogistic 
and  even  misologistic  tendencies,  until  the  counter-assumption, 
that  values  lie  beyond  the  ken  of  knowledge  and  science,  bids 
fair  to  rival  its  opponent  in  honour.  In  our  theory  as  well  as  in 
our  practice  we  have  reached  a  point  of  equivocation,  not  to  say 
contradiction,  at  which  we  must  either  take  refuge  in  a  new 
doctrine  of  "  two-fold  truth  "  or  else,  if  there  is  any  practical 
meaning  in  the  principle  of  the  dialectic,  await  with  patience  a 
middle  ground  of  unification. 

In  the  meantime,  this  dilemma  of  the  Time-Spirit  demands  a 
new  and  rigid  alignment  of  principles,  and  any  book  which  enters 
the  "  fighting-zone  "  owes  it  to  itself  to  be  clear  on  this  point. 
The  present  work  places  itself  frankly  on  the  side  of  knowledge 
rather  than  of  edification,  in  the  full  belief  that  the  latter  pre 
supposes  the  former.  For  better  or  for  worse,  we  are  caught  in 
the  grip  of  an  immitigable  will  to  know,  a  will  to  know  which 
claims  for  its  province  "  the  human  soul  and  its  limits,  the 
entire  range  of  its  hitherto  acquired  experiences,  the  entire  history 
of  the  soul  and  its  still  unexhausted  possibilities."  Short  of 
this  we  cannot  stop  except  by  the  loss  of  the  energy  of  this  will 
which  means  the  decadence  of  values  as  well.  Nor  does  it  shun 
the  bringing  together  of  the  concepts  of  value  and  science.  If 
science  when  it  has  followed  this  track  has  sometimes  shown 


Preface  ix 

qualities  meriting  the  term  "  dismal,"  it  does  not  follow  that 
when  it  has  penetrated  more  deeply  it  may  not  be  "  gay."  We 
may  well  believe  that  the  period  of  a  crude  and  external  domi 
nance  of  the  human  spirit  by  an  inhuman  conception  of  science 
is  approaching  its  zenith,  if  indeed  it  has  not  already  begun  to 
decline  ;  but  it  would  be  a  mistake  to  infer  that  with  such  decline 
will  go  that  deeper  and  more  intimate  control  of  the  individual 
and  social  will  alike  which  is  made  possible  through  the  inter 
pretation  of  their  meanings  in  terms  of  reflective  thought.  Be 
cause  certain  limited  conceptions,  as  well  as  unlimited  claims, 
of  science  are  being  modified,  it  does  not  follow  that  the  power 
of  science,  in  the  older  and  better  use  of  that  term,  will  be  lessened. 
Rather  may  we  confidently  look  for  its  increase.  For  faith  and 
feeling  also  make  unlimited  claims  which  only  the  discipline  of 
the  scientific  spirit  enables  us  properly  to  appraise.  There  will 
always  be  new  ventures  in  faith  and  science  alike,  and  new 
ventures  must  always  be  followed  by  new  evaluations.  But 
such  evaluations  are  not  to  be  secured  by  reference  to  a  closed 
system,  either  of  truths  or  of  values,  but  only  by  that  orderly 
progression  from  actualities  to  possibilities  and  certainties 
which  is  the  method  of  science.  In  the  interest  of  true  evalua 
tions,  the  present  time  is  committed  to  the  full  development  of 
all  that  is  implied  in  the  concept  of  science  rightly  understood. 

In  comparison  with  this  task — in  its  larger  aspects  by  no 
means  that  of  the  merely  technical  philosopher,  the  aim  of  the 
following  pages  is  much  more  modest.  Limited  as  they  are  to 
certain  "first  works"  which  must  in  the  nature  of  the  case  be 
largely  technical,  the  writer  must  be  content  merely  to  hope  that 
they  may  affect,  at  least  indirectly,  the  larger  issues. 

The  first  six  chapters,  beginning  with  "  definition  and  analy 
sis,"  and  culminating  in  the  chapter  on  the  general  laws  of 
valuation,  seek  to  lay  the  foundations  for  an  understanding 
of  the  different  types  of  value  judgments,  their  implications 
and  their  limits.  In  the  development  of  this  portion,  chapters 


x  Preface 

IV.  and  v.  are  in  a  sense  an  interruption,  since  they  are  wholly 
psychological ;  and,  while  they  treat  of  matters  necessary  to  the 
complete  understanding  of  the  more  general  topics  that  follow, 
might  have  been  handled  in  separate  sections  as  required.  On 
the  whole,  it  seemed  better  to  bring  them  together  in  one 
systematic  treatment.  Chapters  vn.  to  xm.  inclusive  apply  these 
results  to  a  genetic  and  synthetic  investigation  of  the  conscious 
ness  of  value  from  its  lowest  to  its  highest  levels.  Without 
laying  claim  to  comprehensiveness  or  completeness,  it  may  be 
said,  I  think,  that  there  is  no  significant  form  of  worth  experience 
which  is  not  adequately  enough  treated  to  show  its  relation  to 
the  general  system  of  values.  Finally,  in  chapter  xiv.  an 
attempt  is  made  to  show  the  bearing  of  these  results  on  what  I 
have  ventured  to  call  the  axiological  problem  of  evaluation. 
The  views  which  I  have  felt  justified  in  presenting  in  the 
limited  space  of  a  single  chapter  represent  but  in  outline 
certain  philosophical  conclusions  to  which  the  study  of  the 
phenomenology  of  valuation  has  led.  The  work  was  originally 
planned  to  be  merely  such  a  phenomenological  study,  and  as 
such  it  must  be  judged,  but  the  unity  which  the  entire  work 
gained  by  the  addition  of  this  chapter  seemed  to  compensate 
for  any  inadequacies  which  the  chapter  might  appear  to  have 
when  viewed  by  itself.  In  this  connection  I  wish  to  express 
my  regret  that  Miinsterberg's  Philosophic  der  Werthe  appeared 
too  late  for  the  utilisation,  except  in  the  last  chapter,  of  any 
of  the  valuable  suggestions  which  I  have  got  from  its  study. 
While  my  general  position  has  remained  unmodified,  I  could  not 
have  remained  uninfluenced,  in  many  details  at  least,  by  his 
brilliant  and  persuasive  presentation  of  a  view  which  is  essentially 
opposed  to  my  own. 

In  acknowledging  my  indebtedness  to  recent  writers  for 
many  of  the  ideas  contained  in  these  pages,  I  have  first  of  all 
to  express  my  specific  obligations  to  the  oral  as  well  as  written 
teachings  of  Professor  Baldwin.  My  deeper  sense  of  obligation 


Preface  xi 

he  has  most  kindly  allowed  me  to  express  in  the  form  of  dedica 
tion.  Of  my  indebtedness  to  Meinong  and  Ehrenfels,  as  well  as 
to  others  of  that  school,  my  references  to  their  works  give  visible 
proof.  These  do  not,  however,  adequately  suggest  the  valuable 
help  that  I  have  received,  not  only  from  their  researches  in  this 
special  field,  but  from  all  their  writings.  To  the  French  psy 
chologists,  Ribot  and  Paulhan,  whose  studies  in  "  feeling " 
have  yet  to  be  properly  valued,  I  am  also  greatly  indebted. 
In  conclusion,  I  have  to  express  my  gratitude  to  Professor 
J.  H.  Muirhead,  the  editor  of  the  Library  of  Philosophy,  for  his 
very  kind  interest  in  the  work,  as  well  as  for  many  valuable 
suggestions  and  criticisms.  My  colleagues  Professor  G.  A. 
Kleene  and  Professor  Arthur  Adams  have  also  rendered  me 
great  assistance,  the  former  with  expert  suggestions  on  points 
in  economics,  the  latter  in  connection  with  the  reading  of  the 
proof. 

Chapters  u.,  m.,  and  part  of  iv.  have  already  appeared  as 
articles  in  the  Psychological  Review.  They  have  in  each  case 
been  considerably  modified  to  suit  their  present  purpose.  The 
general  Introduction  is  an  expansion  and  modification  of  a 
paper  printed  in  the  Philosophical  Review  under  the  same  title. 


TRINITY  COLLEGE,  HARTFORD,  CONN. 
December,  1908. 


CONTENTS 

CHAPTER   I 

INTRODUCTION 

PAGE 

The  Function  of  a  General  Theory  of  Value  :  its  Nature  and  Sources  .  I 

The  Psychological  Problem  and  Method — The  Presuppositional  Method  of 

Psychological  Analysis  :  a  form  of  the  Genetic  Method  .  .  -9 

The  Axiological  Problem  and  Method — Facts  and  Norms — Genesis  and 

Validity       .  .  .  .  .  .  .  .  .16 

The  Relation  of  the  Psychological  and  Axiological  Points  of  View  .  .18 


CHAPTER   II 

DEFINITION   AND   ANALYSIS   OF  THE   CONSCIOUSNESS   OF   VALUE 

The  Judgment  of  Value — Worth  Predicates  as  Affective- Volitional  Meanings 
— Equivocations  in  Judgments  of  Value  leading  to  Analysis  of  Pre 
suppositions  .  .  .  .  .  .  .  .21 

Values  as  Meanings  of  an  Object  for  a  Subject — The  Subject  of  the  Value 
Judgment — The  Subject  in  Different  Attitudes — Classification  of  Atti 
tudes  :  Simple  Appreciation,  Personal  and  Over-Personal  or  Impersonal 
— The  Object  of  the  Value  Judgment — Objects  of  Condition,  Personal 
and  Over- Individual  Value — The  Relation  of  Subject  and  Object  .  26 

Psychological  Analysis  of  Worth  Experience — Worth  as  Feeling  with  certain 
Cognitive  Presuppositions — The  Worth-Fundamental  is  Feeling,  not 
Desire — The  Presupposition  of  Reality  not  exclusively  Existential  Judg 
ment  :  Criticism  of  Theory  of  Existential  Judgment  underlying  this  View 
— The  Presupposition  of  Reality:  Presumption,  Judgment  and  Assump 
tion  of  Existence  of  Objects — Analysis  of  these  Cognitive  Attitudes  .  35 

The  Genesis  and  Relations  of  these  Presuppositions — Genetic  Levels  of 
Valuation  .  .  .  .  .  .  .  .  -49 

Resume  of  Definition  and  Analysis     .  .  .  .  .  -S3 


CHAPTER   III 

MODES    OF    THE   CONSCIOUSNESS   OF   VALUE — PRIMARY   AND    ACQUIRED 

Appreciative  Description  of  Feelings  of  Value  :  its  Nature  and  Method        .     5  5 
The   Fundamental  Appreciative  Distinctions  in   Feeling — Descriptions  of 
Reality-Meanings     of     Feelings     of     Value,     not     of     Simple     Feeling 


Contents 

abstracted    from    Cognitive    Presuppositions— The    Three-Dimensional " 
Theory  of  Feeling :   Interpretation— Correlation  of  Appreciative  Mean 
ings  with  Cognitive  Presuppositions  ...  50 

Meanings  Acquired  by  Development  from  the  Fundamental  Modes— Value- 
Movement— Acquired  Meanings  of  Simple  Appreciation— The  Impellent 
Mode :  Feelings  of  Obligation— The  Semblant  Mode :  ^Esthetic  Feeling  67 

Acquired  Meanings  of  Characterisation  and  Participation:  Personal  and 
Over-Individual  Values  ....  7I 

The  Quantitative  Meanings  of  Worth  Feeling— Analysis  of  the  Concept  of 
Degree— Degree  of  Worth  (Depth  and  Breadth)  and  Degree  of  Intensity  : 
Independently  Variable— Intensity-less  Appreciation  :  Illustrations- 
Theories  of  this  Relation — Suggestion  of  a  Genetic  Theory  to  be 
developed  later  .....  72 

The  Bearing  of  this  Analysis  on  Further  Problems     .  .  78 


CHAPTER    IV 

THE   PSYCHOLOGICAL  BASIS   OF   A  THEORY  OF  VALUATION 

The  Nature  of  Feeling  and  Will  and  their  Relations— Dualistic  Theories  of 
Feeling  and  Will :  Criticism— Monistic  and  Genetic  Theory— Interpreta 
tion  of  the  Monistic  Theory :  its  Relation  to  the  Definition  and  Analysis 
of  the  Consciousness  of  Value  .  .  .  .  .  g  r 

Further  Analysis  of  Feeling:  Structural  and  Functional— Feeling  as  a 
Kind  of  Sensitivity — The  Appreciative  Distinctions  in  Feeling  as 
"  Forms  of  Combination "  of  the  Elements  .  .  .  .96 

Correlation  of  Structural  and  Functional  Analysis — Changes  in  Functional 
Meaning  and  Changes  in  Sensitivity:  Passion  and  Emotion.  Sentiment 
and  Mood,  and  Affective  Sign— Theory  of  their  Genetic  Relations  .  103 

Corollaries  from  the  Preceding  Theory — Their  Significance  for  the  Theory 
of  Valuation  ....  .108 


CHAPTER   V 

THE   CONTINUITY   OF   AFFECTIVE-VOLITIONAL   MEANING 

The  Acquirement  of   Recognitive  and  Generic   Meanings  on   the  Part  of 
Feelings :  The  Problem      .  .  .  .  .  .  .   1 1  r 

Affective  Memory — Types  of  Affective  Memory — The  Criterion  of  Recogni 
tive  Meaning  as  applied  to  Feelings — Theory  of  Affective  Memory         .   113 

The   Generic   Meanings  of    Feeling:    Affective  Generalisation — The   Phe 
nomena  of  Affective  Continuity:  Substitution,  Subsumption,  Transition   120 

Psychological  Theory  of  Generic  Meanings   of  Feeling — Imageless   Appre 
hension    and    Intensity-less   Appreciation — The    Process   of    Affective 
Generalisation        .  .  .  .  .  .  .  •   121 

The  Role  of  Affective  Generals  in  Processes  of  Valuation — Worth  Continui 
ties:   Illustrations      .  .  .  .  .  .  -133 


Contents  xv 

CHAPTER  VI 

THE   LAWS   OF  VALUATION 

PAGE 

The  Laws  of  Valuation :  Their  Nature  and  Range  of  Application—4 

Economic  and  Extra-Economic  Values — Classification  and  Interpretation  142 

The  Law  of  the  Threshold :  General  Meaning — Economic  Thresholds, 
Upper  and  Lower  :  The  Existence-Minimum  and  Final  Utility — Modific 
ation  of  the  Threshold  through  Acquired  Meanings :  Complementary 
Values — General  Function  of  Threshold  in  Processes  of  Valuation — The 
Independent  Variability  of  Hedonic  and  Worth  Thresholds  .  .146 

The  Law  of  Diminishing  Utility :  Its  Historical  Significance  and  Relation 

to  a  General  Theory  of  Value  ,  .  .  .  .  .156 

The  Psychological  Basis  of  the  Law :  Dulling  of  Sensitivity  with  Repetition 
and  Satiety — Critical  Examination  of  these  Laws — Their  Application 
limited  to  Sense-Feelings  and  to  the  Hedonic  Redundancies  of  Feelings 
of  Value — Restatement  of  the  General  Law  of  Limiting  or  Diminishing 
Value  .  .  .  .  .  .  .  .  .158 

Extent  of  the  Application  of  this  Law :  Its  Application  to  all  Instrumental 
Utility  Values,  but  not  necessarily  to  Intrinsic  Ideal  Values — The  Law 
of  Marginal  Utility  and  its  Explanation — Certain  Limitations  of  this  Law  167 

The  Law  of  Complementary  Values — As  Modifying  the  Law  of  Limiting 
Value :  In  Economic  Valuation ;  In  Extra-Economic  Valuation, 
Ethical  and  ..Esthetic — General  Characterisation  of  the  Law:  Its 
Psychological  Basis — Interpretation  .  .  .  .  .  173 

The  Application  of  these  Laws  to  Ideal  Objects  of  Intrinsic  Value — The 
Limits  of  this  Application  .  .  .  .  .181 

General  Conclusions :  The  Problem  of  the  Limits  of  Acquirement  of  Value — 
Inferences  from  this  Study  of  the  Laws  of  Valuation — Their  Bearing 
upon  Further  Studies  .  .  .  .  .  .  .185 


CHAPTER   VII 

VALUES   OF   SIMPLE   APPRECIATION 

Values  of  Simple  Appreciation  :  their  Origin  and  Nature — Objects  of  "  Con 
dition"  Worth,  Primary  and  Derived  .....  190 

Value  Movements  in  General :  Definition  and  Classification — Their  Relation 
to  the  Laws  of  Valuation  .  .  .  .  .  .  .  194 

Value  Movements  of  Simple  Appreciation :  Ethical  and  ^Esthetic  Values 
thus  Acquired — Modification  of  Economic  Values  of  Acquisition  and 
Consumption  ........  205 

The  Ethical  or  Impellent  Mode  of  Simple  Appreciation  :  As  the  Result  of 
Inward  Value  Movement — Pre-Ethical  and  Quasi-Ethical  Impulsions 
and  Obligations :  their  Sub-Personal  and  Sub-Social  Character — Illus 
trations  of  Instinctive,  Quasi-Ethical  Obligations — Modification  of  Econ 
omic  Valuation  by  these  Acquired  Obligations  .  .  .  .  207 

^Esthetic  or  Semblant  Mode  of  Simple  Appreciation :  as  Special  Form  of 
the  Value  Movement  toward  Activity — Pre-^Esthetic  and  Quasi- 
jEsthetic  Forms  of  Activity  :  Their  Individual  and  Sub-Social  Character 
Conditions  of  Movement  to  Activity  in  its  Pre-^Esthetic  Form — The 
Special  Differentia  of  .the  ^Esthetic  .  .  .  .  .216 

Modification  of  Economic  Valuation  by  Acquired  ^Esthetic  Values  .  .  229 


xvi  Contents 

CHAPTER  VIII 

PERSONAL    AND     OVER-INDIVIDUAL    VALUES — THEIR    ORIGIN 
AND    NATURE 

PACE 

Personal  and  Over-Individual  Values :  their  Origin  and  Nature — Their 
"  Common  Meaning  "  Presupposes  Sympathetic  Participation  .  .232 

Sympathetic  Participation  (Einfuhlung):  its  Relation  to  Simple  Apprecia 
tion  and  Feeling  .  .  .  .  .  .  .  -234 

The  Psychology  of  Sympathetic  Projection:  Its  Nature  and  Conditions — 
The  Inducing  Conditions  of  Affective  Projection  .  .  .  .236 

Levels  of  Sympathetic  Participation — The  Presuppositions  of  Sympathetic 
Feelings  :  (a)  Organic  Sympathy,  with  Vague  Presumption  of  Existence  ; 
(b)  Rise  of  Assumption-Feelings  and  Emergence  of  Distinctions  in  Pre 
suppositions — Sembling  ;  (c)  Judgment-Feelings  :  Conceptual  Reconstruc 
tion  of  the  Inner  Life  in  Terms  of  Dispositions  ....  244 

Sympathetic  Participation  (Einfuhlung)  as  a  Process  of  Valuation — The 
Nature  of  the  Feelings  of  Value  Involved :  the  Projected  Feelings 
"Real"  Feelings — Value  Movements  in  Participation  and  Characterisa 
tion  .........  249 

The  Distinction  between  Personal  and  Impersonal  Participation — Intensive 
and  Extensive  Projection :  Resulting  in  Differences  of  Presuppositions 
of  Feelings  ......  *  253 

CHAPTER   IX 

PERSONAL   WORTHS  :    VALUES   OF  CHARACTERISATION 

Definition  :  The  Personal  Attitude  in  Valuation — The  Presuppositions  of 
this  Attitude — The  Ideal  of  the  Personality — How  assumed  in  all  Judg 
ments  of  Personal  Worth  .......  260 

The  Character  of  the  Ideal  Person  as  determined  by  the  Processes  in  which 
it  is  Constructed — Idealisation  involved  in  Sympathetic  Participation — 
Division  of  the  Personality — Extrusion  of  the  Negative  Elements — 
Further  Stages  in  Idealisation  :  ^Esthetic  Individuation  of  the  Person  ; 
Acquirement  of  Complementary  Value  .....  263 

The  Laws  of  Valuation  as  applied  to  Personal  Worth — The  Problem  in  the 
Light  of  our  General  Study — Special  Examination  of  Feelings  of  Personal 
Worth  .........  270 

The  Activities  of  Idealisation  and  Characterisation — Not  Subject  to  the  Law 
of  Diminishing  Value  :  Feelings  of  Personal  Worth  not  Subject  to  Dulling 
of  Sensitivity  and  Satiety  .......  272 

The  Effect  of  Idealisation  on  our  Actual  Feelings  and  Judgments — The 
Semblant  Mode  in  Characterisation — Dispositions  Created  which  affect 
Actual  Judgment  .  .  .  .  .  .  .  275 

Absolute  Personal  Values  :   They  exist  as  Practical  Absolutes  .  .  277 

CHAPTER   X 

PERSONAL  WORTHS — Continued 

Analysis  and  Interpretation  of  the  Concrete  Judgments  of  Personal  Worth  : 
Imputation  and  Obligation — Definition  of  the  Subject,  Object,  and  Terms 
of  Estimation  Presupposed  in  such  Judgments  «  .  '  -»  .  282 


Contents  xvii 


PAGE 


Difference  between  the  Personal  or  Ethical  and  the  Impersonal  or  Moral 
Standpoints  .........  288 

The  Relative  Estimation  of  Personal  Worth — The  Thresholds  and  Norms 
of  Personal  Obligation  and  Imputation  :  Their  Origin  in  the  Processes  of 
Sympathetic  Participation  .  .  .  .  .  .  .291 

The  Normal  Threshold  :  The  Norm  of  Characterisation — The  Upper  and 
Lower  Limits  of  Personal  Worth  :  The  Characterisation-Minimum — How 
Presupposed  in  all  Judgments  of  Personal  Worth  ....  293 

Laws  Governing  Feelings  of  Personal  Worth :  as  Illustrated  in  Imputation 
of  Merit  and  Demerit  and  in  Personal  Obligation  ....  298 

The  Axiological  Question  of  the  Validity  of  the  Implicit  Assumptions  or  Pre 
suppositions  of  these  Judgments  ......  309 


CHAPTER   XI 

IMPERSONAL    OVER-INDIVIDUAL   VALUES 

Definition  :  Over-Individual  Values  of  Participation  and  Utilisation — The 
Impersonal  Attitude  in  Valuation — The  Morally  Qualified  Act  and  the 
Morally  Qualified  Judgment — Relativity  of  the  Distinction  between  the 
Personal  and  Impersonal  Attitude  .  .  .  .  .  311 

The  Presuppositions  of  the  Impersonal  Attitude  :  Over-Individual  Demands 
— Over-Individual  Demands  as  Collective  Desire  and  Feeling — Their 
Origin  and  Nature — Social  Synergies  :  Demand  and  Supply — The  Re 
lation  of  Subjective  to  Objective  Participation  Value  .  .  .  317 

Subjective  Participation  Value — The  Individual's  Feelings  of  Participation 
Value  as  Determined  by  Social  Sympathy — Extensive  Sympathetic  Par 
ticipation — The  Laws  of  Social  Sympathy  .  .  .  .  320 

The  Objective  Participation  Value  of  Dispositions  as  Deduced  from  the 
Laws  of  Social  Sympathy — The  Law  of  Marginal  Participation  Value  .  328 

Corollaries  from  the  Law  of  Marginal  Participation  Value — The  Laws  of 
Social  Synergy — Social  Value  Movements — Social  Differentiation  and 
Segregation — The  Norms  and  Limits  of  Participation  Value  .  .  331 

The  Limits  of  Acquirement  of  Social  Over-Individual  Value — The  Question 
of  Absolute  Social  Values  .......  342 


CHAPTER  XII 

IMPERSONAL    OVER-INDIVIDUAL   VALUES — Continued 

Analysis  and  Interpretation  of  Concrete  Moral  Judgments — The  Moral 
Value  of  an  Act  as  determined  by  its  Participation  Value — Estimation 
of  Moral  Value  in  Terms  of  Egoism  and  Altruism — The  Significance  of 
these  Terms  .  .  .  .  .  .  .  .  •  35° 

The  Thresholds  and  Norms  of  Moral  Judgment — The  Norm  of  Participation  : 
The  "  Correct  "  :  The  Normal  Expectation  of  Social  Participation — The 
Lower  Threshold  :  The  Participation-Minimum — How  Developed  in 
Social  Participation  and  Presupposed  in  all  Judgments  of  Moral  Value  .  356 

Laws  Governing  Objective  Participation  Value  as  reflected  in  Imputation 
of  Praise  and  Blame  and  in  Moral  Obligation — Moral  Obligation  Relative, 
not  Absolute  ........  358 


xviii  Contents 

CHAPTER  XIII 

SYNTHETIC  PREFERENCE 

PAGE 

The  Relative  Value  of  Different  Classes  of  Worth  Objects— The  Levels  of 
Valuation  .........  366 

Rationalistic  and  Monistic  Theories  of  Valuation — Criticism — Intuitionism 
and  Scepticism  ........  368 

Analysis  of  the  Facts  of  Synthetic  Preference  as  exhibited  in  Judgments  of 
Obligation  and  Imputation  of  Value — Conflicts  between  the  Ideals  and 
Obligations  of  the  Different  Levels  of  Condition,  Personal  and  Impersonal 
Over-Individual  Values — The  Resolutions  of  these  Conflicts  in  Actual 
Experience :  In  so  far  as  Uniform  and  Unequivocal,  Reflections  of  the 
General  Laws  of  Valuation — Break-down  of  these  Laws  at  the  Limits  .  371 

Conclusions  from  the  Analysis  of  Synthetic  Preference — Restatement  of 
the  Criticism  of  Monistic  Theories — The  Doctrine  of  Supreme  Moments 
or  Practical  Absolutes — The  Supreme  Moments  of  Intrinsic  Appreciation 
Transcend  all  Distinctions  of  Appreciative  Description — The  Bearing  of 
these  Conclusions  on  Larger  Philosophical  Problems  .  .  .  378 


CHAPTER  XIV— Conclusion 

VALUATION   AND   EVALUATION 

Restatement  of  the  Axiological  Point  of  View — Reflective  Evaluation — 
Normative  and  Factual  Objectivity  .  .  .  .  -384 

Analysis  of  Axiological  Distinctions  of  Reality  and  Truth  employed  in  Re 
flective  Evaluation — The  Meanings  of  Existence  :  Outer  and  Inner — 
The  Meanings  of  Truth  :  Outer  and  Inner  .....  386 

The  Relation  of  Normative  to  Factual  and  Truth  Objectivity — Their 
Meanings  only  Partially  Identical :  The  Objectivity  of  Values  not  wholly 
Reducible  to  Existence  and  Truth — Proof  of  this  Conclusion  in  the  Value 
Judgments  of  Religion  .......  390 

The  Sufficient  Reason  or  Sanction  of  Valuation — The  Ground  of  Value — 
The  Rationalistic  Criterion  ;  The  Pragmatic  Criterion  ;  Criticism — 
Formulation  of  the  Sufficient  Sanction  of  Valuation  in  the  Light  of  the 
Ultimate  Meaning  of  the  Presupposition  of  Reality  and  of  its  Fulfilment  395 

Axiological  Necessity  and  Sufficiency — The  Well-Founded  Value — Axio 
logical  Possibility  and  Compossibility — Formulation  of  a  Criterion  of  the 
Well-Founded  Value  .  .  .  .  .  .405 

Application  of  this  Criterion  to  Specific  Practical  Problems — Again  the 
Monistic  Ideals  .  .  .  .  .  .  •  4J4 

Philosophical  Conclusions  — Truth,  Value,  and  Reality  :  Their  Ultimate 
Relations  .......  •  422 

INDEX  .  .  .  429 


VALUATION 

ITS   NATURE   AND   LAWS 

CHAPTER    I 

INTRODUCTION 

I.  THE  FUNCTION  OF  A  GENERAL  THEORY  OF  VALUE 

THERE  has  seldom  been  a  time  in  the  history  of  thought  when 
the  problem  of  "  value  "  has  so  occupied  the  centre  of  attention 
as  at  present.  Fundamental  changes  in  the  actual  values  of 
mankind,  giving  rise  to  what  has  been  well  called  "  our  anxious 
morality,"  with  its  characteristic  talk  of  creating  and  conserving 
values,  have  brought  with  them  what  may,  without  exaggeration, 
be  described  as  a  gradual  shifting  of  the  philosophical  centre  of 
gravity  from  the  problem  of  knowledge  to  the  problem  of  values. 
The  problem  of  knowledge  has  itself  become,  in  some  quarters 
wholly,  in  others  partially,  a  problem  of  value. 

The  historical  causes  of  this,  until  recently  silent,  change  of 
attitude  are,  in  a  general  way,  clear  enough.  The  change  from 
intellectualism  to  voluntarism,  the  rigorous  discipline  of  the 
human  soul  through  the  almost  universal  application  of  the 
concepts  of  evolution  and  the  struggle  for  existence,  with  their 
ideas  of  selective  and  survival  values — these  are  explanations 
which  immediately  suggest  themselves  ;  and  yet  they  are  but 
general  and  superficial  characterisations  of  a  still  more  funda 
mental  crisis  of  the  social  will,  a  crisis  which  has  its  roots  deep  in 
the  necessities  of  things,  and  which  we  are  as  yet  scarcely  able 
to  understand. 

Whatever  the  causes,  the  effects  are  everywhere  in  evidence. 
This  gradual  change  in  actual  values  has  found  a  mouth 
piece,  if  somewhat  rhetorical  and  rhapsodical,  in  Nietzsche's 
cry  of  "  transvaluation  of  all  values."  But  this  cry  has  been 


2  Valuation  :   its  Natiire  and  Laws 

echoed  by  other  hearts  and  minds,  and  that  which  began  as  a 
species  of  poetry  has  passed  into  sober  prose.  Of  chief  import 
ance  is  the  transition  from  the  accumulation  of  knowledge  to  its 
evaluation.  To  say  nothing  of  the  growing  attempt  to  evaluate 
the  results  of  physical  science  in  the  interests  of  a  more  compre 
hensive  natural  philosophy — a  movement  which  may  or  may  not 
have  some  connection  with  Nietzsche's  arraignment  of  science  in 
its  present  form,  we  may  find  sufficient  evidences  of  this  change 
of  heart  in  the  social  and  moral  sciences,  where  the  problem  of 
value  lies  closer  to  the  surface.  "  While  formerly,"  we  are  told, 
"  it  was  almost  wholly  the  external  structure  of  the  social  life, 
and  the  economic  values  which  it  produces,  that  received  atten 
tion,  now  it  is  the  meaning  of  this  life  for  the  human  soul,  its 
spiritual  origin  and  spiritual  effects,  which  finds  expression."1 
In  short  it  is  the  problem  of  evaluation. 

Corresponding  to  this  change  in  practical  attitude,  has  ap 
peared  the  more  theoretical  consciousness  of,  as  it  were,  a  new 
side  of  reality.  We  have  been  scarcely  aware,  so  we  are  told, 
that  our  entire  life,  on  its  conscious  side,  is  one  continuous  series 
of  feelings  of  value  and  evaluations,  of  explicit  judgments  and 
implicit  assumptions  of  value  ;  and  that  it  is  only  by  reason  of 
the  very  fact,  that  they  are  valued,  that  the  mechanically  deter 
mined  elements  of  reality  in  any  sense  have  meaning  for  us. 
Far  from  being  a  mere  fact  among  other  -facts,  that  which  we 
mean  by  our  evaluation  of  objects  is  something  independent  of 
this  world,  and  so  little  merely  a  part  of  it  that  it  is  rather  the 
whole  world  seen  from  a  special  point  of  view.  Over  against  a 
world  of  facts  is  set  a  world  of  values. 

But  if  this  growing  consciousness  of  the  problem  of  value  has 
indeed  reached  a  point  where  we  are  conscious  of  a  world  of 
values,  where  the  terms  ethical,  aesthetic,  and  even  "  truth  " 
values,  are  in  every  mouth,  and  where  the  thought  of  a  special 
"  theory  of  value  "  is  no  longer  novel,  with  it  has  also  come  the 
realisation  that  philosophy,  and  the  philosophical  disciplines 
which  are  traditionally  concerned  with  values,  are,  in  their 
present  form,  not  quite  in  a  position  to  take  possession  of  the  new 
world.  It  is  true  that  for  some  time  metaphysics  has  seemed  to 
many  to  be  but  a  theory  of  value  ;  but  the  traditional  problems 

1  This  quotation  is  taken  from  the  "  Prospekt"  of  Die  Gesellschaft  (Verlag 
der  Literarischen  Anstalt,  Ruten  &  Loenig,  in  Frankfurt  A.  M.),  a  collection  of 
social-psychological  monographs  in  which  the  various  institutions  of  society  are 
studied  from  the  point  of  view  of  their  values  for  the  individual.  Some  of  the 
titles  are  Religion,  Speech,  Custom,  Commerce,  The  State,  Politics,  War,  The 
Strike,  etc. 


Introduction  3 

as  well  as  the  traditional  methods  of  that  discipline  are  still  such 
as  to  make  the  question  of  values  subordinate  to  the  question 
of  "  being."  Nor  are  the  special  sciences  which  deal  with 
facts  of  value  able,  as  such,  to  cope  with  the  changes,  in 
both  form  and  content  of  discussion,  which  this  new  setting  of 
the  problem  has  brought  about.  An  harmonious  division  of 
labour  between  economics,  ethics,  and  aesthetics  has  produced 
results  which,  for  various  and  sufficient  reasons,  do  not  meet 
the  need.  It  is  rather  precisely  because  of  this  division  of 
labour,  unwisely  conceived,  that  the  results  are  unsatisfactory. 
More  and  more  the  conviction  gains  ground  that  a  general  theory 
of  value,  which  shall  comprehend  in  a  systematic  and  scientific 
way  all  types  of  human  values,  is  an  absolute  necessity. 

II.  THE  SOURCES  OF  SUCH  A  THEORY 

It  has  been  said  that  the  most  fruitful  metaphysical  thought 
of  the  present  is  to  be  found  in  the  special  sciences.  While 
perhaps  not  quite  true,  such  a  statement  has  this  element  of 
truth,  that  it  is  within  the  special  sciences  that  the  most  signifi 
cant  questions  of  philosophy  first  make  their  appearance.  Simi 
larly,  the  necessity  of  solving  certain  special  questions  of  value 
within  the  sciences  of  economics,  ethics,  and  aesthetics,  has 
developed  concepts  the  significance  of  which  extends  far  beyond 
these  limits,  and  which  therefore  afford  the  material  for  more 
general  and  systematic  reflections. 

Of  first  importance  is  the  "  theory  of  value  "  which  economics 
has  developed  for  its  special  purposes.  Narrow  as  this  theory 
is  (for  it  is  not  so  long  ago  that  an  economist,  F.  von  Wieser, 
although  of  the  opinion  that  he  had  fulfilled  his  intention  of 
' '  treating  exhaustively  the  entire  sphere  of  worth  phenomena  with 
out  an  exception,"  did  not  once  in  his  investigations  go  beyond 
the  region  of  economic  goods),  nevertheless,  the  very  limitation  of 
its  activities  to  a  narrow  range  of  problems  has  led  to  an  intensive 
analysis  of  certain  facts  and  laws  of  valuation  which  should  have 
long  since  furnished  an  example  to  ethics,  and  which  must  now 
furnish  both  the  stimulus  and  the  discipline  for  any  one  who  seeks 
to  comprehend  the  larger  field.  But  this  limitation  of  interest 
has  obscured  wider  relations,  knowledge  of  which  would  have 
been  fruitful  for  the  special  work  of  the  economist  himself,  and, 
in  some  cases,  has  led  to  fallacies  of  both  observation  and  infer 
ence,  which  a  more  philosophical  treatment  of  facts  would  have 


4  Valuation  :   its  Nature  and  Laws 

corrected.  These  limitations  are,  however,  being  overcome.  The 
necessity  of  translating  economic  into  sociological  conceptions, 
of  correlating  economic  with  larger  social  values,  has  brought 
about  a  notable  change.  Indeed,  much  of  the  movement  in  the 
direction  of  a  more  general  theory  comes  from  economics  itself. 
Gradually  the  opposition  to  theory  in  this  sphere  is  giving  way, 
and  at  the  same  time  the  feeling  increases  that  economic  values 
are  but  a  special  class  of  human  worths,  and  that  they  can  be 
understood  only  in  their  relations,  especially  in  their  relation  to 
ethical  values.1 

Ethics,  likewise,  has  its  contributions  to  make  to  a  general 
theory  of  value.  Chief  among  these  are  its  appreciative  analyses 
and  descriptions  of  qualitatively  different  attitudes  and  disposi 
tions,  and  its  elaboration  of  a  doctrine  of  the  norms  of  obligation 
and  virtue  in  which  the  appreciative  distinctions  of  the  race  have 
been  fixed.  To  this  must  be  added  the  development  of  hypotheses 
as  to  the  nature  of  the  ultimate  good,  which,  while  they  have 
not  led  to  any  final  solution,  have  nevertheless  served  to  develop 
and  organise  the  normative  point  of  view.  But  it  is  precisely 
because  of  this  preoccupation  with  ultimate  norms  and  abstrac 
tions  that  ethics  is  in  no  position  to  meet  the  advances  of  econom 
ics.  For  ethics,  as  it  is  commonly  understood,  still  remains 
too  much  in  the  traditions  of  the  Greeks,  and,  instead  of  seeking 
a  theory  of  value  founded  upon  an  adequate  psychology,  con 
tents  itself  with  a  theory  of  abstract  goods,  consisting  in  an 
external  and  often  arbitrary  classification  and  evaluation  of 
objects  of  desire  without  a  sufficiently  vital  sense  of  the  great 
problems  involved  in  the  processes  and  laws  of  desire  themselves. 

Especially  harmful,  moreover,  has  been  the  Kantian  distinc 
tion  between  the  "  empirical  "  and  the  "  intelligible  "  will,  and 
the  narrowing  effect  of  the  concept  of  abstract  imperatives. 
Although  no  longer  held  in  its  original  form,  it  still  exercises 
influence  through  the  unfortunate  antithesis  of  facts  and  values, 
of  genesis  and  validity.  For  where  such  distinctions  are  made 
ultimate,  where  the  laws  of  the  empirical  will  are  conceived  to 
be  irrelevant,  or  even  hostile,  to  the  will  that  values,  there  a 
science  of  values  is  impossible. 

Where,  on  the  other  hand,  ethics  has  broken  loose  from  these 
bonds,  the  new-found  freedom  has  given  rise  to  such  a  multitude 
of  irreconcilable  principles  that  it  is  immediately  apparent  that 

1  Compare  in  this  connection   Hadley's  article  on  "  Economic  Science "  and  the 
present  writer's  article  on  "  Worth  "  in  the  Dictionary  of  Philosophy  and  Psychology. 


Introduction  5 

the  certainty  of  method,  which  makes  possible  internal  unity 
of  principles  and  harmonious  external  relations  with  other 
sciences,  is  still  lacking.  It  has  even  been  seriously  doubted 
whether  ethics  can  maintain  its  place  as  a  special  science — 
whether  it  is  not  doomed  to  break  up,  on  the  one  hand,  into  a 
part  of  psychology,  the  task  of  which  shall  be  to  analyse  the 
individual  feelings,  judgments,  and  acts  of  will,  the  content  of 
which  has  the  moral  predicate,  and,  on  the  other  hand,  into  a 
part  of  sociology,  which  shall  portray  the  forms  and  content  of 
the  common  life  which  stand  in  relations  to  the  ethical  obligation 
of  the  individual.  Its  double  character  will,  it  is  thought,  ulti 
mately  prove  its  undoing.1 

Doubtful  though  such  predictions  may  rightly  be  held  to 
be — for  the  boundaries  of  sciences  are  determined  by  other 
motives  than  those  of  mere  logic,  and  there  are  practical  reasons 
which  will  plead  strongly  for  the  integrity  of  ethics  as  a  separate 
discipline,  still  there  can  be  no  doubt  that  the  inconsequent 
character  of  the  science,  in  its  present  state,  unfits  it  for  leader 
ship  in  the  attempt  to  conceive  valuation  in  its  more  general 
aspects.  Like  economics  it  has,  to  be  sure,  recently  been  looking 
beyond  its  narrowly  conceived  province,  and  seeking  points  of 
contact  with  its  neighbours— the  breaking  up  of  its  solidarity  is, 
in  one  sense,  but  an  outward  sign  of  an  inward  grace ;  but  this 
is  in  itself  not  sufficient  to  make  of  ethics  the  science  of  values 
par  excellence. 

Nor  is  such  a  science  to  be  developed  by  a  merely  external 
fusion  of  elements  from  both  of  the  preceding  sciences,  with 
perhaps  the  addition  of  a  few  judicious  reflections  upon  aesthetic 
and  religious  values.  To  meet  the  obvious  necessities  of  the 
situation  there  is  required,  rather,  a  systematic  treatment  of 
human  values  in  their  mutual  relations,  together  with  the  psy 
chology  of  feeling  and  will  upon  which  such  a  theory  must  rest. 
What  is  needed  is  a  point  of  view  and  method  which  shall  go 
beyond  the  special  motives  of  economics  and  ethics,  and  thus  find 
common  ground  in  a  conception  and  purpose  which  unite  them 
both.  Thus,  while  economics  has  been  thought  to  be  a  descriptive 
and  explanatory  science,  and  has  contented  itself  with  description 
of  the  empirical  laws  of  valuation  for  the  purposes  of  control,  it 
has  really  been  shot  through  with  assumptions  of  a  normative 
character,  and  has  been  fruitful  in  disclosing  actual  standards  of 
value  which  ethics  has  often  failed  to  estimate  at  their  proper 

1  Simmel,  Einleitung  in  die  Moralwissenschaft,  Berlin,  1893,  Vol.  I,  Preface. 


6  Valuation  :   its  Nature  and  Laws 

worth.  On  the  other  hand,  ethics,  although  claiming  to  be  a 
normative  science,  has  found  it  necessary  to  investigate  the 
phenomenology  of  feeling  and  will,  without,  however,  as  I  shall 
seek  to  show  later,  succeeding  in  making  these  investigations 
sufficiently  fruitful  for  its  more  ultimate  purposes.  The  de 
sideratum,  therefore,  seems  to  be  to  find  a  method  which  shall 
unite  in  some  more  fruitful  way  the  descriptive  and  the  norma 
tive  points  of  view,  a  method  which  shall  know  how  to  interpret 
the  norms  of  the  so-called  "  intelligible  "  will  in  terms  of  the 
laws  of  the  "  empirical  "  will. 

Is  SUCH  A  SCIENCE  POSSIBLE  ? 

The  preceding  statement  of  the  problem  of  a  general  theory 
of  value  shows  it  to  consist  in  two  main  problems,  closely  con 
nected,  the  descriptive  or  psychological  and  the  normative  or 
axiological.  And  such  a  conception  of  the  problem  seems 
necessitated  by  the  facts  with  which  we  are  concerned.  For 
the  function  of  valuation  has  two  aspects.  On  the  one  hand, 
we  fed  the  values  of  objects  ;  on  the  other  hand,  we  evaluate 
these  objects  and  ultimately  the  experiences  of  value  them 
selves.  The  first  aspect  is  a  process,  the  conditions  and  laws 
of  which  are  to  be  determined  ;  the  second  is  a  function,  the 
meaning  and  norms  of  which  are  to  be  developed.  To  a  pre 
liminary  characterisation  of  these  two  problems  we  might  now 
proceed  immediately,  were  it  not  for  the  dogma  of  the  anti 
thesis  of  the  "  intelligible  and  empirical "  will  which,  in  its 
various  forms,  has  stood  in  the  way  of  a  science  of  values,  and 
must  therefore  receive  brief  consideration. 

This  dogma,  appearing  under  various  names,  now  as  the 
antithesis  of  genesis  and  validity,  again  as  the  antithesis  of  facts 
and  values,  has  become  especially  familiar,  not  to  say  insidious, 
in  its  latter-day  formulation,  Appreciation  versus  Description. 

It  is  not  difficult  to  understand  the  motives  of  the  antithesis 
in  its  present  form.  The  gradual  usurpation  of  the  whole 
field  of  description  by  certain  specialised  scientific  methods 
brought  with  it  inevitable  disappointment.  The  psycho-physical 
and  biological  methods,  approaching  as  they  did  the  problem 
from  without,  and  finding  irrelevant  all  aspects  of  experience 
except  such  as  could  be  connected  with  the  conceptions  of  these 
sciences,  soon  showed  their  inadequacy  as  means  of  describing 
experiences  of  value.  The  simplest  solution  of  the  difficulty 


Introduction  7 

seemed,  therefore,  to  be  in  looking  upon  values  as  merely  ap 
preciable  and  not  communicable  in  terms  of  any  objective 
description.  Value  is  always  the  meaning  of  an  attitude  of  a 
subject,  and  is  therefore  not  describable  in  terms  of  mental 
elements.  An  attitude  can  be  merely  appreciated. 

It  would  seem  that  the  antithesis  is  falsely  conceived,  and 
that  it  arises  primarily  from  the  fact  that  we  have  to  do  here 
with  a  false  way  of  setting  the  problem.  Instead  of  going  directly 
to  facts,  the  point  of  view  here  disclosed  starts  with  a  wholly 
arbitrary  and  narrow  conception  of  description.  Having  assumed 
this,  and  finding  a  mass  of  experience  which  escapes  its  categories, 
the  logic  of  the  situation  leads  to  the  conclusion  that  there  is 
appreciation  without  description.  Let  us  first  consider  the 
abstract  merits  of  the  antithesis,  and  then  we  may  turn  to  a 
critical  examination  of  the  concept  of  scientific,  and  therefore 
psychological,  description  which  underlies  it.  From  this  we  may 
be  able  to  determine  the  function  of  psychology  in  a  general 
theory  of  value. 

As  a  preliminary  distinction  the  antithesis  does  well  enough. 
For  a  moment,  perhaps,  one's  appreciation  seems  to  be  one's 
own  "  incommunicable  dream,"  but  the  need  of  participating 
with  others  in  the  social  concourse  presses  upon  us  the  necessity 
of  objectifying  our  experience,  of  searching  for  presentations 
with  which  the  experience  may  be  connected.  Through  them 
the  attitude  becomes  objectified  to  consciousness  and  communi 
cated  to  others,  and  behold  appreciation  has  itself  increased. 
The  very  condition  of  continuous  and  progressive  appreciation 
is  some  sort  of  description. 

It  is  no  less  true  that  there  can  be  no  description,  even  the 
most  scientific,  without  an  appreciative  element.  Here  again 
the  ideal  of  a  scientific  description  without  the  element  of 
appreciation  is  merely  an  ideal  limit,  set  for  certain  purposes, 
but  a  limit  not  realised  in  actual  experience.  It  would  not  be 
difficult  to  show  that,  when  in  any  science  we  make  abstractions, 
the  direction  and  extent  of  these  abstractions  are  determined 
by  an  act  of  appreciation.  All  abstraction  is  in  the  last  analysis 
purposive.  Whether  the  product  of  our  abstraction  is  in  any 
sense  the  concrete  thing  with  which  we  started,  or  has  any  useful 
relation  to  it,  is  finally  to  be  decided  by  an  act  of  appreciation. 

So  much  for  the  antithesis  in  its  general  form.  The  twofold 
assumption  that  there  may  be  appreciation  without  some  form 
of  description,  or  description  without  an  ultimate  moment  of 


Valuation :   its  Nature  and  Laws 

appreciation,  proves  untenable  in  both  aspects.  It  is  true  that 
there  may  be  a  blind  sort  of  feeling  of  significance  before  the  rise 
of  any  cognitive  acts  objectifying  the  experience,  but  that  feeling, 
with  all  its  brute  immediacy,  has  scarcely  reached  the  level  of  ex 
plicit  appreciation  or  feeling  of  value.  It  is  also  true  that  the  total 
meaning  of  an  appreciation  is  never  exhausted  in  any  description. 
There  is  always  an  element  which  just  escapes.  But  some  of  the 
meaning  is  conserved ;  otherwise  it  is  not  description.  We  may, 
it  is  true,  describe — satisfactorily  for  certain  limited  purposes— 
and  at  the  same  time  ignore  certain  aspects  of  the  total  appre 
ciation,  but  never  all,  otherwise  it  is  no  description. 

From  these  general  conclusions  more  specific  inferences  may 
be  drawn  of  immediate  and  practical  importance  for  the  dis 
cussions  that  follow.  In  the  first  place,  there  are  varying  types 
of  description  of  any  phenomenon,  types  determined  by  the 
purposes  of  the  description,  and  therefore  by  the  degree  of 
appreciation  retained  in  the  descriptive  terms.  The  antithesis 
between  appreciation  and  description  is  accordingly  reducible 
to  a  distinction  between  two  types  of  description,  appreciative 
and  scientific,  and  we  may  probably  infer  that  there  is  at  least 
an  appreciative  description  of  experiences  of  value. 

In  another  connection *  I  have  sought  to  show  that  there  is 
such  appreciative  description — and  communication — of  individual 
experiences  of  value,  and  to  develop  its  characteristics  and  princi 
ples.  A  brief  summary  _of  the  conclusions  will  be  sufficient 
for  the  present  purpose.  rSuch  description  has  as  its  object  the 
communication  and  objectification  of  the  intrinsic  meaning  of 
individual  experience  in  the  interest  of  facilitation,  either  con 
servation  or  increase,  of  appreciation.  This  communication 
and  description  are  accomplished  through  connection  of  individual 
experiences  with  ideal  psychical  objects,  already  shared  and  over- 
individual,  projected  affective  -  volitional  meanings  embodied 
in  ideal  persons  and  states.  Through  identification  or  contrast 
with  these  the  individual  experience  is  communicated,  both  in 
its  quality  and  degree.  Such  communication  and  description 
is  essentially  norm  construction.  For  the  ideal  object,  thus  pro 
jected  and  shared,  contains  the  funded  meaning  of  past  experi 
ences,  and  constitutes  not  only  the  presupposition,  the  medium, 
of  all  communication  of  present  experience,  but  also  the  norm 
of  its  control. 

1  "Appreciation  and   Description  and   the   Psychology   of  Values,"   Philosophical 
Review,  November,  1905. 


Introduction  .    9 

When  we  ask  what  it  is  that  this  appreciative  description 
seeks  to  communicate,  aad  what  it  is,  according  to  the  view  criti 
cised,  thalrraTHiot-be-eomnittnicated'and  described,  we  find  it  to  be 
certain  references  of  the  attitude  beyond  the  present  experience, 
meanings  acquired  in  individual  and  social  processes.  These 
are  always  references  of  the  present  state  to  something  pre 
supposed.  They  may  be  described  in  general  terms  as  trans- 
gredient  and  immanental.  The  transgredient  reference,  as 
expressed,  for  example,  in  such  appreciative  categories  as  obliga 
tion  and  desert,  is  a  present  feeling,  but  includes  in  it  a  reference 
beyond  the  present  state.  The  immanental  reference — as,  for 
instance,  the  worth  suggestions  of  aesthetic  states — is  a  present 
state,  referring,  not  beyond  the  present  state,  but  to  something 
more  deeply  implicit,  something  presupposed  in  it.  These  re 
ferences,  communicated,  as  we  have  seen,  by  connection,  either 
through  identification  or  contrast,  with  projected  ideals  or  norms, 
really  point  to  the  psychical  processes  in  which  these  norms 
were  constructed  and  in  which  their  meaning  was  acquired. J 
Whether  these  processes  can  be  described,  whether  the  genesis 
of  these  attitudes  and  meanings  can  be  reconstructed,  is  the 
problem  of  the  psychology  of  worth  experience,  and  this  includes 
the  question  of  the  relation  of  appreciative  to  scientific 
description. 

III.  THE  PSYCHOLOGICAL  PROBLEM  AND  METHOD 

The  first  task  of  a  general  theory  of  value  is  psychological 
analysis.  Strictly  speaking  there  is  no  problem,  scientific  or 
unscientific,  which  does  not  have  its  psychological  side.  Not 
only  the  questions,  but  also  the  objects  in  connection  with  which 
these  questions  arise,  belong  in  the  first  place  to  the  psychical 
life.  It  is  further  apparent  that  the  fact  of  value  itself  cannot  be 
described  otherwise  than  by  reference  to  determinations  which 
are  taken  from  the  psychical  life  and  which  therefore  belong  to 
psychology.  The  most  convincing  proof  of  this  is  the  fact 
that  it  has  never  occurred  to  economics,  which  is  in  the  main 
free  from  abstract  questions  of  methodology,  to  attempt  to 
define  the  nature  of  value  without  reference  to  psychology. 
Reflections  of  this  nature  would  seem  to  lead  to  the  inference 
that  it  is  in  psychology  and  its  analyses  that  we  must,  in  the  first 
place,  look  for  those  general  categories  of  description  which  shall 
form  the  basis  of  a  general  theory  of  value,  and  for  the  general 


io  Valuation:   its  Nature  and  Laws 

laws  of  psychical  process  which  shall  enable  us  to  correlate  the 
different  types  of  valuation.  Such,  indeed,  is  a  plain  inference 
from  the  facts,  but  here  again  certain  a  priori  theories  of  the 
nature  and  function  of  psychology  call  a  temporary  halt. 

Upon  the  general  question,  as  it  relates  to  psychology  as  a 
whole,  there  are  a  multitude  of  counsels  at  the  present  time. 
There  are  those  who  see  in  psychology  and  its  descriptions 
chiefly  a  propaedeutic  to  the  interpretation  and  appreciation  of 
actual  psychical  reality,  the  categories  of  which  are  teleological. 
It  is  the  fundamental  Geisteswissenchajt.  It  is  upon  the  basis 
of  such  a  conception  of  psychological  purpose  and  method  alone 
that  we  may  ascribe  to  psychology  the  role  of  the  science  of 
abstract  mental  laws  which  shall  make  possible  the  interpretation 
of  the  concrete  mental  reality  with  which  the  sciences  of  ethics, 
aesthetics,  etc.,  are  concerned.  In  direct  opposition  to  this  view, 
both  historically  and  logically,  is  the  view  which  denies  the  possi 
bility  of  description  except  through  connection  of  the  psychical 
with  physical  objects,  and  which  therefore,  in  view  of  the  artificial 
transformation  of  the  psychical  which  results,  also  denies  its  func 
tion  as  interpreter  of  the  psychical  objects  of  ethics  and  aesthetics. 
Finally,  there  are  those  who,  while  perhaps  not  sure  as  to  the 
precise  logical  basis  for  the  recognition  of  two  distinct  types  of 
method  within  the  same  science,  are  yet  forced  by  a  broad 
view  of  the  facts  to  recognise  two  distinct  purposes  in  the  re 
constructions  of  psychology,  the  one  having  as  its  end  the  con 
struction  of  abstract  concepts  which  shall  be  instrumental 
in  the  interpretation  of  actual  historical  psychical  reality  as  a 
process  of  acquirement  of  meaning,  the  other  the  control  of  the 
psychical  through  its  connection  with  mechanical  process. 

Upon  any  conception  of  the  function  of  psychology  other  than 
the  second,  the  facts  of  worth  experience  are,  as  such,  the 
material  of  psychology.  That  view  is  obviously  but  a  special 
application  of  the  general  antithesis  between  appreciation  and 
description,  and  must  in  the  last  analysis  share  its  fate,  but  a 
brief  consideration  of  this  special  expression  of  the  dogma  will 
clear  the  way  for  our  positive  conceptions  of  the  function  of 
psychology  in  a  general  theory  of  value. 

Briefly  it  runs  as  follows.  All  description  and  explanation 
have  as  their  motive  the  communication  and  ultimately  the  control 
of  experience.  Such  communication  and  control  of  the  subjective 
and  individual  are  possible  only  through  linkage  with  objects 
which  have  common  meaning  and  which,  therefore,  are,  through 


Introduction  1 1 

abstraction,  as  far  removed  as  possible  from  individual  apprecia 
tions.  The  only  objects  which  fulfil  these  conditions  are  the 
physical  objects,  the  abstractions  of  physical  science  which  have 
lost  all  intrinsic  meaning  and  have  become  wholly  instrumental. 
The  ideal  of  psychology  is  therefore  connection  of  the  psychical 
with  physical  objects,  and  only  in  such  connection  do  we  have 
description  and  explanation  which  merits  the  name  of  scientific. 

The  consequences  of  this  are  obvious.  For  if  the  only 
description  which  merits  the  name  scientific  is  connection  with 
physical  objects,  then  in  the  reconstruction  of  our  immediate 
appreciations  abstraction  must  be  made  from  all  appreciative 
moments  in  the  psychical,  and  the  immediate  experience  must 
be  broken  up  into  non-appreciative  elements,  preferably  sensa 
tions,  which  may  be  connected  with  the  non-appreciative  elements 
of  the  physical  construction.  What  this  means  for  the  psychology 
of  those  aspects  of  the  psychical  which  form  the  basis  of  worth 
experience  is  evident.  Feeling  and  will,  the  basis  of  this  experi 
ence,  intend,  in  both  the  transgredient  and  immanental  references 
of  their  states,  psychical  objects  as  well  as  physical,  and  can 
communicate  these  intents,  these  acquired  meanings,  only  through 
connection  with  such  ideal  objects.  These  objects  are,  however, 
always  projected  will  and  feeling,  and  scientific  description,  if  it 
is  of  the  nature  assumed,  can  make  no  use  of  these  psychical 
objects,  and  therefore  no  use  of  the  abstract  conceptions  of  feeling 
and  will  in  its  reconstructions.  Such  continuity  as  it  may 
establish  is  not  psychical,  but  must  be  wholly  in  terms  of  physio 
logical  dispositions.  If  this  view  of  psychological  description  is 
justified,  Miinsterberg  has  drawn  the  only  conclusion  possible, 
viz.,  that  there  is  no  psychology  of  the  worth  experience  possible, 
and  therefore  no  relation  between  appreciative  and  scientific 
description. 

Clearly  the  whole  question  revolves  about  the  more  ultimate 
problem  of  the  purpose  of  psychological  description.  The 
question  is  not  whether  physical  objects  are  the  only  media 
for  any  description — we  have  seen  that  they  are  not,  but 
rather  whether  they  are  the  most  useful  for  the  purposes  of 
psychology.  That  there  are  other  objects  than  the  physical 
through  which  communication  and  description  are  possible,  we 
have  seen.  If  our  initial  assumption,  that  appreciation  without 
description  and  description  without  appreciation  are  but  abstrac 
tions  and  ideal  limits,  that  all  concrete  thought  activities  con 
tain  both  moments  in  different  degrees,  is  valid,  then  there  can 


1 2  Valuation :   its  Nature  and  Laws 

be  scientific  constructions  making  use  of  terms  in  which  the 
process  of  abstraction  from  appreciative  connotation  shows 
different  degrees  of  completeness,  according  as  the  purposes  of 
the  reconstruction  require.  It  follows  that  the  absolute  contrast 
between  appreciative  and  scientific  description  disappears,  and 
we  have  left  merely  the  practical  problem  of  the  degree  to  which 
appreciative  differences  shall  be  retained  in  our  constructions. 
This  is  wholly  a  question  of  the  purpose  of  the  description. 

Historically,  and  in  present  practice  in  so  far  as  it  is  fruitful, 
the  motive  of  psychology  is  primarily  one  of  interpretation. 
The  region  of  possible  control  of  mind  through  its  connections 
with  the  body,  although  we  cannot  limit  it  a  priori,  is  small 
indeed  in  comparison  with  the  regions  of  possible  interpretation 
through  psychical  conceptions.  Even  in  the  region  where  the 
motive  is  primarily  one  of  direct  control,  the  conclusion  is  rapidly 
gaining  ground  that  control  is  possible  only  through  appreciative 
interpretation  of  the  mental  life.  Especially  noteworthy  is  the 
change  of  view  in  the  field  of  mental  pathology,  where  the 
necessities  of  practice  are  leading  to  a  reaffirmation  of  the 
fundamental  conceptions  of  emotion  and  conation,  as  against 
the  purely  neurological  conceptions.1  Even  if  this  were  not  so, 
it  is  impossible  to  ignore  the  larger  region  where  the  function 
of  psychology  is  wholly  interpretative.  And  here  the  motive 
determines  the  method.  For  concepts,  in  order  to  be  instru 
mental  for  interpretation,  must  conserve,  and  contain,  at  least 
implicitly,  the  acquired  meaning  which  they  seek  to  describe. 
The  explanation  must  be  in  functional  terms,  and  functional 
terms  are  in  the  last  analysis  refinements  of  appreciative  descrip 
tion.'2  The  question  whether  there  is  any  relation  between 

1  Compare  in  this  connection  the  psychological  conceptions  of  Janet,  i.e.  of  emotion 
and  conation,  with  the  neurological  of  Wernike  and  Ziehen.  Also  the  introductory 
paragraphs  of  James's  Presidential  Address,  The  Energies  of  Men. 

8  Moreover,  with  the  growing  recognition  of  the  close  relation  of  economic  objects 
and  values  to  other  psychical  objects  and  their  values,  already  referred  to,  comes  the 
recognition  of  the  fact  that  the  psychology  of  values  is  concerned  with  the  interpretation 
of  individual  and  social  worth  processes  and  only  indirectly  with  their  control.  With 
this  recognition  of  the  interpretative  function  comes  the  necessity  of  the  use  of  terms 
which  may  be  instrumental  in  interpretation,  terms  with  appreciative  connotation.  As 
soon  as  the  economic  philosopher  seeks  to  use  his  constructions  as  a  means  of  interpreta 
tion  of  concrete  reality,  to  connect  economic  with  ethical  and  nssthetic  worths,  he  must 
restore  the  appreciative  aspects.  Interesting  illustrations  of  this  appear  in  the  works 
of  Veblen — A  Theory  of  the  Leisure  Class  (Macmillan)  and  A  Theory  of  Business 
Enterprise  (Scribner),  one  of  the  most  significant  aspects  of  whose  method  consists  in 
making  technical  essentially  appreciative  terms,  and  more  marked  still  in  Simmel's 
Philosophic  des  Geldes.  It  is  an  interesting  fact  that  while  official  psychology  in  some  of 
its  tendencies  seeks  to  exclude  all  appreciative  descriptions,  the  economic  sciences  are 
becoming  more  psychological  by  restoring  them. 


Introduction  1 3 

appreciative  and  scientific  description  is  then  the  better  known 
problem  whether  psychology  should  be  a  "  functional  "  or  a 
"  content  "  psychology.  For  the  latter  a  psychology  of  value 
is  impossible  ;  for  the  former  it  is  possible,  and  what  is  more, 
a  present  fact. 

When,  therefore,  we  narrow  this  general  problem  of  psycho 
logical  method  to  the  more  specific  question  of  the  psychology  of 
worth  experience,  it  is  possible  to  draw  certain  inferences  as  to 
method  which  find  substantiation  in  the  actual  procedure  of  psy 
chology.  In  the  first  place,  values  are  facts,  to  be  described  as  any 
other  mental  facts.  The  sharp  antithesis  of  facts  and  values  might 
temporarily  delay  the  appeal  to  psychology,  but  the  simple  and 
inevitable  necessity  of  the  situation — that  every  assertion  of  a 
worth  involves  likewise  the  assertion  of  its  conformity  with  actual 
or  possible  experiences  of  feeling  and  will — makes  the  appeal  to 
psychology  ultimately  unavoidable.  But  the  recognition  of  this 
abstract  proposition  does  not  help  us  until  we  have  a  clearer 
conception  of  psychical  fact.  Whether  psychological  analysis 
can  serve  as  an  instrument  of  interpretation  and  appreciation 
of  values  depends  upon  what  we  conceive  its  function  to  be. 
Here  the  principle  maintained  throughout  this  discussion  must 
be  final.  If  experience  is  conceived  to  be  cognisable  as  a  fact 
only  when  it  is  viewed  as  a  passive  state  loosed  from  all  relations 
with  the  object  which  it  cognises  and  appreciates,  then  the 
experience  of  value  cannot  be  cognised  in  this  sense.  If,  on  the 
other  hand,  psychology  is  a  science  which,  although  it  makes 
use  of  abstractions,  still  deals  with  reality,  then  a  value  can  be 
cognised  as  a  fact,  as  well  as  appreciated.  To  cognise  it  as  a  fact, 
related  in  certain  uniform  ways  to  facts  of  the  same  order, 
requires  that  it  shall  be  subsumed  under  general  concepts  ; 
but  in  order  that  these  concepts  shall  really  define  it,  they  must 
have  meanings  common  to  all  appreciation.  Psychological 
description  must  then  start  with  appreciative  description.  To 
describe  thus  appreciatively  a  valuation,  its  meaning  as  an 
attitude  must  be  communicated ;  but  this  meaning  can  be  com 
municated  only  by  connecting  it  with  the  ideal  objects  toward 
which  the  attitude  is  taken,  and  by  characterising  the  pre 
dominant  moment  in  the  attitude,  whether  feeling  or  will,  and 
the  cognitive  presuppositions  which  determine  it.  Consequently, 
when  we  attempt  to  describe  and  classify,  and  to  derive  genetically 
the  various  attitudes  in  valuation,  we  must  retain  in  our  abstrac 
tions  such  concepts  as  feeling  and  will,  functional  terms  which 


14  Valuation:   its  Nature  and  Laws 

have  still  enough  of  appreciative  connotation  in  them  to  be 
instrumental  in  the  interpretation  of  appreciations. 

Here  then  finally  we  see  the  relation  of  appreciative  to 
scientific  psychological  description.  Appreciative  description 
communicates  the  meanings  of  feelings  acquired  through  con 
nection  with  psychical  objects  toward  which  the  feeling  is 
directed.  These  objects  are,  as  such,  not  the  material  of  psy 
chology  alone  any  more  than  are  the  physical  objects.  As 
objects  they  belong  to  the  normative  sciences.  But  while  they 
are  not  the  material  of  psychology  alone  —  inasmuch  as 
they  are  projections  beyond  the  individual,  nevertheless  the 
processes  by  which  they  have  been  projected  and  objectified 
and  the  processes  by  which  the  individual,  when  once  they  have 
become  psychical  objects,  participates  in  their  meanings — in 
other  words,  the  presuppositions,  conative  and  cognitive,  which 
determine  his  feeling  attitudes  toward  them — are  distinctly 
the  material  of  psychological  study. 

THE  PRESUPPOSITIONAL  METHOD — A  FORM  OF 
GENETIC  METHOD 

The  method  of  psychological  worth  analysis  we  may  then 
characterise  as  the  Presuppositional  Method.  It  begins  with 
analysis  of  presuppositions.  The  key  to  the  position  is  to  be 
found  in  the  fact  that  worth  experience  is  always  an  attitude. 
An  attitude  is  an  immediate  experience  which  contains  in  it  a 
reference — in  our  terms,  either  transgredient  or  immanental — to 
presupposed  psychical  process.  Its  determinants  are  the  actual 
cognitive  acts  of  the  present  experience — for  all  worth  ex 
perience  presupposes  some  form  of  cognition  of  reality — and  the 
conative  dispositions,  the  assumptions  and  postulates,  which 
form  the  platform  of  the  present  experience.  The  varying 
worth  attitudes  must  be  defined  in  terms  of  their  presupposi 
tions,  actual  and  dispositional.  But  further,  if  any  worth 
attitude,  when  viewed  thus  psychologically,  is  an  immediate 
feeling  plus  the  acquired  meaning  or  reference,  then  this 
reference,  which  is  for  immediate  appreciation  the  sign  of  worth 
continuity,  must  find  a  psychological  interpretation  in  a  con 
tinuity  which  is  psychical,  and  not  in  one  established  indirectly 
through  connection  with  physiological  dispositions.  With  this 
our  method  becomes  genetic. 

The  genetic  method  in  psychology,  broadly  viewed,  is  capable 


Introduction  1 5 

of  different  formulations.  Thus  Baldwin,  Paulhan,  and  Stout, 
although  making  the  idea  of  development  fundamental,  make 
use  of  somewhat  different  principles.  Accommodation  and 
Habit,  Systematisation  and  Arrest  of  Conative  Tendencies  — 
such  are  the  differing  functional  conceptions  with  which  they 
work.  But  the  conception  which  underlies  them  all  is,  I  think  it 
may  he  safely  said,  that  the  progressive  differentiation  and 
correlation  of  the  content  of  consciousness,  by  which  new  mean 
ings  are  acquired  and  appreciatively  distinguished,  must  be 
referred  for  explanation  to  functional  readjustments  of  conscious 
ness  as  an  organic  whole.  Different  levels  of  meaning  are  thus 
distinguished  and  transitions  from  one  level  to  another  accounted 
for.  One  of  the  most  important  consequences  of  this  genetic, 
functional  method  is  that  what  appears  from  one  point  of  view 
as  habit  or  disposition,  may  in  a  new  adjustment  find  a  place  in 
the  content  as  a  meaning ;  so  that  the  functional  meaning  of 
disposition,  viewed  as  habit,  is  conserved  on  the  new  platform 
of  accommodation  in  the  new  arrangement  of  content.  This 
unification  of  functional  habit  and  content  under  the  genetic 
categories  finds  its  most  notable  expression,  on  the  cognitive 
side  of  experience,  in  the  doctrine  of  the  general  concept  it  is  able 
to  contribute.  Whereas  from  the  purely  analytical  point  of  view 
the  general  concept  finds  no  satisfactory  psychological  equivalent, 
from  the  genetic  its  functional  meaning  receives  due  recognition. 
"  Selective  thinking  may  then  be  viewed  as  the  systematic  or 
progressive  and  continuous  determination  of  the  stream  of 
thought  in  the  individual's  mind."1 

When  now  this  genetic  method  is  applied  to  the  worth  aspect 
of  consciousness,  a  similar  problem  presents  itself — how  valuation 
may  be  conceived  as  a  continuous,  progressive,  and  systematic 
determination  of  the  stream  of  conation  and  feeling  in  the 
individual's  mind.  Here  again,  psychical  continuity  is  the 
important  point,  and  the  special  form  in  which  it  here  appears 
stands  in  close  connection  with  the  problem  of  the  derivation 
of  the  different  attitudes  of  appreciation  as  genetic  modes  of  one 
continuous  process  of  acquirement  of  meaning — of  showing  how 
the  acquired  meaning  of  one  attitude, having  become  dispositional, 
functions  as  an  assumption  or  presupposition  of  new  feelings 
and  modes  of  valuation.  This  involves  the  concept  of  psy 
chical  development  or  value-movement,  the  conditions  and  laws 

1  Baldwin   in  the   Dictionary  of  Philosophy  and  Psychology,   Vol.    II,   article  on 
"Selective  Thinking." 


1 6  Valuation  :   its  Nature  and  Laws 

of  which  must  be  explained  in  terms  of  still  more  general  laws 
of  feeling  and  will. 

The  presuppositional  method  as  thus  described,  lies  midway,  so 
to  speak,  between  the  teleological  analysis  and  explanation  of  the 
normative  sciences — which  assume  an  end  or  ends  as  the  in 
struments  of  analysis  of  the  stages  of  meaning  and  the  ordering 
of  the  psychical  objects,  and  the  causal  method  which  abstracts 
from  all  meaning,  and  may  thus  break  up  the  whole  concrete 
attitude,  including  the  functional  presuppositions,  into  as  many 
parts  as  it  finds  convenient  in  the  working  out  of  the  relation  of 
mind  and  body.  The  presuppositional  method  assumes  no 
specific  end  for  psychical  process.  It  contents  itself  with  carrying 
over  from  the  sphere  of  appreciation  the  merely  functional 
concept  of  "  acquirement  of  meaning."  But  assuming  conative 
continuity  in  which  meaning  is  acquired,  it  takes  the  differences 
in  meaning  distinguished  by  appreciative  description,  which 
would  be  ignored  in  the  purely  causal  analysis,  and  asks  what 
functional  adaptation  is  presupposed  by  this  difference.  And 
since  all  adaptation  which  is  psychical  consists  of  conative  and 
cognitive  acts,  its  problem  is  to  analyse  out  the  conative  and 
cognitive  presuppositions  of  worth-feelings. 

IV.  THE  AXIOLOGICAL  PROBLEM  AND  METHOD — FACTS  AND 
NORMS — GENESIS  AND  VALIDITY 

The  second  task  of  a  theory  of  value  is  the  reflective  evaluation 
of  objects  of  value.  We  not  only  feel  the  value  of  objects,  but 
we  evaluate  these  objects  and  ultimately  the  feelings  of  value 
themselves.  Clearly  another  point  of  view  than  the  psychological 
is  here  involved,  a  point  of  view  which  requires,  not  only  to  be 
clearly  defined,  but  also  to  be  properly  related  to  the  psycho 
logical.  If  our  problem  were  that  of  the  determination  of  the 
validity  of  objects  and  processes  of  knowledge,  it  would  be  best 
described  as  logical  or  epistemological,  but  the  term  epistemology 
is  too  narrow  to  include  the  problem  of  the  evaluation  of  values, 
and  we  may  therefore  make  use  of  a  special  term  to  define  the 
problem  as  it  here  presents  itself.  On  the  analogy  of  the  term 
epistemology  we  have  constructed  the  term  axiology/  and  may 
hereafter  speak  of  the  relation  of  the  axiological  to  the  psycho 
logical  point  of  view. 

The  chief  problem  for  axiology,  as  for  epistomology,  is 
bound  up  with  the  distinction  between  subjective  and  objective, 


Introduction  1 7 

a  distinction  made  use  of  in  connection  with  judgments  of  value 
as  well  as  judgments  of  knowledge.  We  recognise  values  as  in 
some  way  independent  of  individual  acknowledgment,  for  between 
the  subject  and  the  object  there  are  relations  of  feeling  and  will, 
felt  as  demands  and  obligations,  just  as  inviolable  as  those 
of  the  sense  impressions  imposed  upon  us  from  without.  Between 
the  subjectively  desired  and  the  objectively  desirable  in  ethics, 
between  subjective  utility  and  sacrifice  and  objective  value  and 
price  in  economic  reckoning,  between  the  subjectively  effective 
and  the  objectively  beautiful  in  art,  in  all  these  cases  there  is  a 
difference  for  feeling  so  patent  that  in  naive  and  unreflective 
experience  the  feelings  with  such  objectivity  of  reference  are 
spoken  of  as  predicates  of  the  objects  themselves. 

For  reflection,  however,  there  is  a  difference  between  the 
meaning  of  this  distinction  in  the  sphere  of  values  and  that  which 
it  has  in  the  sphere  of  truth,  and  it  is  at  this  point  that  the 
specific  character  of  the  axiological  problem  appears.  In  the 
theory  of  knowledge  the  dispute  still  rages,  and  is  especially 
fierce  at  the  present  time,  as  to  whether  there  is  an  objectivity 
which  transcends  all  subjective  process,  whether  qualities  inhere 
in  the  thing  apart  from  experience.  For  the  theory  of  value 
the  problem  is  simplified.  All  values  are  in  one  sense  subjective. 
All  are  founded  on  some  process.  But  we  recognise  that  our  con 
cept  of  subjectivity  must  make  room  for  a  kind  of  objectivity, 
that  the  feelings  or  desires  developed  in  one  process  may  exercise 
a  control  over  feelings  and  desires  determined  by  other  processes, 
and  that  this  control  gives  them  a  form  of  objectivity. 

When  we  seek  a  name  for  this  form  of  objectivity,  we  find 
one  at  hand  in  the  concept  of  the  norm  and  of  normative  judg 
ments.  The  practical  significance  of  an  objective  value  is  that 
it  forms  the  norm  for  subjective  feelings  of  value,  that  it  deter 
mines  subjective  feeling  in  some  way.  An  examination  of  the 
character  of  this  determination  indicates  its  uniqueness.  The 
norm  is  the  product  of  appreciative  description  and  construction 
of  subjective  feeling  ;  but  when  it  is  thus  objectified  and  projected, 
it  becomes  a  demand,  the  acknowledgment  of  which  is  the  con 
dition,  or  presupposition,  of  further  appreciations,  or  subjective 
feelings.  The  acknowledgment  of  the  normal  exchange  value, 
the  price  of  an  object,  is  the  condition  of  its  further  utilisation 
by  the  individual,  the  acknowledgment  of  permanently  desirable 
dispositions  is  the  condition  of  the  realisation  of  certain  subjective 
ethical  values,  the  acknowledgment  of  objective  beauty  is  the 


1 8  Vahiation  :   its  Nature  and  Laws 

condition  of  permanent  aesthetic  satisfactions.  Still  more 
apparent  is  the  relation  in  the  case  of  the  extreme  objectifica- 
tions  of  religion.  Ideals  of  a  supernatural  character  are  the 
product,  phenomenologically  speaking,  of  individual  and  racial 
appreciative  constructions ;  but  the  assumption,  or  postulation, 
of  their  existence  is  the  presupposition  of  certain  subjective 
feelings  of  value,  such  as  reverence  and  inner  peace.  In  general 
the  norm  is  an  assumption  or  postulate  of  existence  representing 
the  permanent  aspects  of  desire,  underlying  changeable  feelings 
and  judgments.  Its  function  is  the  control  of  appreciation. 

The  norm  is  thus  seen  to  have  the  double  character  of  sub 
jectivity  and  objectivity.  The  normative  judgment  represents 
at  the  same  time  a  subjective  appreciation  and  an  objective 
description.  Its  subjective  reference  is  seen  in  the  fact  that 
it  is  only  through  these  projected  ideal  objects,  assumed  to  exist 
independently  of  the  subject,  that  the  subject's  individual 
feelings  can  be  communicated.  Norm  construction  is,  as  we  have 
seen,  a  product  of  appreciative  communication  and  description. 
As  such  the  norm  has  a  psychological  genesis  and  character ; 
it  is  an  assumption,  a  postulate,  determined  by  certain  disposi 
tions.  Its  objective  character  is  apparent,  on  the  other  hand, 
in  that,  having  passed  beyond  the  subjective  control  of  the 
moment  and  become,  through  its  character  as  a  presupposition 
of  belief,  the  condition  of  further  subjective  appreciations,  it  in 
turn  exercises  control  over  these  feelings. 

From  this  double  character  of  normative  objectivity  certain 
characteristics  of  axiology  and  axiological  method  may  be 
inferred.  The  axiological  problem  is  the  reflective  evaluation  of 
values,  and  this  evidently  consists  in  determining  the  validity 
of  distinctions  between  subjective  and  objective  already  de 
veloped  in  worth  experience.  Now,  the  distinction  being  what 
it  has  been  shown  to  be,  it  is  clear  that  the  question  of  the 
validity  of  any  such  distinction  is  bound  up  wholly  with  the 
question  whether  the  objectivity  postulated  fulfils  its  function 
as  the  necessary  presupposition  of  the  continuity  of  valuation,  in 
its  two  aspects  of  acquirement  and  conservation  of  value.  Other 
questions  may  indeed  be  raised — as  for  instance  whether  the 
reality  which  an  object  of  value  thus  has  is  equivalent  to  existence 
apart  from  subjective  process,  but  they  are  not  axiological.  It 
is  also  evident  that  such  a  criterion  must  remain  wholly  abstract 
and  general  until  the  phenomenology  of  valuation,  its  processes, 
its  objects  and  laws,  has  been  developed.  To  the  application 


Introduction  1 9 

of  the  criterion  we  shall  return  in  the  concluding  chapter,  con 
tenting  ourselves  for  the  present  merely  with  its  formulation. 

When,  however,  the  problem  of  axiology  is  stated  in  this  way, 
it  is  immediately  apparent  that  a  certain  definite  relation  to 
psychology  is  involved.  For  immediate  experience  this  norma 
tive  objectivity  appears  in  an  immediate  appreciation  of  value 
which  has  as  its  cognitive  presuppositions  certain  assumptions 
or  postulates,  but  for  reflection  these  very  assumptions  show 
themselves  to  be  the  product  of  a  selective,  genetic  differentiation 
of  our  desires — through  arrest,  effort,  and  consequent  readapta- 
tions  and  reconstructions,  in  which  some  of  our  desires  have 
developed  into  permanent  and  objective  demands.  Out  of  the 
general  level  of  immediate  appreciation  has  emerged  a  develop 
ment  which  has  its  conclusion  in  a  new  kind  of  objectivity  or 
reality.  It  is  clear,  then,  since  all  values,  whether  subjective  or 
objective,  are  founded  on  some  process,  that  the  ultimate  question 
as  to  their  validity  is  whether  they  are  well-founded  or  not.  It 
is  also  clear  that  whether  they  are  well-founded  or  not  depends 
upon  their  conformity  to  certain  ultimate  laws.  Every  assertion 
of  a  value  implies  at  the  same  time  an  assertion  of  its  conformity 
to  the  laws  of  feeling  and  will. 

To  this  inference  from  the  preceding  study  of  the  nature 
of  the  normative  judgment,  the  dualism  between  apprecia 
tion  and  description  is  likely  again  to  reply  with  a  doctrine  of 
absolute  values,  and  to  insist  that  the  evaluation  of  objects 
of  value  is  wholly  independent  of  their  genesis,  the  norms 
of  their  evaluation  being  in  no  sense  related  to,  or  determined  by, 
psychological  laws.  It  is  unnecessary  to  recall  our  previous  dis 
cussion  in  order  to  point  out  that  the  axiom  "  no  description 
without  appreciation  "  has  as  its  converse  "  no  appreciation 
without  description,"  a  proposition  which  we  have  also  accepted 
as  justified.  If  it  is  true  that  there  is  no  description  and  com 
munication  without  its  stimulus  and  control  in  appreciation, 
it  is  also  true  that  there  is  no  appreciation  except  through  the 
media  and  the  control  of  objective  descriptions.  But  what 
is  reflective  evaluation  but  the  highest  form  of  appreciation, 
and  how  can  that  reflective  evaluation  proceed  in  its  task  of 
distinguishing  between  subjectively  and  objectively  determined 
values  without  a  study  of  the  genesis  of  these  differentiations  ? 
The  situation  may  be  stated  in  still  another  way.  Whatever 
may  be  the  abstract  formulae  for  the  normative  sciences  of 
the  norms  of  validity,  they  cannot  be  anything  else  than  the 


2O  Valuation  :   its  Nature  and  Laws 

development  in  other  terms,  and  for  other  purposes,  of  what, 
from  another  point  of  view,  we  call  psychological  laws.  We  may 
well  believe  that  psychological  description  is  not  the  whole  of  a 
theory  of  value,  but  it  certainly  is  not  irrelevant  to  the  normative 
problem.  It  is  at  least  necessary  that  the  assumptions  and 
postulates  embodied  in  these  norms  shall  be  psychologically 
possible,  that  "  they  shall  be  in  harmony  with  the  general  laws 
of  the  conscious  life  and  only  special  and  detailed  developments 
of  what  lies  in  these  laws."  * 

1  Quoted  from  Hoffding's  discussion  of  the  relation  of  logical  laws  to  psychology, 
The  Problems  of  Philosophy,  p.  76,  translation,  Macmillan,  1905. 


CHAPTER   II 

DEFINITION  AND  ANAL  YSIS   OF  THE 
CONSCIOUSNESS   OF  VALUE 

I.  WORTH  PREDICATES  AS  FUNDED  AFFECTIVE- VOLITIONAL 
MEANINGS — ANALYSIS 

i.  The  Judgment  of  Value. 

A  CURSORY  examination  of  the  more  general  terms  of  worth 
description,  good  and  bad,  useful  and  useless,  beautiful  and 
ugly,  noble  and  ignoble,  etc.,  or  indeed  the  terms  worth  and 
worthless,  valuable  and  valueless  themselves,  and  the  manner 
in  which  they  are  applied,  makes  us  immediately  aware  of  the 
fact  that,  for  the  unreflective  worth  consciousness,  they  are  at 
first  tertiary  qualities,  as  much  a  part  of  the  object  as  the  so- 
called  primary  and  secondary  qualities  are  parts  of  the  physical 
object  of  cognition.  This  is  especially  noticeable  in  the  case  of 
the  ethical  and  aesthetic  predicates,  but  it  is  no  less  true  of  the 
unreflective  use  of  the  terms  utility  and  value,  as  for  instance 
when  we  say  that  iron  has  utility  or  value  even  when  the  con 
ditions  of  its  applicability  are  lacking.  The  intrinsic  worth 
judgment  is  psychologically  the  more  fundamental,  whatever 
may  be  inferred  upon  closer  inspection  and  reflection. 

But  while  as  qualities  of  objects,  as  terms  employed  in 
appreciative  description,  they  have  a  certain  kind  of  objectivity, 
they  are  nevertheless  felt  to  differ  from  the  other  qualities  in 
that  they  are  subjectively  conditioned  in  a  way  that  the  so- 
called  primary  and  secondary  qualities  are  not. 

The  judgment  of  value  has  accordingly  been  described  as  a 
mere  assertion  of  the  meaning  of  the  object  for  the  subject,  or  as 
an  appreciation.  When  I  say  that  the  object  is  good  or  beautiful 
or  noble,  I  assert  a  direct  relation  of  the  object  to  my  feeling  and 
will,  a  vharmony  between  the  object  and  my  subjective  dis 
positions  which  is  relatively  independent  of  my  judgment  of 

21 


22  Valuation  :   its  Nature  and  Laws 

existence  of  the  object  or  judgment  of  the  truth  of  the  idea  I  have 
of  the  object.  Existence  is  perceived  ;  truth  is  thought ;  value 
is  felt.  But  while  the  worth  predicates  are  in  the  first  place 
felt  and  not  cognised,  while  they  are  at  the  third  remove  from 
pure  objectivity,  nevertheless,  there  is  presupposed  in  every 
appreciation,  in  every  judgment  of  value,  a  reference  to  reality 
and  truth.  This  reference  comes  to  the  surface  as  soon  as  I  ask 
such  questions  as  these  :  is  the  object  really  good  or  useful  ? 
is  it  truly  noble  or  beautiful  ?  The  feeling  of  value  includes  the 
feeling  of  reality.  Appreciative  meanings  presuppose  reality 
meanings. 

2.  Equivocations  in  the  Value  Judgment,  leading  to  Axiological 
Distinctions  in  Existence,  and  Reality  Meanings. 

Accordingly,  when  we  attempt  a  further  analysis  of  these 
predicates  we  find  certain  references  to  reality,  always  implicit 
in  the  judgment,  which  demand  to  be  made  explicit.  Prior 
to  such  reflective  analysis  they  give  rise  to  equivocations  in  their 
meaning,  equivocations  so  confusing  that  more  than  one  thinker 
has  counselled  entire  scepticism  in  the  matter,  not  without  a  show 
of  reason,  it  must  be  admitted.  But  that  this  initial  scepticism 
is  merely  a  salutary  warning  becomes  apparent  as  we  follow 
these  equivocations  to  their  sources,  for  it  is  precisely  in  this 
process,  this  study  of  the  grammar  of  the  worth  consciousness, 
that  we  find  both  the  nature  of  these  predicates  and  the  basis 
of  their  classification.  The  character  of  the  confusion  may  be 
seen  at  a  glance  by  observing  the  distinctions  which  worth  analysis 
has  developed,  in  all  the  concrete  worth  sciences,  economics, 
ethics,  aesthetics,  for  the  removal  of  the  equivocations.  Worths 
are  said  to  be  subjective  or  objective,  real  or  ideal,  actual  or 
imputed,  intrinsic  or  instrumental. 

The  first  distinction,  between  subjective  and  objective 
worths  or  values,  gives  the  key  to  the  situation.  The  same 
objects,  let  us  say  diamonds,  may  have  little  worth  or  indeed  be 
distasteful  to  me  personally,  although  in  another  attitude  I 
may  ascribe  great  value  to  them  and,  indeed,  think  of  them  as 
intrinsically  valuable.  My  friend's  action  may  be  sanctioned 
by  me  in  immediate  appreciation,  although  from  an  objective, 
moral  point  of  view  I  must  needs  condemn  it.  Such  contradic 
tions  can  only  be  resolved  by  a  distinction  between  subjective 
and  objective  values.  Closely  connected  with  this  equivocation 


Definition  of  the  Consciousness  of  Value  23 

is  that  which  arises  when  the  distinction  between  intrinsic  and 
instrumental  values  is  ignored.  An  object  which  is  worthless, 
or  indeed  the  object  of  negative  worth  judgments  of  harmful 
or  bad,  may  acquire  the  worth  predicate  when  it  becomes  in 
strumental  to  some  object  of  immediate  or  intrinsic  value. 
Similarly,  within  the  sphere  of  instrumental  values  or  utilities  of 
economics,  we  find  an  equivocation  which  can  be  removed  only  by 
the  use  of  the  distinction  between  subjective  and  objective. 
On  the  one  hand,  if  any  thing  is  of  worth  because  it  is  utilisable, 
it  is  always  so  for  a  subject  and  with  reference  to  concrete  con 
ditions.  But  on  the  other  hand,  we  are  led  to  ascribe  value  to  an 
object,  for  instance  when  we  say  that  iron  has  value,  irrespective 
of  its  relation  to  an  individual  subj  ect  and  to  concrete  conditions  : 
by  a  process  of  abstraction  we  give  the  object  value  in  itself. 
For  these  differences  in  meaning  the  economists  have  used  the 
terms  subjective  and  objective  value ;  or  the  latter  is  sometimes 
called  objective  exchange  value. 

From  these  illustrations  we  see  that  the  attitude  expressed 
by  a  worth  judgment,  whether  the  worth  be  described  as  sub 
jective  or  objective,  is  an  attitude  of  a  subject,  but  the  difference 
in  attitude  is  determined  by  the  inclusion  or  exclusion  of  certain 
presuppositions,  the  nature  of  which  is  to  be  determined.  The 
other  distinctions,  between  real  and  ideal,  actual  and  imputed 
values,  show  the  same  desire  to  remove  the  equivocations  inherent 
in  worth  predicates. 

Sometimes  we  attribute  worth  to  an  object  when  we  mean 
that  it  deserves  to  be  valued,  irrespective  of  its  actual  valua 
tion  by  any  person  or  group  of  persons.  Such  value  is  said 
to  be  ideal.  Again,  there  are  objects  of  valuation,  the  exist 
ence  or  non-existence,  or  the  possibility  or  probability  of 
realising  which  is  not  inquired  into,  but  which  are  abstractly 
valued  and  said  to  be  ideal  values  in  contrast  to  the  real 
value  of  objects  where  the  judgments  of  existence  or  possibility 
are  true  or  grounded  judgments.  In  both  cases  the  real  and  the 
ideal  values  are  equally  functions  of  the  relation  of  the  object 
to  the  subject.  The  difference  lies  in  the  attitude  of  the  subject, 
in  the  different  presuppositions  of  the  feeling  in  the  two  cases. 
Confusion  of  meaning  arises  only  when  these  presuppositions 
are  not  made  explicit. 

The  distinction  between  actual  and  imputed  values,  like  the 
other  distinctions  considered,  is  one  which  is  not  found  in  the 
immediate  worth  experience  itself,  but  which  develops  when 


24  Valuation :   its  Nature  and  Laws 

the  presuppositions  of  the  worth  judgment  are  made  explicit 
through  reflective  analysis.     The  total  worth  predicated  of  an 
object  is  often  seen  to  have  more  than  one  determinant  and 
under  certain  circumstances,   the   element  in   the   total  value 
corresponding  to  one  subjective  determinant  will  be  described  as 
actual,  while  the  other  element  will  be  described  as  imputed. 
Thus  the  elements  of  a  total  complex,  food  for  instance,  will  each 
be  said  to  have  its  actual  value  arising  from  its  capacity  to 
satisfy  separate   desires,   or  to  satisfy  desire  when  consumed 
separately.     On  the   other    hand,   such   worth   as   an   element 
may  get  from  its  combination  with  the  other  elements  is  said 
to  be  an  imputed  value. ,  In  a  similar  way,  when  an  act  of  a 
person  has  value  as  manifesting  a  disposition  instrumental  to  the 
fulfilment  of  social  ends,  this  is  described  as  its  actual  value, 
while  an  additional  value  attributed  to  it  as  a  part,  or  manifesta 
tion  of  the  total  personality,  is  described  as  an  imputed  value 
over  and  above  the  actual  value  of  the  act.     It  is  obvious  from 
these  illustrations  that  the  different  moments  in  the  total  worth  of 
the  object  have  different  subjective  determinants,  and  that  these 
go  back  to  the  different  objects  or  aspects  of  the  object  upon 
which  judgment  is  directed,  in  other  words,  to  the  cognitive 
presuppositions. 

3.   Interpretation  of  these  Equivocations — As  due  to 
Different  Presuppositions. 

The  meanings  thus  differentiated  may  be  described  as  the 
reality-meanings    of    the    worth    predicates.      As    distinguished 
from  the  purely  appreciative  meanings  previously  considered, 
they  represent  modifications  in  worth  predication  determined 
by   differences  in  cognitive  attitude  toward  the  object.     The 
necessity  of  such  distinctions  arises  from  the  fact  that  the  appre 
ciative  meanings  are  not  wholly  independent  of  the  reference  to 
reality    involved.     As    simple    acts    of    appreciation,    the    pre 
suppositions    of    existence   may   not   be   explicit;    indeed   the 
most  primitive  judgments  of  worth  are  assertorial— without  any 
conditional  element  whatever.     But  as  soon  as  the  question  of 
evaluation  of  the  worth  predicates  themselves  is  considered,  as 
soon  as  the  axiological  problem  of  the  differentiation  of  subjec 
tively  conditioned  values  from  objectively  conditioned,  is  raised, 
then  the  presuppositions  of  reality  must  be  made  explicit.     By 
making  them  explicit  is  understood  the  acknowledgment  of  the 


Definition  of  the  Consciousness  of  Value  25 

presupposition  of  reality,  present  in  all  judgments  of  value,  in  all 
appreciations,  in  specific  judgments  of  existence.  In  what  way 
are  the  values  real  ?  In  what  way  are  the  objects  of  value 
existents  or  realities  ? 

From  this  study  of  the  various  meanings  of  the  worth  predi 
cates,  it  becomes  clear  that  the  worth  judgments  express  not 
attributes  of  objects  apart  from  the  subject,  even  when  the  value 
is  described  as  actual  and  objective,  but  rather  functions  of  the 
relation  of  subject  to  object.  When  we  speak  of  an  object 
as  having  absolute  or  objective  value,  it  is  only  by  a  process 
of  temporary  abstraction  from  the  subject  in  some  specific 
attitude,  not  from  the  subject  itself.  The  other  differences  of 
meaning  in  the  worth  predicates  reflect  the  same  fact.  Thus 
when  I  attribute  value  to  an  object,  meaning  that  it  is  actually 
valued,  my  attitude  is  determined  by  certain  presuppositions 
of  judgment,  which  are  the  product  of  participation  in  the  worth 
judgments  of  others.  When,  however,  my  judgment  means  that 
the  object  is  ideally  of  worth,  deserves  to  be  valued,  that 
judgment  expresses  a  modification  of  attitude  brought  about 
either  by  exclusion  of  certain  presuppositions  of  judgment, 
as  when  I  pass  my  judgment  in  opposition  to  temporary 
judgments  about  me,  or  by  inclusion  of  other  presuppositions, 
as  when,  for  instance,  I  appeal  from  a  narrower  actual  worth 
judgment  to  a  possible  more  universal  judgment.  The  situation 
is  the  same  in  the  case  of  the  distinction  between  actual  and 
imputed  values.  The  actual  value  is  always  the  meaning  of 
the  object  for  a  subject  in  some  attitude — never  an  attribute  of 
the  object  itself.  The  imputed  value  added  to  the  actual  value 
arises  from  attitudes  of  the  subject,  negligible  or  irrelevant  from 
the  standpoint  from  which  the  actual  value  is  determined. 

From  all  this  it  is  apparent  that  whatever  meaning  we  may 
ultimately  give  to  the  objectivity  of  worth  predicates,  whatever 
validity  may  be  assigned  to  the  presupposition  of  reality  im 
plicit  in  all  judgments  of  value,  we  may  unhesitatingly  assert 
that  these  predicates  are  meanings  pre-determined  by  antecedent 
psychical  processes.  While  at  first  sight  they  appear  to  be 
tertiary  qualities  of  the  object,  on  closer  inspection  they  are 
seen  to  be  acquired  meanings  of  the  object  for  the  subject,  as,  in 
fact  are  some  of  the  so-called  primary  and  secondary.  As  thus 
predetermined  they  may  be  described  as  funded  meanings,  in 
that  they  represent  the  accumulated  meaning  of  these  processes. 
Furthermore,  we  may  now  see  that  the  equivocations  in  the 


26  Valuation  :    its  Nature  and  Laws 

value  judgment  which  have  been  analysed  arise,  just  as  do  certain 
contradictions  in  cognitive  predication,  through  abstraction  of 
the  predicates,  as  qualities  of  the  objects,  from  the  processes 
of  acquirement  of  meaning  in  which  the  meaning  was  funded. 

But  this  analysis  enables  us  to  add  something  more  to  our 
definition  and  characterisation  of  worth  predicates.  We  have 
defined  them  as  funded  meanings,  pre-determined  by  antecedent 
psychical  process.  It  is  possible  to  limit  stiU  further  the  concept 
by  defining  them  as  affective-volitional  meanings,  thus  distin 
guishing  them  relatively  from  the  attributes  employed  in  cognitive 
predication.  Relations  to  judgments  of  truth  and  existence  are 
presupposed  in  all  appreciations  and  judgments  of  value,  but 
as  we  have  seen,  in  the  first  place  only  implicitly.  As  assertorial 
judgments,  they  assert  a  relation  of  the  object  to  feeling  and  will 
—either  an  immediate  actual  experience  or  a  possible  experience 
of  feeling  or  will— that  is,  belief  in  the  power  of  the  object  to  call 
such  experiences  into  being. 

II.  FURTHER  ANALYSIS  OF  THE  VALUE  JUDGMENT  AS  EXPRESSING 
AFFECTIVE-VOLITIONAL  MEANING  OF  THE  OBJECT  FOR  THE 
SUBJECT 

i.  Axiological  Distinctions  as  Clues  to  Analysis. 

Two  important  consequences  follow  from  this  conception 
that  worth  or  value  is  the  funded  meaning  of  the  object  for  the 
subject  in  different  attitudes,  or  as  predetermined  by  different 
dispositions  and  interests.  In  the  first  place,  while  the  dis 
tinctions  we  have  been  discussing  are  developed  from  the  axio- 
logical  standpoint  of  the  determination  of  the  relative  validity 
of  worth  judgments,  we  have  in  the  analysis  underlying  these 
distinctions  at  the  same  time  a  clue  to  the  psychological  analysis 
and  classification  of  the 'different  attitudes  involved.  In  all  these 
differences  of  meaning  the  sources  of  the  difference  were  found  in 
the  nature  of  the  cognitive  presuppositions.  All  valuation,  as 
attitude  of  the  subject,  is  primarily  an  act  of  immediate  apprecia 
tion  ;  but  this  primitive  attitude  may  be  modified  to  give 
various  meanings  by  the  inclusion  of  various  types  of  judgments, 
existential,  instrumental,  judgments  referring  the  object  to  the 
self  or  others,  judgments  of  possibility  or  probability  of  acquisi 
tion  and  possession,  etc.  While  for  the  axiological  point  of  view 
the  truth  of  these  presuppositions  is  significant,  for  psychological 


Definition  of  the  Consciousness  of  Value  27 

analysis  their  importance  lies  in  the  changes  in  worth  experience, 
which  follow  upon  changes  in  these  presuppositions.  In  the 
second  place,  as  a  result  of  this  conception  of  worth  as  the 
affective- volitional  meaning  of  the  object  for  the  subject  in 
different  attitudes,  the  way  is  now  open  for  an  analysis  of  the 
worth  subject  and  for  a  classification  of  the  fundamental  worth 
attitudes. 

2.  The  "  Subject  "  of  the  Value  Judgment— The  Subject  in  Different 
Attitudes — Classification  of  Attitudes. 

The  equivocations  in  the  meaning  of  the  worth  predicates 
already    considered,    indicate    certain    fundamental    differences 
for  the  subject  of   the   experience.     The   distinctions   between 
subjective  and  objective  worth,  between  actual  and  ideal,  are 
reducible  to  differences  in  the  judging  subject.     These  differ 
ences  have  led  to  the  conception  of  different  subjects  for  differ 
ent   types   of  worth   judgments.     Thus  Kreibig1   distinguishes 
between  a  primary  and  secondary  worth  subject,  the  primary 
being  the  individual  as  such,  the  secondary  being  the  group  or 
race  consciousness.    So  also  Meinong,2  in  treating  of  the  difference 
between  ethical  and  moral  judgments,  distinguishes  the  more  per 
sonal  ' '  ethical ' '  from  the  impersonal, ' '  moral '  'subj  ect.  The  former 
is  the  concrete  ego  in  his  relation  to  the  alter  ;    the  latter  is 
neither  the  ego  nor  the  alter  but  an  abstraction,  a  third  person,  the 
"  impartial  spectator  "  who  sits  in  judgment  upon  both.     These 
distinctions,  appearing  as  they  have  in  the  effort  to  do  justice 
to  fundamental  differences  in  worth  predication,  point  in  the  right 
direction.     But  they  are  nevertheless  open  to  the  criticism  which 
attaches  to  all  conceptual  constructions  employed  as  instruments 
of  analysis,  that  they  are  in  danger  of  being  hypostatised  into 
separate  realities  and  conceived  as  real  even  when  abstracted 
from  the  individual  subject.     For  certain  purposes  of  social  and 
ethical  philosophy,  we  may,  perhaps,  speak  of  a  group  conscious 
ness,  of  a  general  or  over-individual  will,  without  a  serious  dis 
tortion  of   the  facts  ;   but  for  the  empirical  analysis  of  worth 
judgments,  it  is  nearer  the  truth  to  say  that  the  subject  in  the  role 
of  the  individual,  of  the  group  or  race,  or  of  the  impartial  spectator, 
is  the  individual  in  different  attitudes.     The  problem  is  then  to 

1  Kreibig,  Psychologische  Grundlegung  eines  Systems  der  Wert-theoric,  Wien,  1902, 


P 


Meinong,  Psychologisch-ethische    Untersuchungen  zur   Wert-lheoriet  Graz,   1894, 
pp.  72,  163,  216. 


28  Valuation:   its  Nature  and  Laws 

account  for  the  origin,   differentiation,   and  fixation  of  these 
relatively  permanent  attitudes. 

The  worth  judgment  of  an  individual  may  then  express  the 
affective- volitional  meaning  of  an  object  for  the  subject,  as 
qualified  by  the  subject's  participation  in  and  explicit  cog 
nition  of  the  worth  attitudes  of  others,  of  single  persons,  of 
social  groups,  or  perhaps  of  an  over-individual  worth  conscious 
ness  which  transcends  even  group  distinctions,  giving  the  im 
personal  attitude  of  the  "  impartial  spectator."  The  differ 
ence  in  attitude  is  determined  by  the  inclusion  or  exclusion  of 
judgments  as  part  presuppositions  of  the  meaning.  The  psycho 
logical  problem  is  the  tracing  of  the  processes  by  which  this 
participation  in,  and  cognition  of,  the  attitudes  of  others  is  real 
ised,  the  more  specific  problem  of  worth  analysis  itself  being  to 
determine  how  this  modification  of  the  attitude  of  the  subject 
also  modifies  the  worth  predicated  of  the  object. 

In  a  preliminary  way  we  may  distinguish  three  fundamental 
attitudes  of  the  self  or  subject  of  worth  judgment :  (i)  Simple 
appreciation  of  the  affective-volitional  meaning  of  an  object  for 
the  self ;  (2)  the  personal  attitude,  in  which  the  worth  of  the 
object  is  determined  by  explicit  reference  of  the  object,  whether 
a  physical  possession  or  a  psychical  disposition,  to  the  self  or 
the  alter,  and  in  which  characterisation  of  the  self  or  the  alter 
is  presupposed;  and  (3)  the  impersonal  attitude,  in  which  the 
subject  of  the  judgment  is  identified  with  an  impersonal  over- 
individual  subject  and  the  value  of  the  object  is  determined  by 
explicit  reference  to  the  over-individual  demand.1  All  three  are 
forms  of  appreciation  of  worth,  but  while  the  first  is  simple 
appreciation,  in  that  the  presuppositions  are  simple,  the  personal 
and  impersonal  attitudes  are  complex  and  derived,  having  as 
their  presuppositions  judgments  and  assumptions  which  have 
had  an  historical  genesis. 

3.  The  "  Object "  of  the  Value  Judgment — Classification  of  Worth 

Objects. 

As  the  subject  of  value  experience,  one  of  the  moments  in 
the  value  function,  is  constantly  changing,  expanding  and  con 
tracting  through  inclusion  and  exclusion  of  presuppositions  of 

1  This  classification  corresponds  in  principle  with  Baldwin's  classification  of  cogni 
tive  meanings  in  the  first  volume  of  his  J^hought  and  Things,  chap,  vii,  p.  148,  where  he 
distinguishes:  (i)  Simple  and  private;  (2)  aggregate  and  con-aggregate;  (3)  social  and 
public  meanings. 


Definition  of  the  Consciousness  of  Value  29 

judgment,  so  also  the  object  of  valuation  undergoes  modifica 
tion.  Broadly  speaking,  the  object  of  worth  belongs  to  the 
presentational  side  of  consciousness,  is  the  object  of  immediate 
apprehension  with  its  implicit  presupposition  or  explicit  judgment 
of  existence.  It  is,  therefore,  in  the  first  place,  the  not-self  the 
external  object  of  feeling  and  will,  those  aspects  of  experience 
which  are  from  the  beginning  presentational.  But  there  is 
scarcely  any  aspect  of  consciousness  which  cannot  become  pre 
sentational,  cannot  be  presented  to  consciousness,  and  thus 
become  the  object  of  judgment.  Even  the  attitude  of  valuation 
itself,  which  we  may  describe  as  "psychical"  pre-eminently,  is 
susceptible  of  representation,  translation  into  ideal  terms,  and 
of  thus  taking  its  place  on  the  objective  side  of  the  value  function. 
The  psychology  of  this  representation  of  the  psychical  will 
engages  our  attention  at  those  points  where  we  shall  make  use 
of  the  principle.  Here  it  is  merely  important  to  insist  that  the 
general  class,  worth  objects,  includes  physical  and  psychical 
and,  among  the  latter,  the  attitude  of  valuation  itself. 

A  more  significant  distinction  among  objects  of  valuation  is 
that  between  primary  and  secondary  or  between  simple  and 
founded  objects.  By  a  founded  object  in  general  we  understand 
one  built  up  by  processes  of  ideation  and  judgment  upon 
primary  sensations  and  perceptions.  Such  a  founded  object 
is,  strictly  speaking,  not  the  object  of  perception  but  of  idea 
tion  or  judgment,  and  may  be  said  to  be  pre-determined  by  these 
processes.  Thus  certain  ideal  objects  of  presentation  and  judg 
ment,  while  themselves  neither  sensed  nor  perceived,  may  be  said 
to  be  founded  on  sensation  and  perception.  /  The  processes  of 
sympathetic  realisation  of  the  feelings  of  another,  are  in  the  first 
place  perceptual  in  character,  but  upon  the  basis  of  these  pro 
cesses  certain  ideal  objects,  the  self  and  its  dispositions,  are  built 
up,  and  these  become  the  objects  of  imputed  values.  To  them  is 
imputed  the  funded  meaning  of  the  processes  of  feeling  and 
conation  involved  in  their  construction. 

These  founded  objects  may  be  of  two  kinds,  according  as  they 
are  founded  on  processes  of  perceptual  or  ideational  activity.  Illus 
trations  of  the  former  are  :  (a)  beauty  or  grace  of  form  in  objects 
of  perception  ;  (6)  founded  qualities  acquired  in  the  sensational  and 
perceptual  activities  of  consumption  of  food,  or  more  broadly  of 
various  instinctive  activities,  such  as  cleanliness,  manners.  Any 
harmonious  grouping  or  arrangement  of  the  activities  of  living 
creates  secondary  objects  of  worth,  founded  upon  the  primary. 


3O  Valuation  :   its  Nature  and  Laws 

As  illustrations  of  the  secondary  worth  objects,  founded  on  pro 
cesses  of  ideation  and  judgment,  we  may  take  the  person  and  his 
affective  or  conative  dispositions,  built  up  conceptually  on  the  basis 
of  the  immediate  appreciations  of  sympathy  by  a  process  of  infer 
ence,  which,  in  turn,  become  the  objects  of  secondary  judgments 
of  merit  and  demerit,  etc.  To  these  may  be  added  a  third  group 
of  founded  worth  objects  which  may  be  described  as  over- 
individual.  These  are  the  products  of  the  ideal  re-construction  of 
objects  of  primary  worth,  as  determined  by  participation  in  the 
worth  processes  of  larger  social  groups  or  of  society  at  large. 
To  this  class  belong  the  ideal  moral  and  cultural  goods  of  society, 
economic  goods  as  objects  of  exchange,  including  the  medium 
of  exchange  which  has  over-individual  worth  exclusively.  In 
distinguishing  thus  between  founded  objects  as  products  of 
perceptual  and  ideational  activities,  we  cannot  of  course  make  the 
distinction  absolute,  for  in  the  case  of  many  such  objects  both 
activities  have  been  at  work  in  their  construction. 

A  preliminary  classification  of  worth  objects  would  then 
include  the  following  groups  :  (i)  Objects  of  simple  appreciation. 
These  objects  may  be  either  physical  or  psychical  and  include 
the  founded  psychical  objects  built  up  in  perceptual  activity. 
The  worth  of  these  objects  may  also  be  described  as  "  condition  " 
worth  for  the  reason  that  when  the  feeling  of  value  is  made  the 
object  of  reflection  it  is  referred  directly,  as  feeling  of  pleasantness 
or  unpleasantness,  to  a  modification  of  the  condition  of  the 
organism,  and  is  set  in  contrast  to  personal  and  social  values. 
(2)  Objects  of  personal  worth  such  as  qualities  and  dispositions 
of  the  person  (the  self  or  the  alter)  objects  founded  in  the  pro 
cesses  of  characterisation  of  the  person.  (3)  Objects  of  over- 
individual  or  common  worth  constructed  in  processes  of  social 
participation,  ideal  constructions  developed  in  the  interest  of 
social  participation,  or  of  utilisation  and  exchange  of  objects.  In 
general  these  objects  of  worth  correspond  to  the  fundamental 
attitudes  of  the  subject  of  the  value  experience. 

4.  The  Relation  of  Subject  and  Object — Further  Development  of 
the  Term  "Affective-Volitional  Meaning  ":  its  Extension  and 
Intension. 

The  analysis  of  the  meanings  of  worth  predicates,  and  the 
consequent  differentiation  and  classification  of  the  fundamental 
types  of  the  subject  and  object  of  the  judgment  of  value,  bring 


Definition  of  the  Consciousness  of  Value  3 1 

us  to  a  third  problem,  namely,  a  more  definite  characterisa 
tion  of  the  term  affective-volitional  meaning  and  an  analysis 
and  classification  of  the  modes  of  consciousness  corresponding  to 
its  various  meanings.  As  long  as  we  were  concerned  merely 
with  a  preliminary  differentiation  of  cognitive  meaning  from 
that  aspect  of  meaning  described  as  worth  or  value,  it  was 
sufficient  to  describe  the  latter  as  a  meaning  pre-determined  by 
processes  of  feeling  and  conation,  and  the  judgment  of  value  as 
an  appreciation  or  acknowledgment  of  that  funded  meaning. 
But  when  this  criterion  is  examined  more  closely  and  the  attempt 
is  made  to  determine  more  precisely  just  what  aspect  of  meaning 
is  represented  by  the  different  types  of  worth  judgments,  appre 
ciation,  characterisation,  participation  and  utilisation,  just  what 
the  determining  processes  of  feeling  and  conation  are  in  each 
case,  more  detailed  psychological  analysis  becomes  necessary. 

When  we  seek  to  make  more  specific  this  very  general  descrip 
tion  of  the  worth  relation,  we  are  confronted  with  two  possible 
views  of  worth  which  may  be  described  as  a  broader  and  a 
narrower  view. 

The  narrower  view  recognises  only  two  types  of  value 
judgment,  the  ethical  and  economic,  thereby  limiting  the  term 
value  to  such  feeling  attitudes  as  follow  upon  the  affirmation 
of  the  existence  or  non-existence  of  an  object  for  the  self  or 
its  purposes.  This  limitation  denies,  therefore,  the  character 
of  worth  attitude  to  all  immediate  feeling  of  the  meaning 
of  the  object  for  the  subject,  prior  to  the  distinctions  which 
we  describe  as  economic  and  ethical,  and  likewise  to  all  forms 
of  higher  immediacy  of  feeling  attitude  as  we  have  them 
primarily  in  the  aesthetic  consciousness.  This  view,  which 
has  been  presented  most  definitely  by  Witasek1  and  H.  W. 
Stuart,2  logically  excludes  the  aesthetic  from  the  sphere  of 
values,  in  the  view  of  the  former  because  the  aesthetic  is  pre- 
judgmental,  i.e.,  is  feeling  which  has  merely  presentations  as  its 
content,  in  the  view  of  the  latter  because  he  conceives  it  to 
be  post-judgmental,  an  appreciative  state  in  which  all  judgment 
has  lapsed.  Either  mode  of  cutting  the  aesthetic  attitude  off 
from  its  closely  related  ethical  and  economic  attitudes  is,  we 
shall  find,  open  to  serious  criticism. 

The  reasoning  which  underlies  the  formulation  of  this  criterion 

1  Witasek,  Allgemeine  JEsthetik,  Leipzig,  1904. 

a  Stuart,    Valuation  as  a  Logical  Process,  in  Dewey's  Studies  in  Logical  Theory^ 
Chicago,  1903. 


32  Valuation:    its  Nature  and  Laws 

is  well  expressed  by  Stuart  in  the  following  paragraph  :  "  Our 
general  criterion  for  the  propriety  of  terming  any  mode  of 
consciousness  the  value  of  an  object  must  be  that  it  shall  per 
form  a  logical  function,  and  not  simply  be  referred  to  in  its 
aspect  of  psychical  fact.  The  feeling  or  emotion,  or  whatever 
the  mode  of  consciousness  in  question  may  be,  must  play  the 
recognised  part,  in  the  agent's  survey  of  the  situation,  of  prompt 
ing  and  supporting  a  definite  practical  attitude  with  reference 
to  the  object.  If,  in  short,  the  experience  enters  in  any  way  into 
a  conscious  purpose  of  the  agent,  it  may  properly  be  termed 
a  value." 

Now,  in  examining  this  criterion  one  recognises  immediately 
that  it  provides   a  good   definition   of  a  certain   type   of   re 
flective  value  judgments  which  we  may  call  secondary.     A  very 
large   group  of   our  worth  judgments  are   determined   by  the 
conscious   inclusion    of    the    feeling    or    emotion    as    presented 
content,  and  as  partial  determinant  of  the  judgment.   The  typical 
economic  judgment  takes  place  only  upon  the  occasion  of  adding 
to  or  taking  from  our  store  of  objects,  and  is  motived  by  a  re 
flective  inclusion  of  the  worth  feeling  in  our  total  practical  atti 
tude.    The  ethical  judgment,  in  its  typical  reflective  form,  may  be 
shown  to  be  of  the  same  character  in  that  the  subject's  own  mode 
of  experience  or  way  of  feeling,  presented  in  terms  of  a  disposition 
or  quality  of  the  self,  enters  as  a  determinant  in  the  total  situation. 
But   the  secondary   and  derived  character  of  these  reflective 
judgments   soon    becomes    evident.     How   can   the    feeling    or 
emotion  as  presented  content  "  play  a  recognised  part  "  as  a  value 
"  in  the  agent's  survey  of  the  situation,"  unless,  as  a  motive 
to  previous  unreflective  judgments,  i.e.,  before  it  was  presented 
as  a  conscious  determinant,  it  was  also  a  value  or  at  least  sug 
gestive  of  value  ?     We  may  say,  then,  that,  while  much  of  valua 
tion  is  a  logical  process  in  this  sense,  nevertheless  valuation  in 
itself  has  its  roots  in  experiences  of  simple  appreciation,  where  the 
emotion,  while  determinative,  is  not  so  consciously,  as  object 
of  presentation  or  judgment,  and  must,  therefore,  be  referred  to 
simply  in  its  aspect  of  psychical  fact. 

We  must,  accordingly,  interpret  our  definition  of  value  as 
affective-volitional  meaning  in  a  broader  way — so  as  to  include 
modes  of  feeling  or  desire,  as  the  case  may  be,  which 
are  merely  appreciative  of  the  object,  which  merely  appre 
hend  the  object  with  its  funded  meaning.  We  cannot  con 
fine  it  to  attitudes  in  which  this  meaning,  abstracted  from 


Definition  of  the  Consciousness  of  Value  33 

the  object,  becomes  a  motive  in  the  subject's  survey  of  the 
situation.  We  shall  then  be  enabled  to  include  both  the  attitudes 
of  lower  immediacy,  which  are  pre-judgmental,  and  those  of 
higher  immediacy,  which  are  post-judgmental,  recognising  the 
intermediate  role  of  the  reflective  judgments,  existential,  instru 
mental,  possessive,  etc.,  and  recognising  also  that  the  reflective 
and  the  unreflective,  the  intrinsic  and  the  instrumental,  are  con 
stantly  passing  over  into  each  other,  a  phenomenon  which  we 
shall  later  describe  as  value-movement. 

In  close  relation  to  this  first  problem  which  arises  in  the 
attempt  to  make  more  specific  the  general  definition  of  worth  as 
affective-volitional  meaning,  a  second  problem  arises,  namely, 
the  question  of  the  specific  manner  in  which  we  shall  set  the 
worth-moment  in  relation  to  its  psychological  equivalents,  feeling 
and  conation.  In  the  use  of  the  double  term  affective-volitional 
itself  in  our  preliminary  demarcation  of  worth  experience,  there 
inheres  a  certain  vagueness  which,  while  excusable  when  viewed 
in  the  light  of  the  purpose  of  the  term,  must  be  subjected  to 
explicit  analysis  if  we  are  to  find  equivalents  for  worth 
experience  which  shall  form  the  basis  for  a  scientific  re 
construction  of  the  processes  of  valuation.  The  significance 
of  this  double  term  lay  in  the  fact  that  it  marked  off  a  species 
under  the  generic  term,  meaning.  Not  that  there  could  be  cog 
nitive  meaning  without  worth  references  or  affective-volitional 
meaning  without  cognitive  presuppositions.  Indeed,  we  shall 
see  that  these  terms  are  not  very  clear  at  the  limits.  Merely  to 
indicate  a  relative  distinction,  by  means  of  emphasis  on  different 
aspects  of  meaning,  was  the  purpose  of  this  differentiation. 

In  the  second  place,  the  double  term  was  necessary  for  the 
reason  that  only  in  such  a  definition  could  all  the  attitudes 
toward  objects,  recognised  as  worth  attitudes,  be  included. 
For  our  ordinary  usage,  at  least,  makes  a  clear  distinction  be 
tween  feeling  and  will  and  recognises,  as  objects  of  worth,  objects 
toward  which  both  types  of  attitude  are  directed.  Prior  to 
more  scientific  analysis,  this  double  relation  must  be  taken  as 
descriptive  of  the  worth  attitude.  But  here  again,  when  this 
general  definition  is  subjected  to  psychological  analysis,  we  find 
that  the  distinction  between  feeling  and  conation  in  some  of  its 
forms  is  far  from  clear,  and  it  is  consequently  difficult  to  say 
under  which  of  these  terms  the  immediate  experience  which  is 
the  bearer  of  these  meanings,  is  to  be  subsumed.  On  the  one 
hand,  we  find  experiences  of  preference  and  obligation  where 


34  Valuation  :    its  Nature  and  Laws 

feeling,  in  the  form  of  passive  pleasantness  and  unpleasant 
ness,  is  scarcely  present,  or,  if  present  at  all,  is  irrelevant — 
so  irrelevant  in  fact  that  some  theories  of  worth  experience  (e.g. , 
the  voluntaristic  theories  of  Brentano  and  Schwartz)  find 
the  locus  of  value  in  what  they  describe  as  "  intensity-less 
acts  of  preference,"  denying  the  worth  aspect  to  feeling  and 
its  intensities.  On  the  other  hand,  we  find  worth  experiences, 
such  as  the  aesthetic,  apparently  purely  affective,  where  desire, 
or  conation  in  all  its  forms,  is  at  a  minimum,  and  appears  to  be 
significant,  if  significant  at  all,  merely  as  a  disposition  or  pre 
supposition.  While,  then,  in  view  of  these  facts  the  general 
term  affective-volitional  meaning  was  necessary  to  define  the 
various  meanings  of  objects  included  under  the  term  values,  it  is 
nevertheless  evident  that  the  definition  can  become  serviceable 
for  further  psychological  analysis  and  explanation  only  when 
it  is  determined  which  of  these  elements,  the  affective  or  conative, 
is  primary  and  which  secondary — that  is,  which  is  always  present 
actually  as  conscious  experience  and  which  as  a  merely  dis- 
positional  determinant. 

In  the  light  then  of  these  considerations,  it  would  appear 
that  the  course  of  our  further  analysis  is  clearly  indicated. 
We  are  compelled,  on  the  one  hand,  to  include  both  the 
concepts  of  feeling  and  conation  in  our  psychological  equiv 
alents  for  the  worth  moment ;  otherwise  we  should  not  have 
a  true  equivalent  for  the  funded  meaning  of  the  object  described 
as  worth.  On  the  other  hand,  when  we  seek  to  analyse  the 
content  of  the  experience,  we  find  they  are  present  in  different 
degrees  and  different  ways,  and  the  question  arises  which  is  the 
more  fundamental.  Is  then  the  worth-fundamental  feeling  or] 
desire  ? 

In  the  second  place,  whichever  of  these  two  aspects  be  taken 
as  fundamental,  a  second  question  necessarily  arises — is  worth 
coextensive  with  feeling  or  desire,  or  is  there  a  further  demar 
cation  within  the  sphere  of  feeling  or  desire  ?  In  other  words, 
have  all  feelings  or  desires,  whatever  their  conditions,  however 
fleeting  and  however  caused,  the  transgredient  and  immanental 
references  which  characterise  the  worth  attitude  of  the  subject 
toward  the  object  ? 


Definition  of  the  Consciousness  of  Value  35 


III.  PSYCHOLOGICAL  ANALYSIS  OF  WORTH  EXPERIENCE — WORTH 
AS  FEELING  WITH  CERTAIN  COGNITIVE  PRESUPPOSITIONS 

i.  The  Worth-Fundamental  is  Feeling,  not  Desire — Criticism  of 

Ehrenfels. 

Both  of  these  problems  have  been  in  the  forefront  of 
recent  psychological  analysis  of  value.1  They  are  questions 
which  are  forced  upon  the  attention  as  soon  as  we  attempt  to 
co-ordinate  and  reduce  to  common  terms  the  varying  attitudes 
which  have  been  included  within  the  definition  of  worth 
experience.  It  is  true  that  there  is  a  point  of  view  from  which 
these  finer  distinctions  are  irrelevant.  One  can  see  that  for 
the  limited  purposes  of  economic  analysis,  which  requires  but  a 
short  excursion  into  psychology,  we  might  speak  of  the  worth 
moment,  now  as  feeling,  and  now  as  desire.  Ehrenfels  is  also 
probably  right  in  saying  that  the  general  laws  of  valuation 
and  the  forms  of  mutation  of  values  or  value-movement,  hold 
true  whether  we  define  worth  experience  as  feeling  or  desire, 
and  that  changes  in  judgment  of  value  are  due  to  modifications  of 
feeling  or  desire.  Nevertheless,  a  complete  analysis  of  the  worth 
consciousness,  in  all  its  phases,  requires  the  solution  of  both  these 
problems. 

It  is  in  connection  with  the  first  problem  that  the  first  diver 
gence  in  definition  appears,  as  typified  in  the  different  formula 
tions  of  Meinong  and  Ehrenfels.  Ehrenfels  defines  the  worth 
of  an  object  as  its  desirability  and  makes  actual  desire  the 
fundamental,  assigning  to  feeling  a  merely  dispositional  role ; 
while  Meinong,  on  the  other  hand,  identifies  actual  worth 
experience  with  feeling,  desire  appearing  in  his  definition  only 
as  presupposed  disposition.  In  some  sense,  we  have  seen,  both 
terms,  feeling  and  conation,  must  enter  into  our  psychological 
definition  ;  the  question  is  which  shall  be  given  the  role  of 
fundamental,  actual  experience,  and  which  that  of  disposition. 

Ehrenfels2  takes  desire  as  the  actual  psychological  worth- 
fundamental.  Value,  we  are  told,  is  proportional  to  the  desira 
bility  of  the  object — and  he  continues,  as  though  it  were  self- 

1  For  a  detailed  historical  statement  and  criticism,  see  the  writer's  article  "Recent 
Tendencies   in  the  Psychological  Theory  of  Value,"  Psychological  Bulletin,  Vol.  IV, 
No.  4,  March  15,  1907. 

2  Ehrenfels,    System  der   Wert-theorie,   Leipzig,    1897,  Vol.   I,   chap.    I,  especially 
P-  35- 


36  Valuation  :   its  Nature  and  Laws 

evident, — 'i.e.,  to  the  strength  of  the  actual  desire  which  cor 
responds  to  it."  The  first  part  of  the  definition  is  certainly  true. 
The  funded  meaning  of  an  object  is  its  desirability,  its  capacity, 
under  certain  circumstances,  of  calling  out  desire.  The  second 
part  does  not,  however,  necessarily  follow.  It  does  not  follow 
either  that  judgments  of  worth  are  determined  by  actual  desire, 
or  that  the  worth  of  the  object  is  proportional  to  the  strength  of 
the  actual  desire.  As  to  the  identification  of  value  or  desira 
bility  with  actual  desire,  a  consideration  of  certain  simple  but 
typical  worth  experiences  indicates,  that  it  is  not  exclusively  an 
actual,  but,  ultimately,  merely  a  possible  desire  or  desire-dis 
position  with  which  worth  is  to  be  equated,  a  modification  of  his 
earlier  definition  which  Ehrenfels  himself  accepts.  When  I 
think  of  an  absent  friend,  I  may  feel  his  worth  to  me  without  the 
slightest  trace  of  actual  desire  for  his  immediate  presence, 
although  the  presupposition  of  that  feeling  is  a  disposition  so  to 
desire.  Or  again,  my  consciousness  of  the  objective  value  of 
objects  of  economic  use  may  be  independent  of  any  actual  desire, 
although  not  of  my  knowledge  of  their  desirableness  under  certain 
circumstances.  It  is  equally  true  that  the  degree  of  worth  or 
desirability  of  an  object  cannot  be  straightway  identified  with 
the  degree  of  actual  desire.  It  is  undoubtedly  proportional 
to  the  strength  of  desire-disposition  presupposed,  but  the  strength 
of  a  conative  tendency  or  disposition  is  not  always  measured  by 
the  intensity  of  actual  desire ;  is  often  inferred  indirectly  from 
its  effects  on  volition,  or  through  the  intensity  of  the  emotional  dis 
turbance  following  upon  arrest.  The  assumption  that  the  strength 
of  a  desire-disposition  is  given  directly  in  immediate  modifications 
of  consciousness,  is  one  which  introspection  makes  highly  im 
probable,  and  Ehrenfels,  at  least,  with  whose  definition  we  are 
here  concerned,  does  not  admit  it. 

It  is  clear,  then,  that  while  desire,  and  conative  tendency  in 
general,  must  find  a  place  in  our  worth  definition,  it  cannot  be 
taken  as  the  psychological  fundamental  in  the  sense  that  it  is 
the  conscious  correlate  of  the  funded  meaning  of  the  object. 
This  conscious  correlate  is  feeling.  Ehrenfels  thus  brings  feeling 
into  his  definition.  Desire  is  not  determined  by  mystical  qualities 
of  objects  but  by  aspects  of  our  consciousness  which  can  be 
reduced  to  psychological  terms.  "  All  acts  of  desire  are  deter 
mined,  in  regard  to  their  direction  as  well  as  their  strength,  by 
the  relative  increase  of  pleasure  which  they,  according  to  the 
affective  dispositions  of  the  individual  in  question,  bring  with 


Definition  of  the  Consciousness  of  Value  37 

them  upon  their  entrance  into,  or  continuance  in  consciousness." 
Feeling  is,  therefore,  after  all,  primary.  The  worth  of  an  ob 
ject  is  directly  proportional  to  the  strength  of  desire,  but  this 
strength  of  desire  is  determined  by  the  difference  of  the  place 
of  the  object  in  the  hedonic  scale. 

In  this  conception  of  Ehrenfels  the  whole  psychological 
problem  of  the  nature  of  feeling  and  desire  and  of  their  rela 
tions,  is  involved.  Into  that  larger  question  we  cannot  here 
enter.  It  will  be  sufficient  to  notice  certain  fundamental  diffi 
culties  which  have  been  generally  recognised  by  the  critics  of 
the  position.  The  criticism  turns  upon  the  concept  of  the  deter 
mination  of  desire  by  feeling,  upon  the  idea  of  the  causal  rela 
tion  involved.  It  is  maintained  with  justification  that  for  a  feel 
ing  to  be  a  cause  of  desire  it  must  be  actual,  that  is  a  present 
state  of  consciousness.  But  according  to  Ehrenfels'  conception  it 
is  not  always  a  present  state,  but  often  a  state  which  does  not  yet 
exist,  which  is  said  to  be  the  cause.  It  is  the  existence  of  an  object 
nbt  yet  realised  or  the  non-existence  of  a  present  object,  which  is 
desired.  The  hedonic  accompaniment  of  a  not-yet  existent 
object,  itself  therefore  not  existent,  cannot  in  any  causal  sense 
be  the  determinant  of  desire.  But  it  may  be  said  that  it  is  the 
difference  of  these  two  states  that  is  the  cause.  In  that  case  it 
must  be  either  the  unfelt,  uncognised  difference,  an  abstraction, 
which  is  the  cause,  or  else  a  new  feeling  following  upon  the 
judgment  of  the  difference  between  the  actual  present  feeling 
and  an  imagined  feeling  arising  from  the  assumption  of  the 
existence  or  non-existence  of  the  object.  In  the  first  case,  we 
have  a  conceptual  abstraction  made  the  cause — which  is  impos 
sible.  In  the  second  case,  a  feeling  difference  has  become  the 
object  of  judgment,  and  a  value  moment  is  already  present  prior 
to  desire.  It  is  clear  that  in  some  sense  feeling  or  feeling-dis 
position  is  always  presupposed  by  desire,  but  the  relation  cannot 
be  described  as  causal. 

Ehrenfels  recognises,  that  upon  this  causal  view  of  the  rela 
tion  of  feeling  to  desire,  the  proposition  must  be  modified  to  read  : 
desire  is  determined  by  feeling  or  feeling-dispositions.  But  we 
have  already  seen  that  worth  cannot,  in  every  case,  be  identified 
with  actual  desire,  but  only  with  the  capacity  of  being  desired  or 
desirability.  Thus  Ehrenfels  is  finally  left  without  any  conscious 
correlate  for  the  worth  moment.  Both  the  feeling  and  conative 
aspects  tend  to  become  dispositional. 


38  Valuation :   its  Nature  and  Laws 

2.  The  Criterion  of  Worth  Feeling — Presupposition  of 
Reality. 

For  reasons  of  the  nature  of  those  developed  in  our  criticism 
of  Ehrenfels's  worth  definition,  Meinong *  makes  feeling  the  worth- 
fundamental.  The  sense  of  worth  is  given  in  feeling  -  signs, 
Werth-gefiihle,  which  are  determined  in  character  and  degree  by 
the  nature  of  their  presuppositions  (Voraussetzungen).2  These 
presuppositions  he  further  conceives,  in  the  case  of  worth  feelings, 
to  be  always  judgments  (or  according  to  his  later  formula 
tion,  judgments  and  assumptions — Annahmen),  and  are  there 
fore  distinguishable  from  feelings  which  have  merely  sensations 
or  presentations  as  their  presuppositions.  Leaving  out  of  account 
for  the  moment  the  question  of  this  limitation  of  the  class, 
worth  feelings,  we  may  accept  Meinong's  general  position. 
The  preferability  of  feeling  as  the  fundamental  element  seems 
to  me  to  be  beyond  doubt,  and  for  the  following  reasons. 
In  general  our  argument  would  be  :  There  can  be  no  sense  of 
worth  without  a  meaning  which  may  properly  be  described 
as  felt  meaning,  while  there  can  very  well  be  a  sense  of  worth 
without  that  qualification  which  we  describe  as  desire  and  voli 
tion.  More  specifically,  even  in  those  experiences  which  we  call 
explicit  desire  or  volition,  the  essence  of  the  desire  can  be 
equally  well  described  in  terms  of  feeling  without  doing  violence 
to  our  speech.  'Yhe  essence  of  desire  is  the  feeling  of  lack  or 
want.  We  "  fed  the  need "  of  something.  What  further 
qualifies  desire  is  the  kinaesthetic  sensations  which  are  irrelevant 
accompaniments  from  the  standpoint  of  the  essential  worth 
moment.  But  it  is  by  no  means  in  the  same  sense  true  that 
every  worth  experience  involves  explicit  desire.  We  may 
actually  feel  the  worth  of  objects  without  the  slightest 
trace  of  that  qualification  of  our  feeling  which  we  describe  as 
actual  desire  for  their  presence,  although  a  conative  disposition 
is  presupposed  and  may  become  explicit  under  suitable  conditions. 
The  same  is  true  of  aesthetic  and  mystical  states  of  repose  where 
actual  desire  is  in  abeyance. 

What   this   means   for   our   definition   is   clear.      In   actual 
worth    experience    desire  is  not  necessarily    present    although 

1  Meinong,  Psychologisch-cthische  Untersuchungen,  Part  I,  chap.  I. 

2  In  presenting  Meinong's  position  I  have  translated  Voraussetzung  "presupposi 
tion"  rather  than  pre-condition,  as  better  adapted  to  convey  his  meaning,  and  have 
retained  this  broader  usage  of  presupposition  throughout,  although  in  the  usage  of 
Baldwin   it   is  confined  to  the   higher  reflective   level,    that   is,    if  I    understand   his 
position  correctly,  his  presupposition  is  always  a  "presupposition  of  belief." 


Definition  of  the  Consciousness  of  Value  39 

feeling  is.  The  desire  is  present  often  merely  as  a  dispositional 
moment  which,  however,  may  become  actual  under  certain 
definite  circumstances.  In  so  far,  therefore,  as  our  definition 
includes  the  element  of  desire,  we  must  enlarge  it  to  read— 
an  object  has  worth  in  so  far  as  it  is  either  desired  or  has  the 
capacity  of  calling  out  desire,  has,  in  other  words,  desirability. 
This  definition  includes  the  mystical  and  sesthetic  states  of 
repose  already  referred  to,  for  no  object  can  become  the  object 
of  such  feelings  which  has  not  been  desired  and  may  not  under 
some  circumstances  again  be  desired.  Conation  is  present  dis- 
positionally  (how  we  shall  see  later)  even  in  these  states  of  repose. 
But  the  case  is  different  with  feeling.  In  defining  worth  as 
feeling  with  certain  characteristic  presuppositions  we  mean 
that  every  actual  worth  judgment  implies  actual  feeling — even 
in  those  cases  where  the  worth  attitude  is  scarcely  distinguishable 
from  the  cognitive. 

Feeling  having  been  taken  as  the  actual  conscious  correlate 
of  worth  predicates,  the  second  problem  arises — whether  worth 
feelings  are  coextensive  with  feelings  in  general,  or  whether  some 
further  differentiation  within  the  general  class,  feeling,  is  re 
quired.  It  is  at  this  point  that  the  definition  of  Meinong,  that 
feelings  of  worth  are  exclusively  "  judgment-feelings,"  becomes 
important.  This  view,  which  may  be  described  as  the  in- 
tellectualistic  theory  of  worth  experience,  has  given  rise  to  so 
many  important  developments  in  ethics  and  aesthetics  that 
it  demands  the  most  careful  consideration.  Negatively  viewed, 
it  denies  the  character  of  worth  experience  to  all  feelings  which 
have  as  their  presuppositions  mere  presentations,  to  all  feelings 
which  may  be  adequately  described  as  the  mere  feeling-tone  of 
the  presentation  or  as  the  effect  of  the  entrance  of  the  presenta 
tion  into  consciousness.  It  differentiates  "  worth  feeling  "  from 
mere  "  pleasure-causation,"  i.e.  pleasure  viewed  as  mere  re 
action  to  stimulus. 

Before  considering  in  detail  the  psychological  grounds  for 
this  view,  it  will  be  well  to  observe  the  more  general  fact  that, 
whether  worth  -experience  be  defined  in  terms  of  desire  or  feeling, 
it  cannot  be  made  co-extensive  with  either.  Desire,  in  itself, 
does  not  constitute  the  experience  of  valuation  :  there  are  fleet 
ing  desires  which  do  not  attain  to  the  level  of  valuation,  a  fact 
which  leads  Kr tiger  in  his  definition,  which  is  in  terms  of  desire, 
to  make  the  differentia  of  worth  a  certain  constancy  of  desire. 
Again,  as  Meinong  points  out,  illustrations  are  plentiful  of  valu- 


4°  Valuation :   its  Nature  and  Laws 

ation  without  actual  consciousness  of  pleasure,  while  a  fleeting 
pleasure    does    not    necessarily    involve    valuation.     Reflection 
upon  these  facts  of  experience  leads  to  more  strictly  logical 
considerations  such  as  those  which  appeared  in  our  criticism  of 
Ehrenfels's  definition.     The  sense  of  value  cannot  be  identified 
with  the  mere  feeling  of  pleasure,  although  of  course  a  feeling  of 
pleasure  when  it  is  made  the  object  of  judgment  may  become 
a  value,  for  the  feeling  of  value  is  conditioned  not  only  by  the 
presence  of  objects,  but  also  by  their  absence.    The  mere  absence 
of  the  object  is  not  the  condition  of  the  feeling,  but  the  cogni 
sance    (in    Meinong's    terms    the   judgment)    of   non-existence. 
The  hedonic  state  which  would  be  the  effect  of  the  presence  of 
the  absent  object  is  not  actual,  and  can  therefore  not  be,  in  any 
causal  sense,  the  condition  of  the  desire  and  of  valuation.    More 
over,  the  cause  of  the  pleasure  is  often  quite  distinct  from  the 
object  of  the  feeling  of  value,  often  physiological  and  uncon 
scious.     The  feeling  of  value  can  therefore  not  be  viewed  as  the 
effect   or   accompaniment   of  sensation   or  presentation   of   an 
object,  but  as  conditioned  by  some  presupposition  of  the  existence 
of  the  object.     In  the  case  of  the  reflex  feeling  of  the  value  of  life, 
conditioned    by    organic    sensations,    the    feeling    is    objectless 
but  contains   a  primitive  presumption   of  reality  which   then 
maintains  itself,  as  an  explicit  judgment  of  value,  by  attaching 
itself  to  objects  which  form  the  concrete  content  of  life. 

The  negative  aspect  of  Meinong's  position,  the  denial  of  the 
character  of  worth  experience  to  mere  presentation-feelings, 
appears  justified  from  this  analysis  of  the  facts.  A  funda 
mental  distinction  seems  to  exist  between  feeling  which  is  a 
mere  feeling-tone,  accompaniment  or  effect  of  a  sensation  or 
revived  image,  and  feeling-attitude  which  is  characterised  by 
the  direction  of  the  feeling  toward  the  object.  Feeling-attitudes 
alone  seem  to  contain  the  worth-moment.  It  is  undoubtedly  true 
that  feeling-tone  of  presentation,  when  it  reaches  a  certain  degree 
of  intensity,  gives  rise  to  a  feeling-attitude,  to  the  presentation 
of  the  cause  as  object  and  the  direction  of  judgment  toward  it, 
and  thus  to  feeling  of  worth.  But  this  feeling  or  desire,  as  the 
case  may  be,  is  distinguished  from  the  feeling-tone  by  the  presence 
of  additional  presuppositions ;  whether  exclusively  judgmental 
or  not  is  a  question  to  be  determined. 

A  critical  consideration  of  this  positive  aspect  of  Meinong's 
definition  requires  a  closer  examination  of  his  use  of  the  term 
presupposition  (Voraussetzung) .  Under  this  concept  he  includes 


Definition  of  the  Consciousness  of  Value  41 

all  those  conditions  of  feeling  which  are  psychical  in  character, 
as  distinguished  from  other  causes  of  feeling  which  may  be  dis- 
positional  and  physiological.  In  this  sense  a  presupposition 
may  be  any  psychical  process,  presentation,  judgments,  the 
various  types,  categorical,  hypothetical,  etc.,  and  other  types  of 
function,  perhaps,  such  as  assumption.  In  every  case  where 
the  presupposition  of  a  feeling  is  spoken  of,  the  feeling  is  directed 
toward  an  object  and  is  conditioned  by  some  psychical  act,  of 
presentation,  of  imagination  with  its  assumption  of  reality,  or 
of  judgment,  which  is  for  Meinong  a  fundamental  form  of 
psychical  process.  The  significance  of  this  distinction  is  to  be 
found  in  the  fact  that  the  characteristic  meanings  of  feelings 
which  distinguish  them  as  feelings  of  value,  are  not  to  be  differ 
entiated  in  terms  merely  of  the  objects  toward  which  the  feeling 
is  directed,  nor  yet  in  terms  of  the  causes  of  the  feeling,  but  in 
terms  of  the  cognitive  acts  or  attitudes  which  relate  the  object 
to  the  subject. 

(a)  The  Presupposition  of   Reality  not  exclusively  Existential 
Judgment — Criticism  of  Meinong. 

Is  then  the  presupposition  of  worth  feeling  exclusively  judg 
mental,  as  Meinong  maintains  ?  To  this  question  our  answer 
must  be  negative.  But  we  may  admit,  to  begin  with,  that 
many  types  of  worth  attitude  do  have  existential  judgments 
as  presuppositions,  and  that  all  secondary  modifications  of  worth 
attitude  are  determined  by  the  inclusion  or  exclusion  of  judg 
ments,  existential  and  relational,  as  part  presuppositions  of 
the  feeling.  But  that  there  is  no  primary  immediate  con 
sciousness  of  value  without  explicit  judgment  of  existence  or 
non-existence  of  the  object,  cannot  be  maintained.  As  was 
pointed  out  in  our  discussion  of  the  equivocations  in  the  worth 
predicates,  ideal  and  imputed  values  may  be  attributed  to  ob 
jects  when  the  question  whether  they  exist  or  may  be  acquired 
is  not  raised,  and  when,  accordingly,  the  attitude  can  never 
reach  the  point  of  explicit  judgment.  The  activities  of  imagina 
tion  and  idealisation  abundantly  prove  that  the  feelings  directed 
upon  their  objects  are  really  feelings  of  worth  and  are  determina 
tive  of  worth  judgments,  although  they  presuppose  mere  as 
sumptions  of  the  reality  of  the  objects  which  do  not  require  to  be 
converted  into,  or  explicitly  acknowledged  in,  existential  judg 
ments. 


4 2  Valuation :    its  Nature  and  Laws 

Meinong  has  indeed  found  himself  compelled  upon  further 
reflection    to   modify    his    definition    to    the    extent    that    he 
includes  with  the  judgment-feelings  assumption-feelings     (An- 
nahme-gefiihle).     He    recognises,   that   "  often    one   values    an 
object   at  a  time   when   there   is   entirely   wanting   all   chance 
of  judgments   of    existence   and   non-existence,    because    it    is 
not  yet  determined  whether  the  object  thought  of  as  in  the 
future    will    exist    or    not."     Moreover,    "it    is   possible,    and 
it  frequently  happens,    that  we  value  an  abstractly  presented 
object  without  inquiring  after  its  existence."1     And  in  a  later 
paper2  he   further   qualifies   his   position   by   recognising   that 
it  is  only  some  universe  of  reality  which  is  necessarily  presupposed, 
in  that  the  presuppositions  are  not  necessarily  categorical  existen 
tial  judgments,  but  may  be  hypothetical  or  disjunctive.     Now 
in  all  these  cases  where  the  object  is  "  abstractly  presented," 
assumed  to  exist,  or  asserted  to  exist  conditionally,  reality  is 
presupposed  in  some  sense ;  there  is  some  reference  to  reality. 
It  is  also  clear  that  in  all  these  cases  the  feeling  characterised 
as  feeling  of  value  is,  by  this  very  reference  to  reality  presupposed, 
in  some  way  differently  qualified  from  the  feeling  of  pleasantness 
or  unpleasantness.     The  question  at  issue  is  merely  as  to  the 
proper  characterisation  of  the  reality-meaning,  whether  it  rests 
exclusively  upon  existential  judgment  or  not. 

(b)  Criticism  of  the  Theory  of  Existential  Judgment  underlying  this 
View — Existential  Judgment  merely  Acknowledgment  of  a 
Presupposition  of  Reality. 

This  question  is  still  more  ultimately  conditioned  by  a 
theory  of  the  existential  judgment.  To  this  theoretical  problem 
we  shall  presently  turn,  but  it  will  be  in  the  interests  of  clear 
ness  to  seek  a  preliminary  characterisation  of  this  presupposi 
tion  of  reality.  There  can  be  no  question,  in  the  first  place, 
that  wherever  there  is  the  feeling  of  value,  there  is  reality  feel 
ing.  When  once  an  object,  the  existence  of  which  was  what  I 
desired  or  what  conditioned  my  feeling  of  value,  is  explicitly 
judged  non-existent,  the  object  undoubtedly  loses  its  value  for 
me.  The  essential  condition  of  its  being  valued  is  eliminated. 

1  Meinong,  "  Uber  Werthalten  und  Wert,"  Archiv  fur  Systematise)*  Philosophic, 
l89S»4PP-  327-46.    Also  his  later  work,  Uber  Annahmen. 

"  (Jrtheilsgefuhle,  was  Sic  sind  und  was  Sie  nicht  sind"  Arckiv  fur  die  gesammtc 
Psychologic,  Vol.  VI,  1905. 


Definition  of  the  Consciousness  of  Value  43 

My  appreciation  of  the  worth  of  an  object  does  not,  however, 
necessarily,  and  in  every  case,  rest  upon  such  explicit  judgment 
of  existence,  but  at  most  upon  a  primary  undisturbed  presumption 
of  reality.  By  this  primary  presumption  of  reality,  of  a  reality, 
moreover,  in  which  the  more  specific  existence  meaning  has  not 
yet  been  differentiated,  is  to  be  understood  the  mere  act  of 
acceptance,  taking  for  granted,1  prior  to  the  explicit  taking  up 
of  the  object  into  a  pre-determined  sphere  of  reality  through 
the  existence  predicate,  and  prior  to  the  assumption  of  existence 
of  an  object  in  the  interest  of  continuity  of  any  trend  or  activity, 
whether  of  the  type  of  cognition  or  valuation. 

As  illustrative  of  this  attitude  of  primitive  presumption  we 
may  consider  first  the  reality-feeling  which  attaches  to  percep 
tion  and  presentation  simply  because  of  the  "  recognitive  mean 
ing  "  !  which  they  have.  Distinctions  between  existent  and 
non-existent  arise  later — more  especially  in  the  case  of  pre 
sentations  in  the  fancy  or  imagination  mode.  They  are 
presumed  to  be  real  until  the  entrance  of  illusion-disturbing 
moments  which  require  the  presumption  to  pass  over  into  ex 
plicit  judgment  and  conviction  either  of  existence  or  non-exist 
ence.  The  fairy  world  of  the  child  is  a  wrorld  neither  of  pure 
presentation  nor  of  existential  judgment  but  of  presumption. 
The  same  may  be  said  of  many  ideals  of  the  more  developed 
mind,  as  for  instance,  the  religious,  about  which  questions  of  ex 
istence  and  non-existence  are  not  seriously  asked.  In  all  these 
cases  some  psychically  pre-determined  demand,  whether  arising 
from  a  more  objective  cognitive  factor  of  recognition  or  a  more 
subjective  factor  of  conative  disposition  or  interest,  creates  a 
presumption  of  reality. 

Such  presumption  must  be  carefully  distinguished  from  both 


1  The  use  of  the  term  presumption  to  characterise  this  relation  to  reality  is,  I  think, 
fully  justified  both  linguistically  and  psychologically.     Our  ordinary  speech,  it  is  true, 
frequently  fails  to  distinguish  between  presumption  and  assumption,  and  has,  moreover, 
read  into  the  word  presumption  a  certain  ethical  connotation  which  partially  unfits  it 
for  the  present  use.     On  the  other  hand,  the  original  meaning  of  the  Latin  praesumptio 
is  much  nearer  to  the  use  that  we  have  in  mind — it  had  more  the  meaning  of  taking  for 
granted  prior  to  explicit  judgment  and  was  quite  different  from  the  conscious  assump 
tion  of  reality  as  we  have  it  in  hypothesis.     The  modern  English  dictionaries  give  as 
one  of  the  renderings,  taking  for  granted,  the  meaning  here  emphasised.     The  use  of 
the  term  in  formal  logic,  as  in  fallacies  of  presumption,  while  at  first  apparently  against 
our  usage,  on  closer  inspection  seems  to  favour  it.     A  presumption  is  a  material  fallacy, 
an  unconscious  pre-logical  taking  for  granted.     Finally,  the  value  of  the  introduction  of 
this  term  for  our  immediate  purpose  is  to  be  found  in  the  possibility  it  affords  of  using 
the  prefixes  pm,  sub,  and  ad,  with  the  same  root,  to  designate  modifications  of  cognitive 
attitude. 

2  Baldwin's  distinction  referred  to  above. 


44  Valuation :   its  Nature  and  Laws 

judgment  and  assumption.  The  existential  judgment  arises, 
we  shall  see,  only  after  disturbance  in  a  sphere  of  reality  already 
presupposed ;  it  is  an  act  which  takes  place  only  after  some  dis 
position,  some  tendency  to  recognition,  or  to  renewal  of  attitude 
of  feeling  or  will  meets  with  opposition  or  arrest.  It  must  be 
equally  clearly  distinguished  from  the  later,  more  developed, 
attitude  of  assumption  of  existence  which  presupposes  dis 
positions  already  created  by  actual  judgment.  The  assumption, 
except  when  it  is  what  we  describe  as  an  unconscious  assump 
tion  (and  then  it  is  really  an  approximation  to  presumption), 
recognises  the  possibility  of  the  non-existence  of  the  object,  and 
in  some  modes  of  playful  assumption  (the  "  semblant  modes  " 
of  Professor  Baldwin)  it  is,  so  to  speak,  on  the  verge  of  explicit 
judgment  of  non-existence.  But  in  the  making  of  the  assumption 
the  act  is  determined  by  a  subjective  factor,  a  demand  arising 
from  already  existing  dispositions  and  interests.  The  assump 
tion  is  an  acknowledgment  of  this  demand. 

It  is  obvious,  after  this  analysis,  that  the  definition  of  feeling 
of  value  under  consideration,  as  feeling  with  existential  judg 
ment  as  its  presupposition,  is  possible  only  on  the  theory  that 
the  primitive  form  of  judgment  is  the  mere  act  of  acceptance 
(acknowledgment) *  or  rejection  and  involves  no  relational  aspect, 
no  separation  of  two  elements,  subject  and  predicate.  The 
existential  judgment  is  identified  with  acceptance  and  the  non- 
existential  with  rejection.  If  this  view  of  judgment  (Brentano's)2 
can  be  maintained,  it  follows  necessarily  that  there  can  be  no 
feeling  of  value  without  a  judgment  as  presupposition,  for  all 
attitude  is  primarily  acceptance  or  rejection,  and  the  feeling  of 
value  is  an  attitude,  not  mere  presentation  plus  feeling.  But  can 
mere  acceptance  or  rejection  be  identified  with  judgment  of 
existence  and  non-existence,  and  at  the  same  time  any  useful 
conception  of  judgment  be  retained  ?  I  think  not,  and  for  the 
following  reasons. 

The   view    that    they   can   implies :    (a)  That    presentation 


1  The  use  of  the  terms  acknowledgment  and  rejection  as  correlative  is  most  unfortu 
nate,  for  it  prejudices  the  whole  question.     Rejection,  as  any  dictionary  tells  us,  is  not 
the  opposite   of  acknowledgment.     Acknowledgment   has   as   its   opposite  disavowal, 
while  the  opposite  of  rejection   is   acceptance.      This   linguistic  relation   corresponds 
precisely  to  the  psychological.      Acknowledgment  and  disavowal   both    represent  the 
explicit  judgmental  acts  by  which  a  reality  already  presupposed  is  affirmed  or  denied. 
Mere  acceptance  or  rejection  of  an  object  presupposes  nothing  more  than  a  presumption 
of  reality  or  disturbance  of  that  presumption. 

2  For  a  presentation  and  discussion  of  Brentano's  theory  of  judgment  see  Stout, 
Analytical  Psychology,  Vol.  I,  chap.  V. 


Definition  of  the  Consciousness  of  Value  45 

and  judgment  (acceptance  or  rejection  of  the  existence  of  the 
presented)  are  two  different  and  irreducible  elementary  aspects 
of  consciousness  ;  (6)  that  while  the  affirmation  or  negation 
of  A  (as  function)  adds  something  to  its  mere  presentation 
(as  function),  the  affirmation  or  negation  of  ^4's  existence 
(as  content)  adds  nothing  to  the  affirmation  or  negation  of  A 
(as  content).  The  first  thesis  is  the  key  to  the  position.  Is 
there  such  a  thing  as  simple  apprehension,  presentation  with 
out  acceptance,  or  does  apprehension  involve  apprehension  of 
existence  ?  At  first  sight  the  former  of  the  two  possible 
alternatives  seems  to  be  true.  From  the  standpoint  of  analysis 
alone,  we  seem  to  find  cases  in  which  the  element  of  affirmation 
is  at  a  minimum,  or  is  even  entirely  lacking,  and  in  which 
a  merely  presentational  consciousness  remains.  Leaving  out 
of  account  the  case  of  doubt  or  suspended  judgment  where, 
although  at  a  minimum,  tendencies  to  judgment  still  remain, 
we  may  turn  immediately  to  the  typical  case  of  aesthetic 
contemplation.  Here  it  is  said  we  have  a  strictly  presenta 
tional  consciousness,  at  least  when  the  contemplation  is  pure, 
when  the  aesthetic  is  unmixed  with  other  factors.  This  view  we 
shall  find  it  necessary  to  reject,  and  for  the  following  reasons. 
In  the  first  place,  aesthetic  contemplation  is  an  attitude — not 
mere  presentation  ;  in  it  there  is,  as  Ehrenfels  says,  at  least  a 
resting  in  reality,  "  ein  Haften  an  der  Wirklichkeit,"  either 
outer  or  inner.  As  such  it  is  more  than  mere  presentation. 
While  for  the  purposes  of  the  psychologist  the  idea  of  a  purely 
presentational  consciousness  is  sometimes  a  useful  abstraction, 
every  actual  experience  presupposes  a  minimum  of  acceptance  or 
rejection.  The  procedure,  therefore,  which  takes  this  abstraction, 
made  for  purposes  of  analysis,  as  a  picture  of  reality,  and  from  it 
infers,  for  instance,  the  unreality  of  the  aesthetic  object  and  ex 
perience  and  its  exclusion  from  the  sphere  of  worth  experience, 
is  vitiated  by  serious  fallacy. 

But  if  the  merely  presentational  consciousness  be  but  an  ab 
straction,  there  still  remains  the  question — to  what  extent,  in 
actual  concrete  cases  of  aesthetic  contemplation,  all  acceptance  and 
rejection  may  be  seen  to  be  excluded  and  the  purely  presentational 
approached.  Perhaps  the  difference  is  negligible.  Most  aesthetic 
attitudes,  it  is  recognised  by  all,  fail  to  give  us  this  contempla 
tion  pure.  In  the  sublime  and  tragic,  for  instance,  pseudo- 
aesthetic  factors,  so  called,  enter  in,  in  the  form  of  acknow 
ledgments  and  rejections,  judgments  of  various  kinds.  Even 


46  Vahiation :   its  Nature  and  Laws 

beauty,  in  its  narrower  sense,  contains,  as  partial  moments, 
normative  judgments.  If  we  are  to  find  any  concrete  aesthetic 
experience  of  "  pure  contemplation,"  it  must  be  in  the 
simplest  perceptual  forms  and  form-qualities.  These  are  indeed 
usually  taken  as  the  typical  aesthetic  objects  when  the  aesthetic 
is  thus  denned.  But  even  here  it  is  doubtful  whether  the 
element  of  acceptance  and  rejection,  or  of  conation,  can  be  ex 
cluded.  It  is  true  that  these  forms  and  form-qualities,  when 
abstracted  from  the  elements  in  which  they  inhere,  may  be  viewed 
as  the  objects  of  purely  presentational  activity.  Nevertheless 
their  construction  was  the  product  of  conative  activity  which 
involved  spontaneous  acceptance  and  rejection,  presumption  of 
reality.  Viewed  genetically,  every  aesthetic  feeling  of  form 
presupposes  a  disposition  created  by  preceding  conative  activity. 

The  distinction  between  simple  apprehension  and  acceptance 
is,  then,  even  in  aesthetic  contemplation,  a  relative  one.  What 
shall  be  said  of  the  second  part  of  the  thesis — that  acceptance 
or  rejection  of  an  object,  A,  is  identical  with  the  affirmation  or 
negation  of  the  existence  of  A,  or,  in  other  words,  with  judgment  ? 
Acknowledgment  or  rejection  does  undoubtedly  presuppose  the 
reality,  in  some  sense,  of  the  presentational  content.  This  is 
the  same  as  saying  that  all  conation  is  directed  toward  objects 
presumed  to  be  real.  It  does  not  follow,  however,  that  explicit 
existential  judgment  is  involved.  We  must,  I  think,  look  upon 
the  existential  judgment  as  derived  from  a  simpler  and  more 
ultimate  attitude  toward  a  reality  presupposed  in  all  conation, 
even  on  the  perceptual  level.  Acceptance  and  rejection  involve 
presumption  of  existence  but  not  necessarily  judgment. 

Such  a  distinction  between  presumption  and  judgment  in 
volves  of  course  a  theory  of  the  nature  of  judgment.  Into  the 
logical  questions  here  raised  we  cannot  go  in  detail,  but  this 
much  at  least  may  be  said.  The  position  maintained  by  Sigwart 
(among  other  logicians) — that  judgment  "  must  be  regarded 
as  establishing  a  relation,  even  in  its  existential  form,"  seems 
necessary  if  our  conception  of  it  is  to  retain  any  useful  sig 
nificance.1  When  the  relational  aspect  is  allowed  to  lapse,  judg 
ment  becomes  practically  indistinguishable  from  conation.  It 
is  true  that  the  existential  judgment  occupies  a  unique  position. 
It  does  not  establish  a  relation  between  its  subject  and  the  pre 
dicate,  "  being,"  "  but  between  an  object  as  idea  and  an  object  as 
intuited."  Affirmation  of  existence  or  non-existence  presupposes, 

1  Sigwart,  Logic  (translation),  Vol.  I,  p.  72. 


Definition  of  the  Consciousness  of  Value  47 

as  mere  acceptance  or  rejection  does  not,  at  least  the  beginning 
of  the  differentiation  of  subject  and  predicate.1 


3.  The  Presupposition  of  Reality — Presumption,  Judgment,  and 
Assumption  of  Existence  of  Objects — Analysis  of  these  Cog 
nitive  Attitudes. 

On  the  theory  of  judgment  here  developed,  the  existential 
judgment  and  the  pure  presentation,  in  so  far  as  "  con 
templation  "  is  pure  presentation,  are  secondary  attitudes, 
derived  from  the  primitive  presumption  of  reality  presupposed 
in  all  acceptance  or  rejection  of  an  object.  The  difference 
between  the  presumption  and  judgment  is  that,  while  in  the 
former  we  have  merely  acceptance  and  rejection,  in  the  latter  we 
have  acknowledgment  and  disavowal,  acceptance  and  rejection 
plus  belief  or  disbelief.  Returning  then  to  the  question  of  the 
necessary  presuppositions  of  the  feeling  of  value,  it  is  clear  that 
there  must  be  the  presumption  of  reality,  for  without  it  there  can 
be  no  attitude  toward  the  object,  seeing  that  attitude  involves 
either  acceptance  or  rejection,  or  disposition  to  accept  or  reject. 
But  it  is  equally  clear  that  the  existential  judgment  cannot  be  the 
sole  and  necessary  presupposition  of  the  feeling,  for  there  can  be 
no  such  explicit  judgment  (acknowledgment  or  disavowal)  except 
as  there  is  already  some  reality  meaning,  some  presupposition  of 
reality.  Again,  the  hypothetical  pure  presentation,  in  so  far  as 
there  is  any  such  mode  of  consciousness,  is  equally  secondary  and 
derived.  It  is  the  result  of  abstraction  from  the  primitive 
presumption  of  reality,  the  result  of  arrest  of  this  presumption 
implicit  in  all  conation.  Meinong's  use  of  the  expression  "  ab 
stractly  presented"  is  significant  in  this  connection.  To  pre 
sent  abstractly  means  to  strip  off  the  reality  feeling  involved 
in  the  first  experience.  This  relation  to  reality  feeling  may, 
however,  be  partially  restored  by  a  further  movement  of  conation 
in  which  the  presented  object  is  assumed  to  exist,  an  attitude  we 
find  characteristic  of  certain  secondary  contemplative  aesthetic 
experiences. 

1  The  following  quotation  from  Baldwin's  Thought  and  Things,  Vol.  II  (chapter  II, 
on  "Acknowledgment  and  Belief,"  p.  17),  puts  the  situation  admirably:  "  The  exist 
ence  meaning  which  the  judgment  always  presupposes  in  the  sense  given,  may,  when 
explicitly  asserted,  be  called  a  predicate  but  not  an  attributive  predicate,  not  a  separate 
element  of  presented  context  or  of  recognitive  meaning,  attributed  to  the  subject  matter. 
It  is  only  the  explicit  assertion  of  the  presupposition  of  belief  in  the  sphere  in  which  the 
subject  matter  is  constituted  an  object  of  thought." 


48  Valuation :   its  Nature  and  Laws 

We  are  thus  led  finally  to  a  consideration  of  the  relation  of  the 
attitude  of  assumption  to  the  primitive  presumption  of  reality  and 
to  the  existential  judgment.  This  question  is  important  for  the 
reason  that  the  special  modification  of  feeling  which  has  assump 
tion  as  its  "  presupposition,"  the  feelings  of  the  imagination  (Phan- 
tasiegefiihle)  of  Meinong's  school,  has  been  made  much  of  in  recent 
discussion.  For  one  thing  it  has  been  asserted  that  these  feelings 
are  not  real  and  therefore  not  feelings  of  value,  although  under 
certain  circumstances  they  may  stand  for,  or  represent,  real 
feelings.  Our  own  view,  which  will  be  developed  more  fully  later, 
is  that  they  are  real  feelings  in  any  sense  which  has  significance 
for  psychology  and  that  they  have  a  presupposition  of  reality, 
although  from  the  point  of  view  of  reflective  evaluation  of  the 
objects  of  such  feelings  (the  axiological  point  of  view)  the  judg 
ments  which  spring  from  these  feelings  may  be  invalid.  But  a 
more  adequate  characterisation  of  the  attitude  of  assumption 
itself  is  our  first  problem. 

Assumption,  as  a  cognitive  attitude,  has  two  meanings. 
According  to  its  first  meaning  it  is  an  acceptance,  a  taking  as 
existent,  of  an  object  when  there  is  an  underlying  sense  of  the 
possibility  of  its  being  non-existent.  In  this  sense  it  is  a  half 
way  stage  between  the  primitive  presumption  of  reality  and  the 
existential  judgment.  In  this  sense  also  it  is  a  secondary  move 
ment  or  act  of  cognition  within  a  developing  sphere  of  reality, 
bounded  by  the  primitive  presumption  of  reality  and  the  ex 
istential  judgment,  affirmative  or  negative.  From  the  point  of 
view  of  conation,  it  is  an  act  determined  by  the  momentum  of 
a  subjective  disposition  or  interest.  In  its  second  meaning  it  is 
not  pre-judgmental  but  post-judgmental,  that  is,  a  permanent 
assumption  is  created  by  habitual  judgment ;  it  presupposes 
dispositions  created  by  acts  of  judgment  and  is  derived  from 
the  judgment  attitude.  In  this  case  the  assumption  approaches 
closely  to  the  presumption,  and  for  this  attitude  the  two  terms 
are  often  used  interchangeably.  It  is  important  to  emphasise 
these  two  meanings1  for  the  feeling  attitudes  involved  are  in  many 
respects  quite  different,  and  the  confusion  of  the  two  has  led 
to  misinterpretation  of  worth  experience.  Thus  the  feelings 
which  attach  to  assumptions  of  the  first  type  may  be  described 
as  feelings  of  the  imagination  ;  they  belong  to  the  mode  of 

1  Baldwin's  recently  published  theory  of  "  schematic "  function  recognises  both 
these  modes  of  "assumption,"  the  existential  judgment  lying,  genetically,  between 
them.  Thotighl  and  Things,  Vol.  I,  chap.  v. 


Definition  of  the  Consciousness  of  Value  49 

semblance  or  "  make-believe."  But  those  which  attach  to 
assumptions  of  the  second  type  are  more  accurately  described 
as  feeling-abstracts  or  affective  signs  and  represent  the  acquired 
or  funded  meanings  of  past  judgment  feelings.  To  this  class, 
we  shall  see  later,  belong  all  those  feelings  or  funded  meanings 
which  inhere  intrinsically  in  general  concepts.  Such  terms 
as  truth,  virtue,  duty,  etc.,  have  functioned  in  particular  ex 
istential  judgments,  and  it  was  upon  the  basis  of  these  judgments 
that  the  feelings  of  value  for  which  these  terms  stand  arose.  But 
when  they  are  thus  formed  they  are  abstractly  valued  without 
explicit  judgments  of  existence  or  non-existence.  They  repre 
sent  an  assumption  which  has  arisen  through  formation  of  habit. 
Explicit  judgment  is  always  the  terminus  of  a  process  of  adapta 
tion.  From  the  primitive  presumption  arises,  through  arrest,-? 
assumption,  which  in  turn  passes  into  judgment  and  the  laterjj 
assumption. 

We  are  now  in  a  position  to  summarise  our  view  as  to  the 
nature  of  simple  appreciation  or  primary  feelings  of  value,  in 
so  far  as  it  is  related  to  Meinong's  criterion.  We  agree  with  him 
to  the  extent  that  we  include  among  the  feelings  of  value  only 
such  feelings  as  have  reality  meanings,  that  is,  have  some  pre 
supposition  of  reality.  As  to  the  nature  of  that  presupposition 
of  reality  we  differ.  We  deny  its  limitation  to  existential  judg 
ment  and  include  the  two  attitudes  of  presumption  and  assump 
tion.  This  may  be  said  to  be  the  result  of  our  critical  analysis  of 
the  meanings  of  experiences  of  worth.  There  remains  still  the 
question  of  the  functional  and  genetic  account  of  these  different 
presuppositions.  Before  undertaking  this  we  must  glance  briefly 
at  another  criterion  of  feeling  of  value  recently  developed,  more 
especially  by  Lipps. 


IV.  THE  GENESIS  AND  RELATIONS  OF  THESE  PRESUPPOSITIONS 
OF  THE  FEELING  AND  JUDGMENT  OF  VALUE 

i.  Criticism  of  the  Theory  that  All  Feelings  of  Value  Presuppose 
Reference  to  the  Personality. 

It  is  maintained  that  all  feelings  of  value  are  feelings  of  per 
sonality — that  the  analysis  which  finds  the  criterion  of  feeling 
of  value  in  the  nature  of  the  attitude  toward  a  transcendent 
object,  really  overlooks  the  significant  moment,  which  is  the 
reference  of  the  feeling  to  the  subject,  the  personality.  Feelings 


5O  Valuation:   its  Nature  and  Laws 

of  value  are  feelings  of  activity  of  the  subject,  the  acts  of  judg 
ment,  etc.,  being  of  only  secondary  importance.  Such  a  criterion 
is  presented  in  the  formula  of  Lipps  i1  "  Der  Wert  jeder  Lust  ist 
bedingt  durch  einen  Personlichkeitswert."  Now,  while  it  is  un 
doubtedly  true  that  there  are  types  of  feelings  of  value  which 
have  as  their  presupposition  explicit  reference  to  the  person 
ality — i.e.,  those  feelings  described  as  values  of  character 
isation,  including  feelings  of  obligation,  desert,  etc.  It  must 
nevertheless  be  recognised  that  these  values  are  secondary 
and  acquired,  that  they  presuppose  judgments  referring  the 
attitude  to  the  presented  self,  the  self  being  the  product 
of  an  ideal  construction  based  upon  preceding  experiences  of 
value.  The  only  sense  in  which  Lipps's  statement  may  be  said 
to  be  true  is  that  in  primary  feelings  of  value,  as  distinguished  from 
simple  pleasure,  there  are  certain  modifications,  certain  implicit 
meanings  which,  when  reflected  upon,  lead  to  their  reference  to 
the  self.  Such  a  modification  of  his  view  we  may  accept. 

The  meanings  which  appear  on  the  level  of  simple  appre 
ciation  prior  to  reference  to  the  self,  Kriiger2  has  described  as 
depth  and  breadth  of  the  feeling  in  the  personality,  and  he  con 
ceives  them  to  constitute  a  third  dimension  of  feeling,  besides  its 
intensity  and  duration,  a  dimension  which  is  determined  by  a  rela 
tive  constancy  of  disposition.  His  development  of  the  criterion 
is  both  analytical  and  genetic.  Valuation  is  distinguished  from 
mere  desire  and  simple  "  pleasure-causation  by  a  moment  of 
relative  constancy  of  desire.  Desire  of  itself  does  not  constitute 
valuation,  and  valuation  is  never  mere  desire  or  a  series  of  desires. 
He  further  conceives  the  relation  of  this  "desire-constant"  to 
the  individual  desires  on  the  analogy  of  the  relation  of  concepts 
to  particular  sensations  and  percepts.  A  valuation  always  pre 
supposes  a  relatively  constant  disposition,  and  this  disposi 
tion  appears  as  an  actual  element  in  consciousness  only  in 
a  corresponding  judgment.  Yet  the  judgment  of  value  is  not 
the  valuation  itself.  This  is  given  rather  in  the  characteristic 
modification  of  the  experienced  desire  and  feeling,  which  he  con 
ceives  to  grow  in  depth  with  the  development  of  the  "  desire-con 
stant."  3  He  suggests  that  it  is  probable  that  in  the  first  stages 

1  Lipps,  Die  ethischen  Grundfragen,  Leipzig,  1899,  Chap.  I. 

a  Kriiger,  Der  Btgriff  des  absolut  Wertvollen  als  Grundbcgriff  der  Moral  philosophic, 
Leipzig,  1898,  chap.  Ill  ("Zur  Psychologic  des  VVertes"). 

a  One  point,  however,  he  has  left  undetermined.  Is  the  worth  experience  given  in 
feeling  or  desire  ?  In  some  passages  he  speaks  as  though  the  sense  of  worth  were  given 
in  feeling  as  determined  by  or  as  determining  desire,  in  others  as  though  it  were  given 
in  the  experiences  of  desire  themselves.  As  a  matter  of  fact,  he  does  not  seem  to  have 


Definition  of  the  Consciousness  of  Value  51 

of  conscious  life  only  that  was  consciously  striven  after  which 
brought  with  it  relative  increase  of  pleasure,  and  that  the  con 
sciousness  of  value  has  probably  taken  its  rise  in  such  strivings, 
but  every  desire  has  a  tendency  to  develop  a  relative  constancy 
and  thus  to  pass  into  a  valuation.  It  leaves  behind  in  the  per 
sonality  constant  dispositions,  and  with  them  traces  of  value. 
The  mechanism  of  pleasure-causation  is  thus  broken  through  by 
the  formation  of  values  ;  and  as  soon  as  the  function  of  valua 
tion  is  formed  at  a  single  point,  the  will  ceases  to  be  exclusively 
determined  by  the  intensity  and  duration  of  expected  pleasure. 
Through  the  fact  of  valuation  the  affective-volitional  life  gets,  so 
to  speak,  a  third  dimension;  the  value  of  a  constant  desire  is 
determined  by  its  breadth  and  depth  in  the  personality. 

The  interest  of  this  definition  of  Kriiger's  is  to  be  found  in  the 
fact  that  it  is  an  attempt  to  connect  the  appreciative  distinctions 
which  differentiate  feelings  of  value  from  other  feelings,  which 
lead  ultimately  to  the  characterisation  of  the  self  and  to  the  ex 
plicit  reference  of  the  object  to  the  self,  with  the  functional, 
dispositional  conditions  of  the  feeling,  and  it  has  been  presented 
here  at  some  length  because  this  concept  of  conative  constants 
or  dispositions  as  the  necessary  conditions  of  feelings  of  value, 
feelings  with  depth  and  breadth,  is  precisely  the  concept  which  we 
need  to  connect  these  appreciative  meanings  with  the  reality 
meanings  which  the  preceding  analysis  has  distinguished. 
At  an  earlier  stage  it  was  seen  that  both  the  concepts  of  feeling 
and  conation  must  find  a  place  in  the  definition  of  worth  ex 
perience.  It  is  now  seen  that  feelings  of  value  are  not  com 
pletely  characterised  by  reference  to  their  presuppositions  of 
reality,  presumption,  judgment  and  assumption,  but  that  we 
must  go  more  deeply  into  the  conative  dispositions  which  deter 
mine  these  acts  of  presumption,  judgment,  and  assumption. 

2.  Genetic  Levels  of  Valuation. 

How  then  shall  we  conceive  this  relation  of  the  two  determi 
nants  of  feelings  of  value  ?  If  we  describe  the  acts  of  cogni 
tion  as  the  actual  psychical  presuppositions  and  the  conative 

faced  this  question  at  all,  as  the  following  passage  indicates  :  "  Where  the  capacity  or 
function  of  valuation  is  to  some  degree  realised,  there  the  individual  experiences  of 
feeling  and  desire  are  in  a  peculiar  manner  heightened  and  deepened,  they  have  a 
personal  character.  They  find,  so  to  speak,  in  the  personality  a  fuller  and  more  indi 
vidual  resonance.  We  can  in  such  a  case  speak  of  a  more  highly  developed  '  Gemiits- 
leben  ' "  (p.  50). 


52  Valuation:   its  Nature  and  Laws 

tendencies  as  the  dispositional  conditions,  our  problem  would 
read  :  What  is  the  relation  of  the  actual  presuppositions  to  the 
dispositional  conditions  as  determinants  of  feelings  of  value  ? 
The  answer  to  this  question  must  be  in  genetic  terms.  We  have 
already  seen  that  there  is  a  certain  genetic  relation  between  the 
attitudes  of  presumption,  assumption,  and  judgment.  Each,  in 
its  way,  represents  a  functional  attitude  toward  a  psychically 
pre-determined  object,  the  acceptance  of  a  demand,  acquiescence 
in  a  claim  to  control,  and  each  therefore  is  a  type  of  reality 
meaning.  But  these  demands  or  controls  vary  at  different 
stages  of  the  genetic  series.  An  analysis  of  the  manner  in 
which  the  dispositional  factor  functions  at  the  different  stages 
of  development  should  give  us  a  point  of  view  from  which  to 
unify  the  results  of  our  study. 

The  condition  which  determines  the  primitive  presumption  of 
reality  seems  to  be  that  the  object  shall  have  recognitive  mean 
ing  for  a  conative  tendency.  At  this  point  the  cognitive  and 
conative  moments  can  scarcely  be  distinguished.  As  far  back  as 
we  can  go  in  our  analysis,  interest  and  conation  seem  to  deter 
mine  recognition,  and  recognition  is  the  condition  of  the  first 
reality-meaning  which  characterises  feelings  of  value.  In  the 
primitive  presumption  of  reality  the  dualism  between  subjective 
and  objective  controlling  factors  has  not  yet  emerged.  It  is  with 
the  first  arrest  of  a  conative  tendency,  through  the  development 
of  an  independent  cognitive  interest,  and  through  differentiation 
of  the  recognitive  factor  from  the  conative,  that  the  innocency  of 
primitive  presumption  is  disturbed  and  a  differentiation  of  sub 
jective  and  objective  demands  or  controls  appears.  Here  the 
attitude  of  assumption  emerges,  determined  largely  by  the  sub 
jective  control  factor  of  the  conative  disposition,  often  in  oppo 
sition  to  objective  controls  already  established — but  not  neces 
sarily  so.  Assumption  of  the  existence  of  an  object  is  the 
acceptance  of  a  subjective  demand,  after  arrest  of  primitive 
presumption,  and  constitutes  a  transition  stage  between  pre 
sumption  and  explicit  acknowledgment  of  a  control  as  objective.1 
From  the  assumption  attitude  emerges  the  existential  judgment, 
either  positive  or  negative.  It  represents  not  merely  the  accept 
ance  or  rejection  of  an  object,  but  the  explicit  acknowledgment 
or  disavowal  of  a  certain  control  factor.  It  is  important  to  ob- 

1  I  am  inclined  to  agree  with  Professor  Baldwin  that  a  pure  fancy  mode  or  play  of 
fancy,  described  by  him  as  the  first  semblant  mode,  constitutes  a  genetic  transition 
between  presumption  and  assumption,  but  for  our  purposes  it  is  negligible. 


Definition  of  the  Consciousness  of  Value  53 

serve  that  this  control  factor  may  be  either  an  objective  or  sub 
jective  factor  and  that  the  existential  judgment  may  be 
acknowledgment  of  either.  But  in  the  latter  case  the  subjective 
has,  by  that  very  process,  been  transferred  to  the  objective  side 
of  the  equation. 


V.  RESUM^  OF  PRECEDING  DEFINITION  AND  ANALYSIS 

The  material  is  now  before  us  for  a  summary  restatement 
of  our  original  definition  of  value,  as  funded  affective- volitional 
meaning,  in  terms  of  psychological  equivalents.  The  psycho 
logical  equivalent  of  the  worth  predicate  is  always  a  feeling, 
with  certain  meanings  determined  by  actual  cognitive  presuppo 
sitions  or  types  of  aQgmtive_reacJtion  which  actualise  pre-existent 
conative  dispositions.  The  value  or  funded  meaning  of  the 
object  is  its  capacity  of  becoming  the  object  of  feeling  and 
desire  through  actualisation  of  dispositional  tendencies  by  acts 
of  presumption,  judgment,  and  assumption. 

The  conative  disposition  is  the  fundamental  determinant  of 
the  feeling  of  value  or  appreciative  meaning  of  the  object,  but 
the  disposition  may  be  actualised,  represented  in  function  by 
different  cognitive  attitudes  or  acts,  of  the  types  enumerated,  and 
according  as  it  is  one  or  the  other  of  these  types  is  the  feeling 
qualified  in  the  manner  described.1  Underlying  the  feeling 
of  value  attached  to  the  idea  of  my  friend  is  the  conative  disposi 
tion,  the  interest  created  by  former  desires  for  his  presence  and 
satisfaction  of  those  desires,  but  that  feeling  may  now  arise 
upon  mere  momentary  assumptions  of  his  existence  without  a 
trace  of  desire  for  his  immediate  presence.  All  "  disposition- 
feelings,"  however  actualised,  are  feelings  of  value  because  they 
represent  the  funded  meaning  of  affective- volitional  process, 

1  In  the  consideration  of  the  relation  of  the  actual  presuppositions  to  the  dispositional 
conditions  there  are  still  certain  questions  which  have  considerable  bearing  upon  later 
discussions.  Thus  Witasek  maintains,  that  while  it  is  probably  true  that  feelings  of 
worth  arise  upon  the  mere  presentation  of  an  object  related  to  desire  dispositions,  never 
theless,  since  desire  presupposes  judgment,  and  these  dispositions  have  been  formed  by 
preceding  judgments,  the  worth  feeling  is  ultimately  still  a  "judgment-feeling."  Now 
it  may  be  admitted  that  judgments  enter  into  the  formation  of  these  desire  dispositions, 
but  as  dispositional  they  are  merely  conative  tendency,  for  it  is  the  essence  of  judgment 
to  be  explicit  and  actual.  Again,  it  is  argued  (by  Saxinger),  that  the  dispositions  corre 
sponding  to  judgment  feelings  are  different  from  the  dispositions  correlated  with  assump 
tion  feelings,  and  he  bases  his  arguments  upon  differences  in  the  laws  governing  the  two 
kinds  of  feeling.  Into  the  consideration  of  this  question  we  cannot  enter  here — that 
will  be  reserved  for  later  study.  We  may  simply  emphasise  our  own  position  that 
worth  feeling  is  a  function  of  conative  disposition,  whether  conation  expresses  itself 
explicitly  in  judgment  or  assumption. 


54  Valuation  :    its  Nature  and  Laws 

although  they  have  different  reality  -  meanings.  From  the 
standpoint  of  the  extension  of  the  term,  the  class,  feelings  of 
value,  includes  aesthetic  feelings,  feelings  of  the  imagination 
so-called,  as  well  as  practical  and  ethical  attitudes. 

In  general,  then,  we  may  conclude  that  feeling  of  value  is 
the  feeling  aspect  of  conative  process,  as  distinguished  from 
the  feeling -tone  of  simple  presentations.  And  by  conative 
process  we  understand  the  total  process  of  development  by 
which  affective-volitional  meaning  is  acquired,  the  total  process 
including  actual  and  dispositional  moments.  How  these  dispo 
sitions,  and  with  them  the  feelings  which  they  condition,  are 
modified,  both  qualitatively  and  quantitatively,  at  different 
stages  of  this  development,  by  changes  in  presuppositions,  and 
more  especially  by  the  inclusion  of  secondary  judgments  of 
relation,  etc.,  is  the  problem  of  the  next  chapter. 


CHAPTER    III 

MODES   OF  THE   CONSCIOUSNESS   OF   VALUE- 
PRIMARY  AND  ACQUIRE 

I.  THE  APPRECIATIVE  DESCRIPTION  OF  FEELINGS  OF  VALUE — 
DESCRIPTION  OF  MODES  OF  REALITY  FEELING 

i.  The  Nature  of  Appreciative  Description. 

THE  course  of  the  preceding  chapter  has  led  to  a  demar 
cation  of  those  meanings  described  as  worths  or  values. 
Beginning  with  the  preliminary  definition  of  worth  as  the 
affective-volitional  meaning  of  the  object  for  the  subject, 
we  advanced  by  successive  stages  of  analysis  to  the  more 
specific  statement  that  worth  experience  is  always  a  feeling 
attitude  which  presupposes  the  actualisation  of  some  conative 
disposition  by  acts  of  presumption,  judgment,  or  assump 
tion  (implicit  and  explicit).  This  definition  obviously  involves 
a  certain  theory  of  the  nature  of  feeling  and  of  its  relation 
to  conation.  For  one  thing,  this  broader  use  of  the  term 
feeling  involves  a  relative  distinction  between  feeling-attitude 
and  affective  tone  of  sensation,  a  distinction  which  has  in 
fact  been  insisted  upon,  and  it  also  leads  to  the  view  that 
feeling,  as  worth  feeling,  has  appreciative  distinctions  not 
found  in  passive  affection.  To  this  theory  of  the  nature  of  feel 
ing,  and  to  the  more  abstract  psychological  analysis  which  it 
involves,  we  must  turn  our  attention  later.  For  the  present — and 
indeed  as  a  necessary  preliminary  to  this  later  study — our  problem 
is  the  further  development  of  the  appreciative  distinctions  of 
feeling. 

Earlier  in  our  study  a  distinction  was  made  between  the  "  ap 
preciative  "  and  "reality-"  (including  existence-)  meanings  of 
worth  predicates.  Starting  with  the  analysis  of  the  latter,  we 
developed  the  definition  of  value  in  terms  of  its  functional 
presuppositions.  But  in  the  course  of  that  very  analysis  we 

55 


5^  Valuation :   its  Nature  and  Laws 

came  upon  certain  appreciative  distinctions  in  feeling,  as,  for 
instance,  in  the  study  of  the  criteria  of  Lipps  and  Kriiger,  such 
as  feelings  of  the  personality,  breadth  and  depth  of  feeling  in  the 
personality,  which  were  taken  as  descriptive  of  feelings  of  value. 
Logically,  this  analysis  of  appreciative  descriptions  of  feeling 
should,  perhaps,  have  come  first  in  our  own  study,  but  the  order 
of  presentation  chosen  has  this  advantage,  that  the  critical  studies 
of  the  preceding  chapter  have,  by  their  results  both  positive  and 
negative,  defined  the  sphere  of  worth  experience,  and  have  given 
us  the  clue  to  the  interpretation  of  the  different  qualifications  of 
feeling  which  are  suggestive  of  worth,  that  is,  which  give  rise  to 
those  meanings  of  objects  which  we  call  worth  predicates. 

The  worth   predicates   themselves,    as   tertiary   qualities   of 
objects,  are,  in  their  manifold  modifications,  appreciative  dis 
tinctions  arising  from  differences  in  the  meaning  of  feelings. 
They   are    projections    into  the   object    of   distinctions   within 
feeling.      The    supposition    immediately    presents    itself    that, 
since    they   are   funded   meanings    of    feeling   processes,    they 
correspond  directly  to  fundamental  differences  in  feeling  itself, 
and  that  there  are  as  many  differences  in  feeling  as  there  are 
worth  predicates.   Reflection,  however,  makes  it  clear  that  appre 
ciative  description  of  objects,  while  the  expression  of  worth  feel 
ings,   is  not  necessarily  the   appreciative  description  of  those 
feelings  themselves.     These  predicates  are  what  we  feel  about 
the  object,  not  how  we  feel.     We  feel  beauty,  goodness,  nobility, 
sublimity,  obligation,  but  when  we  describe  how  we  feel  in  such 
cases  a  transition  has  been  made  to  the  appreciative  description 
of  the  feeling  itself.     The  feeling  has  been  made  the   object 
of  presentation  and  description,  and  it  is  quite  possible  that  in 
such  appreciative  description  of  the  feeling  one  of  these  general 
worth  predicates  may  stand  for  different  modifications  of  feeling, 
or  for  several  at  the  same  time.     Thus  the  predicate  good  may, 
when  applied  to  an  act,  have  as  its  equivalent  a  feeling  described 
as  the  tension  of  obligation,  at  another  the  feeling  of  satisfied 
repose.     In  order  to  describe  adequately  the  feeling  I  have  when 
I  call  an  object  sublime,  it  may  be  necessary  to  use  the  terms 
elevation  and  repose ;  and,  if  I  wish  to  add  to  my  description 
quantitative  terms,  to  speak  of  the  depth  of  the  feeling.     It  is 
apparent,  then,  that  what  is  meant  by  the  appreciative  distinc 
tions  in  primary  worth  feeling  are  those  descriptions  of  his  feelings 
which  the  subject  seeks  as  the  equivalents  of  his  worth  predicates 
applied  to  objects.     The  ultimate  terms  in  which  such  feelings 


•Modes  of  the  Consciousness  of  Value  57 

of  simple  appreciation  are  described  should  give  us  the  funda 
mental  modifications  of  worth  feeling. 

2.  The  Relation  of  Appreciative  Description  to  the  "  Scientific  " 
Description  of  Feeling  as  Content — Theories  of  Feeling. 

It  has  been  said  that  there  are  innumerable  nuances  of 
feeling,  and  in  the  same  breath  it  has  been  asserted  that  all  these 
differences  are  reducible  to  differences  in  intensity  and  duration 
of  a  one-dimensional  continuum,  pleasantness-unpleasantness, 
these  differences  being  due  to  differences  in  the  sensational, 
perceptual,  or  ideal  content  with  which  the  feeling  is  connected. 
With  the  first  part  of  this  statement  we  may  agree,  but  the  second 
requires  critical  examination.  The  consciousness  of  the  inade 
quacy  of  this  conception  of  the  dimensions  of  feeling  has  been 
growing  recently,  and  the  demand  for  new  analysis  has  arisen 
from  two  distinct  quarters — from  the  study  of  the  psychology 
of  worth  experience,  on  the  one  hand,  and  from  non-appre 
ciative  psycho-physical  analysis  as  illustrated  in  Wundt's  three- 
dimensional  theory,  on  the  other  hand. 

In  the  case  of  the  "  worth  psychologists,"  with  whom  we  are 
in  this  connection  primarily  concerned,  the  logic  of  this  anal 
ysis  is  clear  enough.  When  they  turn  from  the  worth  predicates 
of  objects  to  a  description  of  the  experiences  which  determine 
these  predicates,  they  find  the  old  terminology,  intensity  and 
duration  of  pleasantness-unpleasantness,  inadequate  for  the 
reconstruction  of  this  experience.  In  the  analysis  of  Kriiger 
which  we  have  already  considered,  worth  feeling,  which  is  dis 
tinguished  functionally  from  pleasure-causation  by  the  fact  that 
it  presupposes  conative  constants,  is  distinguished  appreciatively 
by  a  new  dimension,  depth  and  breadth  in  the  personality. 
Simmel,1  who  likewise  makes  feeling  the  fundamental  element, 
also  finds  it  necessary  to  distinguish  the  aspects  of  depth  and 
breadth  of  feeling  from  intensity.  Another  class  of  analysts, 
who  hold  a  voluntaristic  theory,  find  modifications  of  worth  ex 
perience,  which  cannot  be  correlated  with  feeling  if  feeling  be 
conceived  merely  as  intensity  of  pleasantness-unpleasantness. 
Brentano2  is  compelled  to  assume  quasi-logical  dimensions  of 
acts  of  preference,  to  which  pleasantness  and  unpleasantness 
are  related  merely  as  redundant  passive  phenomena  and  more 

1  Simmel,  Einkitung  in  die  Moralwissensfhaft. 

2  Brentano,  Psychologic.     Also  Ursprung  der  sittlichen  Erkentniss. 


5  8  Valuation :   its  Nature  and  Laws 

recently  and  definitely,  Schwartz1  has  found  it  necessary  to  dis 
tinguish  fundamentally  between  degrees  of  worth  experience, 
satisfaction  (Sattigung  des  Gefallen)  and  intensity  of  feeling, 
and  on  the  assumption  that  feeling  is  passive  pleasantness- 
unpleasantness,  to  seek  a  voluntaristic  basis  for  worth  experi 
ence.  Despite  the  differences  in  theory  of  the  nature  of  the 
worth-fundamental,  it  is  clear  that  these  analyses  all  have  in 
view  the  object  of  doing  justice  to  appreciative  distinctions  in 
worth  experience,  whatever  that  may  be  found  to  be,  in  terms  of 
psychological  equivalents. 

3.  The  Problem  and  Method  of  Appreciative  Description. 

If  then  we  hold  to  the  view  already  developed,  that  worth 
experience  is  feeling  with  certain  characteristic  presuppositions, 
our  task  is  naturally  to  seek  some  conception  of  feeling  which 
lies  between  the  two  views  propounded — both  of  them  unwork 
able  for  our  analysis — the  proposition  that  feeling  has  innumer 
able  modifications,  and  the  view  that  it  is  merely  intensity 
of  pleasantness  and  unpleasantness.  Now  the  key  to  our  pro 
cedure  is  to  be  found  in  the  fact  that  "  pleasantness-unpleasant 
ness  "  covers  but  one  class  of  terms  which  may  be  applied  to  the 
description  of  the  concrete  feeling  attitude,  that  there  are  other 
class  terms  which  are  equally  fundamental  for  the  communi 
cation  of  the  qualitative  differences  in  feeling.  As  a  descrip 
tion  of  the  subjective  experience  corresponding  to  the 
worth  predicate,  the  qualitative  differences,  pleasantness-un 
pleasantness,  are  insufficient.  Moreover,  when  this  has  become 
clear,  it  will  also  appear  that  in  order  to  express  quantitative 
differences  in  worth  feeling  it  will  be  necessary  to  make  use  of 
other  conceptions  than  that  of  intensity,  in  its  narrower  sense, 
which  has  been  transferred  from  sensation  to  the  pleasantness 
and  unpleasantness  which  accompanies  sensation. 

The  problem  then  is — what  are  the  fundamental  nuances 
of  feeling  corresponding  to  the  tertiary  qualities  or  worth 
predicates  attributed  to  objects  ?  The  answer  to  this 
question  would  naturally  take  the  form  of  a  classification 
of  the  appreciative  descriptions  of  feeling  attitudes  and  indeed 
a  desideratum  of  the  greatest  importance  in  the  present  situa 
tion  of  the  psychology  of  feeling  is  precisely  such  a  "natural" 
classification  of  the  appreciative  terms  used  in  the  first  stages 

1  Schwartz,  Psychologic  des  Willens,  chap.  II ;  also  Appendix  I. 


Modes  of  the  Consciousness  of  Value  59 

of  introspection.  As  I  have  elsewhere1  pointed  out,  the  psy 
chology  of  religious,  ethical,  and  aesthetic  feeling  must  build  its 
generalisations  almost  entirely  upon  the  results  of  such  introspec 
tions,  gathered,  for  instance,  by  the  questionnaire  method,  and  its 
possibility  rests  ultimately  upon  the  existence  of  uniformities  in 
such  descriptions.  Partial  contributions  to  such  a  classification 
already  exist — notably  in  the  sphere  of  religious  experience- 
but  in  default  of  any  adequate  view  of  the  whole  range  of  such 
descriptions,  and  in  view  of  the  impossibility  of  attempting 
such  a  classification  here,  we  may  resort  to  the  more  usual  and 
more  direct  method  of  analysing  our  experience  directly  for  the 
primary  fundamental  meanings  of  feeling,  and  then  seeking  to 
develop  the  secondary  derived  meanings  by  "genetic  progressions  " 
from  the  fundamental.  This  special  application  of  the  genetic 
method  of  analysis  will  have  the  advantage  of  presenting  our 
results  in  such  a  form  as  to  connect  them  immediately  with  the 
results  of  the  preceding  analysis  of  functional  presuppositions, 
and  the  two  will  act  as  mutually  supplementary  and  corrective. 


II.  THE  FUNDAMENTAL  APPRECIATIVE  DISTINCTIONS  IN 
FEELING — QUALITATIVE  AND  QUANTITATIVE 

i.  Qualitative  :  Directions  and  References — The  Three- 
dimensional  Theory. 

What,  then,  are  the  primary,  irreducible  aspects  of  feelings 
which  must  be- distinguished  in  order  to  fix  their  place  in  a 
system  of  meanings  ?  As  has  been  suggested,  these  aspects 
must  be  expressed  in  terms  both  of  quality  and  degree.  Our 
first  concern  is  therefore  with  the  qualities  of  feeling.  Every 
concrete  feeling  attitude  has  two  primary  aspects  or  mean 
ings,  its  direction  and  its  reference.  Its  direction  is  either 
positive  or  negative.  Its  reference  is  either  transgredient 
or  immanental.  Of  the  first  aspect  little  need  be  said. 
It  is  that  fundamental  duality  of  quality  which,  when  feel 
ing  is  viewed  retrospectively  as  passive  and  as  abstracted 
from  conation,  is  described  as  pleasantness-unpleasantness.  As 
direction  or  meaning  of  feeling  attitude,  however,  it  presupposes 
relation  of  the  attitude  to  conation.  What  have  been  described 

1  "Appreciation  and   Description  and   the   Psychology   of   Values,"   Phil.   Rev., 
November,  1905. 


6°  Valuation :   its  Nature  and  Laws 

as  the  references  of  feeling  specify  more  completely,  on  the  other 
hand,  this  relation  to  conation  ;  they  are  aspects  of  the  feeling 
which  refer  to  something  presupposed,  to  a  disposition  already 
acquired  for  which  the  object  has  a  meaning.  In  the  case  of 
the  transgredient  reference  it  is  the  sense  of  a  subjective  control 
leading  on  to  other  states.  In  the  case  of  the  immanental,  it  is  a 
sense  of  a  more  objective  control  leading  to  continuance  or 
repose  in  the  same  state.  When  it  comes  to  describing  these 
directions  and  references,  their  different  nuances  and  suggestions, 
use  is  made  of  metaphorical  and  analogical  terms,  the  significance 
of  which  we  must  consider. 

The  simplest  analogy  here  made  use  of  is  that  of  contrast  pairs 
from  the  different  sense  regions.     Feelings  are  described  as  sweet 
or  bitter,  bright  or  dull,  soft  or  hard,  etc.     They  specify  for  finer 
discrimination  and  description  the  two  fundamental  directions  of 
feeling,  the  positive  and  the  negative,  pleasant  and  unpleasant. 
Of  these  it  is  not  necessary  to  speak  in  detail,  for  while  they  help 
to  describe  the  worth  or  affective-volitional  meaning  of  objects 
for  the  subject,  their  external  and  analogical  origin  makes  them 
only  indirectly  the  means  of  communication  of  worth  experiences. 
A  more  important  group  of  terms  employed  in  differentiating 
the  worth  suggestions  of  feeling-attitudes  are  those  which  may  be 
described  as  dynamic.     They  describe  the  dynamic  suggestions 
of  the  feeling,  specify  the  transgredient  reference.     This  trans 
gredient  reference  is  ordinarily  described  metaphorically  in  terms 
of  forms  of  movement  from  the  external  world.     Of  the  large 
number  of  such   forms   made   use   of   in   these  descriptions  a 
slight   study  of  the  literature  of  such  appreciative  introspec 
tion   makes   us   immediately  aware.      It  is   full   of   terms   for 
different  nuances  of  movements  of  the  crescendo  or  diminuendo 
type— of  soaring,  uplifting,  of  sudden   breaking  in  upon  con 
sciousness,   of   dying   away,    of   height   and   depth,  etc.     They 
can  probably  all  be  included  under  the  general  terms  tension, 
restlessness,    and    perhaps    contraction,    the    nature    of    which 
dimensions,   and    the    theory  connected   with    their    classifica 
tion,  we   shall  consider  presently.     From  the  point  of  view  of 
content  such  forms  of  movement  are  probably  complexes  founded 
on  relations  of   intensity  and   duration   among   more  ultimate 
elements.     However  that  may  be,  the  characteristic  of  these 
symbolic,  dynamic  descriptions  is  that  they  describe  transitional 
aspects  of  experience,  transitions  from  one  aspect  of  content  to 
another  by  which  meaning  is  acquired.     By  this  I  mean  that  in 


Modes  of  the  Consciousness  of  Value  61 

the  present  feeling  there  is  always  a  transgredient  reference  to  a 
past  or  future  attitude.  The  present  experience  is  always  the 
foreground  of  a  background,  past  or  future,  which  is  still,  or 
already,  dimly  felt.  Of  course  in  such  a  feeling  there  is  always 
reference  to  conation,  and  it  might  be  objected  that  we  are  here 
dealing  with  impulse  and  desire  rather  than  with  feeling,  if  it  were 
not  that,  as  we  shall  seek  to  show,  feeling  cannot  be  completely 
abstracted  from  conation. 

A  third,  and  qualitatively  opposite,  class  of  terms  is  used 
to  characterise  appreciatively  the  nuances  of  immanental  refer 
ence  of  feeling.  They  may  all  be  grouped,  I  think,  under  the 
general  terms,  repose,  relaxation,  and  expansion.  Feelings  of  ex 
pansion  have  an  unusual  wealth  of  descriptive  terms  at  their 
service.  Favourite  descriptions  are  in  terms  of  pervasion  and 
possession.  The  subject  of  the  emotion  describes  himself  as  per 
vaded — as  by  an  ether,  a  fluid — as  swallowed  up  by  the  emotion, 
and  in  the  mystical  amorous  and  religious  literature,  of  which  such 
descriptions  are  typical,  it  is  with  love,  with  the  glory  or  the  will 
of  God,  that  the  subject  is  filled.  These  suggestions  or  meanings 
of  feelings  are  likewise  probably  aspects  or  qualities  founded  on 
more  elementary  content. 

This  immanental  reference  of  repose,  with  its  cognate  expan 
sion  of  feeling,  is  a  meaning  wrhich  the  feeling  gets  when  the  con- 
ative  tendency  or  disposition,  presupposed,  has  reached  the  stage 
of  habit  after  accommodation.  The  object  of  the  feeling  occupies 
the  whole  consciousness,  but  into  the  meaning  of  the  object  is 
taken  up  all  the  accumulated  meaning  of  the  processes  of  accom 
modation  for  which  the  disposition  now  stands.  The  reference 
of  the  feeling  is  not  beyond  the  present  state,  but  to  something 
more  deeply  involved  in  it. 

In  the  case  of  the  term  expansion  (and  contraction  its  corre 
lative  transgredient  term) ,  it  is  obvious  that  such  descriptions  are 
metaphorical  transferences  from  the  spatial  world  of  perception, 
but  I  think  it  can  scarcely  be  denied  that,  as  appreciative  de 
scriptions,  they  are  as  fundamental  as  the  other  descriptions 
transferred  from  the  experiences  of  intensity  and  duration.  It 
has  been  objected  to  the  three-dimensional  theory  of  feeling 
that,  if  the  analogical  terms,  tension-relaxation,  restlessness- 
quiescence  are  introduced,  there  is  no  reason  why  the  terms 
contraction-expansion  should  not  also  be  applied.  There  is  none 
in  fact — the  only  question  is  whether  they  are  equally  irreducible 
as  terms  of  appreciative  introspection.  With  an  introspection 


Valuation :   its  Nature  and  Laws 

which   is  not   appreciative    we    have    in    this    connection    no 
concern. 

That  contraction  and  expansion  are  in  this  sense  fundamental 
aspects  of  feeling  there  can  I  think  be  no  question.     And  in  this 
connection  it  is  interesting  to  note  the  fact  that,  in  a  recent  study 
ling  by  experimental  methods,  without  these  appreciative 
itmctions,  it  was  found  impossible  to  distinguish  the  feeling- 
tone  of  simple   sensation   from  a  mood  or  disposition-feeling 
former  attaches,   so  to  speak,  to  the  stimulus-complex 
(taste)  while  the  latter  spreads  over  the  whole  consciousness  " 
:  was  further  found  that  they  have  different  pneumographic 
expressions.     The  former  is  attended  by  quickening,  the  latter 
by  slowing  of  respiration.1 

(a)  The  Three-dimensional  Analysis  a  Description  of  Reality- 
Meanings  of  Feelings  of  Value-not  of  Simple  Feeling 
Abstracted  from  Cognitive  Presuppositions. 

The  relation  of  this  analysis  to  the  so-called  three-dimen 
sional  theory  of  feeling  developed  by  Wundt  may  be  stated  as 
)llows.  We  accept  the  analysis ;  but  for  us  the  terms  of  this  theory 
are  descriptive  equivalents  for  appreciative  meanings  of  total 
J  attitudes,  while  for  Wundt  they  are  qualities  of  simple 
The   difference   arises  necessarily  from   the   different 
:s  of  view  from  which  the  description  of  the  same  experience 
s  approached.     The  appreciative  descriptions  try  to  fixate  the 
meaning  of  the  conative  references   (transgredient  and  imma- 
nental)  implicit  in  the  feeling  attitude,  i.e.,  references  to  preceding 
and  succeeding  conation.     The  analysis  of  Wundt,  on  the  other 
hand,  seeks  to  fixate  the  same  experience  by  terms  from  which 
the  worth  connotation  is  more  completely  abstracted,  and  in  which 
the  implicit  reference  to  the  self  is  ignored.    Royce,  it  should  how 
ever  be  noted  in  passing,  finds  the  "interest"  in  the  hypothesis 
m  the  "  statement  it  makes  possible  of  the  relation  of  feeling 
and  conduct,  not  adequately  conceived  on  the  one-dimensional 
theory  "  -a  clear  recognition  of  the  fact  that  he  is  concerned  with 
appreciative  description. 

That  the  three-dimensional  theory  constitutes  a  true  descrip- 

lon  of  total  feeling  attitudes  is  scarcely  open  to  dispute.     The 

slightest   appreciative   introspection   enables   us   to   distinguish 

on  Gefuhl,"  Arckiv  fur  die 


Modes  of  the  Consciousness  of  Value  63 

between  the  exciting  pleasure  of  hope  and  the  tranquil  pleasure 
of  peace,  between  the  painful  tension  of  dread  and  the  equally 
painful  relaxation  of  despair.  The  question  at  issue  is  not,  then, 
whether  these  differences  are  appreciable  among  total  feeling  atti 
tudes  and  thus  constitute  worth  suggestions,  but  rather  whether 
they  are  equally  characteristic  of  sensation-feelings.  On  this 
question  there  is  no  conclusive  answer  to  be  given  at  the  present 
time.  Wundt  has  brought  forward  experimental  evidence  in 
favour  of  the  view  that  these  additional  qualities  belong  also  to 
simple  sensation-feelings  (the  feeling  tone  of  colours  and  sounds, 
for  instance).  As  to  the  value  of  the  evidence,  there  is,  of  course, 
still  doubt ;  some  experimenters  do  not  find  the  modifications 
of  the  curves  corresponding  to  the  three-dimensional  analysis. 
But  even  if  there  were  no  question  in  regard  to  the  facts  them 
selves,  the  meaning  of  these  facts  would  not  be  unequivocal. 
We  cannot,  for  one  thing,  be  sure  that  while  the  stimuli  are  so- 
called  simple  sensations,  the  feeling  reactions  are  simple  feel 
ings.  They  may  be  —  and,  indeed,  probably  are  —  on  the 
emotional  level,  the  organic  and  muscular  sensations  due  to  the 
surplus  excitation.  It  is  certainly  true  that  the  results  are  most 
apparent,  both  in  the  graphic  registration  and  in  introspection, 
as  reference  to  Wundt's  studies  will  show,  in  those  cases  where 
the  reactions  are  on  the  emotional  level.  Besides,  as  has 
already  been  pointed  out,  although  the  feeling  tone  of  sensa 
tion  is  itself  not  worth-suggestive  or  on  the  level  of  worth  feeling, 
nevertheless,  when  the  stimulus  has  reached  a  certain  intensity 
or  duration,  it  gives  rise  to  a  feeling  attitude  which  is  worth- 
suggestive.  Until  the  experimental  evidence  is  more  un 
equivocal,  both  introspection  and  logic  would  rather  lead  to  the 
view  that  these  dimensions  of  feeling  which  seem  to  belong  to 
simple  feeling-tone  of  sensation  are  really  qualities  of  a  secondary 
feeling  attitude  following  upon  pleasure-causation.  Storring's 
analysis,  already  referred  to,  would  seem  to  indicate  the  truth 
of  this  view. a 

1  Recent  criticisms  of  the  three-dimensional  theory  have  been  entirely  justified  in 
saying,  on  the  one  hand,  that  these  qualifications  of  feeling  are  taken  from  the  side 
of  conative  meaning,  and  on  the  other  that  when  we  look  for  equivalents  in 
content,  we  find  them  only  in  sensations,  kinsesthetic  and  organic.  Both  state 
ments  are  true  and  at  the  same  time  consistent  with  each  other,  as  will  appear 
in  our  later  studies  of  feeling.  It  is  only  in  the  appreciatively  described  total 
meaning  of  the  attitude  that  they  appear  as  primary  qualities  of  experience.  When  we 
take  the  abstract  point  of  view  of  function  they  break  up  into  relations  of  affirmation  and 
arrest  of  tendency.  When  we  take  the  abstract  point  of  view  of  content  or  structure,  they 
break  up  into  complexes  or  series  of  sensations.  The  way  of  reconciling  structural  and 
functional  points  of  view  in  psychology  is  to  correlate  them  both  with  the  appreciative 
description  from  which  both  take  their  origin. 


64  Valuation :   its  Nature  and  Laws 

(b)  Worth  Feelings  are  on  the  Emotional  Level. 

Be  this  as  it  may,  I  think  it  may,  nevertheless,  at  least  be 
said  that  these  aspects  of  experience,  whether  that  experience 
be  a  hypothetical  affective  or  sensational  content,  become  worth- 
suggestive,  acquire  the  transgredient  and  immanental  references 
only  on  the  emotional  level,  only  when  the  feeling  is  a  feeling 
attitude  toward  an  object.  And  I  think  it  may  further  be  said 
that  the  criterion  of  such  a  feeling  attitude,  or  emotion  (the  term 
emotion  being  used  in  its  broadest  sense  to  include  passion, 
sentiment,  and  mood,  as  well  as  emotion  proper),  is  the  presence 
of  the  cognitive  presuppositions  already  analysed,  presumption, 
judgment,  and  assumption.  What  is  meant  by  this,  to  state  the 
point  more  fully,  is  that  the  differences  in  feeling-attitude  appre 
ciatively  distinguishable  appear  only  in  total  feeling-attitudes,  and 
are  not  varieties  of  the  mere  feeling-tone  of  sensations.  It  may 
be  that  the  content  which  acquires  these  meanings  is  certain 
simple  affective  or  sensational  elements,  but  it  acquires  these 
meanings  only  on  the  level  of  emotion. 

The  view  here  developed  involves,  further,  that  the  cri 
terion  of  an  emotion  or  feeling- attitude  is  to  be  found  in  the 
presence  of  a  cognitive  act  (presumption,  judgment,  assump 
tion)  as  the  presupposition  of  the  feeling.  Can  this  view 
be  maintained  ?  I  think  it  may  not  only  be  reasonably  main 
tained,  but  is,  in  fact,  inevitable,  if  we  approach  the  study  of 
emotions  as  above  defined  from  the  standpoint  of  their 
meaning.  There  is,  to  be  sure,  another  point  of  view,  that 
of  more  abstract  study  of  content  and  emotional  expression, 
from  which  this  scarcely  seems  to  be  the  proper  criterion,  as 
for  instance,  in  the  case  of  the  inherited  instinctive  emotions, 
of  which  the  instinctive  fear  of  animals  is  a  good  illustration. 
But  while  this  is  true — and  with  this  view  of  the  facts  our  pre 
sent  analysis  must,  in  its  proper  place,  be  brought  into  harmony, 
it  is,  nevertheless,  also  true  that,  as  a  meaning,  an  emotional  atti 
tude  always  presupposes  such  cognitive  acts.  Joy  and  sorrow, 
the  two  typical  and  fundamental  emotional  attitudes  which  have 
these  worth  suggestions  or  meanings,  become  meaningless  when 
conceived  apart  from  these  presuppositions.  They  are  usually 
judgment-feelings,  although  not  always  such  (as  Meinong  main 
tains),  for  they  may  follow  upon  simple  presumption  or  assump 
tion  of  reality.  The  joy  in  the  presumed,  assumed,  or  asserted 
reality  of  an  object  is  toto  genere  different  from  the  pleasantness 


Modes  of  the  Consciousness  of  Value  65 

of  a  sensation.  And  the  same  is  true  of  those  modes  of  emotional 
attitude,  such  as  fear,  dread,  despair,  hope,  elation,  in  which 
the  cognitive  act  is  further  modified  in  the  direction  of  possibility 
or  necessity.  It  is  further  to  be  observed  that,  from  the  point 
of  view  of  appreciative  analysis,  these  emotional  attitudes  are 
variously  specified  according  as  the  fundamental  positive  or 
negative  direction  has  transgredient  reference  with  its  tension 
or  restlessness,  or  immanental  reference  with  its  relaxation  and 
repose.  Joy  or  sorrow,  as  we  have  seen,  may  be  of  either  type. 
The  inevitable  conclusion  seems  to  be  that  these  meanings  arise 
only  when  there  is  that  totalisation  of  attitude,  the  condition  of 
which  is  the  actualisation  of  conative  dispositions  through  acts 
of  the  type  described.1 

(c)  Objectless  Feelings  not  an  Exception. 

There  are,  however,  certain  phenomena  which  constitute  an 
apparent  exception  to  this  law,  namely,  objectless  feelings  (emo 
tions,  sentiments,  and  moods),  which,  although  objectless,  are 
clearly  worth  suggestive  and  find  expression  in  worth  judgments. 
Practically  all  the  concrete  emotional  attitudes — joy,  sadness, 
anger,  fear — may  appear  as  worth  feelings  without  concrete  per 
ceptual  or  ideal  objects.  A  nameless  sadness  or  fear,  an  objectless 
anger,  may  arise  in  consciousness  with  all  the  worth  suggestions 
of  enhanced  or  thwarted  conation,  but  without  any  object  towards 
which  it  is  definitely  directed.  This  does  not  mean  that  there 
are  no  adequate  conditions  (physiological,  and  even  psycho 
logical),  but  merely  that  there  is  no  presupposition,  no  judg 
mental  reference  to  the  existence  or  non-existence  of  objects. 
They  appear  at  first  sight  to  be  without  such  presuppo 
sitions.  In  reality,  however,  they  are  to  be  viewed  as  in  the 
main  analogous  to  exclamations  and  some  forms  of  impersonal 

1  Wundt  (and,  it  may  be  added,  Hoffding  also)  makes  much  of  the  principle  of 
totalisation,  of  total  resultant,  in  his  analysis  and  theory  of  feeling.  Whatever  be 
the  nature  of  the  simple  feelings,  they  all  tend  to  merge  in  a  total  resultant,  a  unitary 
feeling.  This  principle  of  "  Einheit  der  Gefiihlslage"  is  referred  to  the  principle  of 
unity  of  apperception  for  its  explanation,  all  feeling  being  viewed  as  the  subjective 
aspect  of  apperception.  The  truth  of  this  general  proposition  is  beyond  question,  but 
there  are  different  grades  of  apperception  and  different  degrees  of  totalisation.  Un 
doubtedly  when  attention  is  held  by  a  sensation  of  sound  or  colour,  or  by  an  organic 
sensation,  its  feeling-tone  tends  to  dominate  consciousness  and  to  fuse  with  it  all  other 
feeling-tones.  But  it  is  not  until  there  is  explicit  reference  of  the  sensation,  as  object, 
to  a  conative  disposition  through  judgment  or  assumption,  that  the  totalisation  of  atti 
tude  takes  place  which  gives  rise  to  the  worth  suggestions  of  feeling.  In  such  a  totalisa 
tion  the  feeling-tone  of  sensations,  as  such,  becomes  irrelevant  and  subordinate  to  the 
worth  feelings  of  the  attitude  as  a  whole. 


66 


Valuation :   its  Nature  and  Laws 


judgment  in  the  sphere  of  cognition.  As  in  these  cases  there  is 
no  directly  asserted  subject  of  the  predicate  discoverable,  so  in 
objectless  emotions  and  moods  there  is  no  directly  asserted 
object  of  judgment  to  which  the  worth  predicates  implied  in 
the  feelings  of  joy,  sorrow,  etc.,  are  applied.  Reality  is  implied 
-the  feelings  are  real  and  "earnest,"  but  there  is  no  existential 
judgment  about  any  definite  object  in  reality.  There  is  merely 
an  undifferentiated  presumption  or  assumption  of  reality  as  pre 
supposition  ;  but  this  is  sufficient  to  make  them  worth  feelings. 

The  psychology  of  the  impersonal  judgment  scarcely  leaves 
us  room  to  doubt  of  its  nature.     There  is  for  such  judgment 
neither  subject  nor  predicate,  nor  reference  of  the  one  to  the 
other.     It  is,  so  to  speak,  the  amorphous,  protoplasmic  germ  of 
later  reflective  judgments  which  do  involve  a  separation  of  subject 
and  predicate.   Whatever,  in  the  interests  of  systematic  logic,  we 
may  seek  to  supply  as  the  subject  of  such  judgment  in  order 
to   bring   it   within   the    classifications   of    a   logical  system— 
whether  we   describe   the    subject    as    universal  and   undeter 
mined,  the  whole  of  reality,  or  as  a  determined  and  particular 
sensation^  of^  the  moment,  the   fact  remains  that  psychologic 
ally  the  "  it "  of  the  impersonal  judgment  is  contentless.     Simil 
arly,  in  the  objectless  worth  feeling  the  object  is  no  presentation 
either  universal  or  particular,  no  sensation  either  peripheral  or 
organic.     Subject  and  predicate,  presentation  and  feeling,  are 
not  discriminated.     We  have  to  do  here  with  a  protoplasmic 
worth   attitude  without  judgmental  presuppositions  but  which 
may,  nevertheless,  become  definite  through  the  inclusion  among 
its  presuppositions,  which  are  now  merely   conative   and   dis- 
positional,  of  some  explicit  act  of  judgment. 

3.  Correlation  of  Appreciative  Meanings  with  Cognitive 
Presuppositions. 

Can  we,  then,  correlate  the  meanings  of  worth  feelings, 
thus  described,  with  specific  types  of  cognitive  presup 
positions  ?  The  necessary  presupposition  of  worth  feeling 
is,  as  we  have  seen,  the  actualisation  of  a  conative  disposition 
through  acts  of  presumption,  assumption,  and  judgment.  Can 
we  connect  the  specific  type  of  reference  of  the  feeling  with  a 
definite  type  of  actual  presupposition  ? 

The  two  directions  of  worth  feeling  (positive  and  negative),  as 
distinguished  from  mere  pleasantness-unpleasantness,  contain 


Modes  of  the  Consciousness  of  Value  67 

some  presupposition  of  reality— witness  our  study  of  joy  and 
sorrow,  love  and  anger,  hope  and  despair.    And  as  we  shall  see 
later,  positive  and  negative  worth  may  vary  independently  of 
pleasantness-unpleasantness.     But  it  is  with  the  other  qualifica 
tion  of  feeling,  the  reference  to  conation,  that  we  are  chiefly  con- 
When  we  turn  to  the  transgredient  reference,  with  its 
tension,  restlessness,  contraction,  and  to  immanental  reference 
with  its  relaxation,  repose,  and  expansion,  we  find  that  they  are 
closely  connected  with  changes  in  the  presupposition  of  reality. 
In  general  the  transgredient  reference  appears  in  all  those 
emotional  attitudes  where  an  habitual  presupposition  of  reality 
5  with  opposition  or  arrest,  where,  for  instance,  primitive 
presumption  passes  into  assumption  and  judgment.     In  such  a 
case  it  may  be,  either  the  subjective  control  factor— the  conative 
imposition  which  is  felt  in  the  background  and  which  gives  rise 
to  the  assumption— or  the  more  objective  factor  of  control,  the 
recognitive,  determining  and  giving  rise  to  judgment.     In  either 
ase,  however,  the  transgredient  reference  is  to  a  disposition  in 
the  background,  in  the  process  of  determining  a  new  accom 
modation. 

The   immanental   reference   to   reality,    on   the   other   hand 
represents  the  emotional  attitude  which  goes  with  accommoda 
tion   realised.     It   is   the   feeling  which  attaches  to  judgment- 
t  or  to  the  assumption  of  the  second  type  arising  out  of  that 
The  fact  that  habit  has  its  own  feeling,  its  own  worth 
suggestions,  is  a  point  which  must  be  emphasised  throughout. 


III.  MEANINGS  ACQUIRED  BY  DEVELOPMENT  OF  THESE 
FUNDAMENTAL  MODES— VALUE-MOVEMENT 

i.    Acquired   Meanings   of   Simple   Appreciation. 

With  the  analysis  of  these  primary  aspects  or  meanings 
which  feelings  disclose,  we  are  led  to  the  problem  of  derived  or 
acquired  feeling  attitudes. 

There    are    two    possible    conceptions     of    the    nature    of 
these   attitudes   and  of   the   process   of   their  derivation.     The 

:  of  these  is  the  concept  of  fusion   or  mixture  of  feelings 
purely    analytical    in    character.      On    this    view    the    aspects 

feeling,    the    meanings    of    appreciative     description      are 
hypostatised    as    elements,     and    all    acquired    meanings    are 


68  Valuation:    its  Nature  and  Laws 

conceived  as  fusions  or  mixtures  of  these  elements.  The 
second  concept,  genetic  and  functional  in  character,  looks 
upon  the  derived  attitude,  the  acquired  meaning,  as  a  new 
aspect,  the  product  of  a  new  "  totalisation  "  of  consciousness  in 
which  the  old  aspects  are  taken  up  into  the  new,  but  in  which 
the  new  meaning  is  not  exhausted  by  its  analysis  into  the  old 
elements.  The  new  feeling-attitude  is  a  new  accommodation, 
a  development,  in  terms  of  worth  theory,  a  value-movement. 

The  former  of  these  views,  of  very  limited  applicability  at 
the  best  in  any  region  of  psychological  explanation,  is  wholly 
inapplicable  to  the  explanation  of  the  meanings  of  feeling- 
attitudes.  Wundt  is,  unfortunately,  despite  his  three-dimensional 
theory,  still  too  much  under  the  influence  of  this  conception, 
although  in  applying  his  fundamental  law  of  psychical  causality, 
the  law  of  resultants,  he  explicitly  asserts  that  there  is  an  ac 
quired  meaning  in  the  resultant  complexes  or  fusions  not  found 
in  the  elements.  It  is  better  to  abandon  the  concept  of  elements 
entirely  in  this  connection  and  to  make  use  wholly  of  the  genetic 
concept  of  acquirement  of  meaning  through  change  in  pre 
suppositions. 

The  acquired  meanings  of  feelings  may  be  divided  into  two 
groups  :  (i)  the  acquired  meanings  of  simple  appreciation,  and 
(2)  those  of  characterisation  and  participation.  If  we  recall 
these  distinctions,  previously  made,  it  will  be  remembered  that 
simple  appreciation  of  an  object  is  an  appreciation  of  its  affective- 
volitional  meaning  or  worth  prior  to  explicit  reference  of  the 
object  to  the  ego  or  the  alter  or  to  other  objects,  prior,  in  other 
words,  to  secondary  possessive  or  instrumental  judgments.  On 
the  level  of  simple  appreciation  appear,  then,  certain  qualifica 
tions  of  the  general  transgredient  and  immanental  references  of 
feeling. 

(a)  The  Impellent  Mode-Feelings  of  Obligation. 

The  first  of  these  acquired  meanings  to  be  considered  is  the 
feeling  of  oughtness  or  obligation.  The  feeling  of  oughtness  that 
a  thing  should  be,  that  an  act  should  take  place,  is  a  specific  form 
of  the  feeling  of  value.  As  such,  upon  our  view,  it  should  be 
defined  in  terms  of  its  presuppositions.  Appreciatively  de 
scribed,  it  is  an  acquired  modification  of  the  general  feeling  of 
transgredient  reference  or  of  tension,  and  may  be  best  described 
as  the  Impellent  Mode.  Apart  from  appreciative  description 
it  is  an  experience  of  mere  strain — perhaps,  from  the  point  of 


Modes  of  the  Consciousness  of  Value  69 

view  of  content,  a  mere  strain  sensation.  Its  differentia  is  to 
be  found  in  the  precise  character  of  the  transgredient  reference, 
and  therefore  in  the  character  of  its  cognitive  presuppositions. 
Now  the  feeling  of  oughtness,  in  its  simplest  form,  attaches  to 
objects,  to  things.  It  is  felt  that  if  a  thing  does  not  exist  it 
ought  to.  As  thus  applied,  for  instance,  by  a  child  who  as  yet 
has  practically  no  sense  of  personal,  ethical  obligation,  it  means 
little  more  than  that  the  thing  is  desired.  But  just  that  little 
additional  meaning  is  the  important  modification.  Is  it  possible 
to  define  this  additional  meaning  ? 

The  point  of  difference  is  to  be  found,  I  think,  in  the  fact  that 
the  presuppositions  of  the  feeling  of  oughtness  are  not  simple, 
as  in  the  case  of  a  simple  mode  of  feeling  or  desire.  The  feeling 
of  oughtness  is,  in  fact,  a  transition  mode  between  two  existential 
judgments,  in  which  an  existential  feeling  is  qualified  by  an 
assumption- feeling.  The  object  does  not  exist,  and  we  have 
the  corresponding  feeling  or  desire,  but  so  strong  is  the  conative 
disposition  presupposed  that  it  gives  rise  to  an  assumption  of 
existence.  This  assumption  is  felt  to  be  not  merely  possible, 
but  necessary,  and  thus,  as  Simmel  has  said,  obligation  is  in  one 
aspect  a  mode  of  thought  lying  midway  between  possibility  and 
necessity.1  The  source  of  this  assumption  is  the  subject's 
conative  disposition  and  the  feeling  of  oughtness  is  the  feeling  of 
that  subjective  control ;  but  since  the  subjective  control  is  not 
explicitly  acknowledged  in  judgment,  the  oughtness  is  felt  as 
a  tertiary  quality  of  the  object. 

The  transgredient  reference  of  the  assumption  is  therefore  to 
the  disposition.  To  refer  again  to  the  figure  of  the  foreground 
and  background  of  consciousness,  the  judgment  of  existence  or 
non-existence  of  the  object  is  in  the  foreground,  the  modification 
of  the  feeling  which  we  describe  as  oughtness  having  reference  to 
an  object  in  the  background  which  at  first  is  revealed  merely  in 
this  modification  of  feeling,  but  which  later,  through  the  ac 
tivities  of  ideal  construction  and  judgment,  becomes  an  explicit 
ideal  object,  the  self  or  the  social  will.  When  this  is  developed 
ethical  obligation  is  felt.  In  a  sense  the  simple  feeling  of  ought- 
ness  is  objectless  until  this  stage  of  ideal  construction  is  reached, 
and  corresponds  to  the  impersonal  judgment. 

1  Simmel's  masterly  study  of  the  modes  of  oughtness,  das  Sollen  (Einkitting  in  die 
Moralwis  sense  haft,  chap,  i)  can  be  merely  referred  to  in  passing,  fuller  treatment  being 
reserved  for  another  connection.  The  important  point  is  that  it  is  a  fundamental  mode, 
at  the  same  time  cognitive  and  affective-volitional. 


7O  Valuation  :   its  Nature  and  Laws 

(b)  The  Semblant  Mode-JEsthetic  Feeling. 

Corresponding  to  the  feeling  mode  of  oughtness,  the  primary 
impellent  mode  out  of  which  ethical  obligation  develops,  we  find 
a  second  mode  of  simple  appreciation  which  represents  a  special 
qualification  of  the  immanental  reference  of  feeling,  the  "  sem- 
blant  "  or  aesthetic  mode.1  This  mode,  the  aesthetic  feeling,  is 
always  appreciatively  described  in  terms  of  repose  and  expansion, 
and  its  worth,  in  so  far  as  the  experience  is  purely  aesthetic,  is  im 
manental.  Here  again  we  have,  not  a  simple  aspect  of  feeling 
with  simple  presuppositions,  but  an  attitude  implying  transition 
and  accommodation,  and  therefore  characterised  by  typical 
changes  in  cognitive  presuppositions. 

The  characteristics  of  this  mode  of  feeling,  its  repose,  relax 
ation,  and  expansion,  have  their  origin  in  the  fact  that  the  judg 
ments  of  existence  and  non-existence,  and  with  them  explicit 
conation,  desire,  are  inhibited,  reduced  to  a  minimum,  and  remain, 
in  fact,  merely  as  a  dispositional  presupposition,  while  conscious 
ness  is  largely  absorbed  in  presentational  content.  With  the  laws 
which  govern  the  ordering  of  that  content,  and  which  condition 
the  arrest  of  desire  and  the  inducing  of  repose,  we  are  not  at 
this  point  concerned  ;  it  is  sufficient  to  note  the  general  fact  that 
formal  principles  of  aesthetics  owe  their  significance  psycho 
logically  to  the  fact  that  they  are  instrumental  in  producing  this 
effect.  But  as  has  already  been  pointed  out,  it  is  not  an  adequate 
view  of  aesthetic  feeling  to  regard  it  as  a  purely  presentational 
consciousness.  While  explicit  judgment  is  reduced  to  a  minimum, 
its  place  is  taken  by  assumptions  which  relate  the  object  to  a 
desire  which  is  now  merely  dispositional.  These  assumptions, 
we  have  seen,  may  be  of  two  types,  the  explicit  and  the  implicit. 
In  the  first  case  we  have  the  primitive  semblant  mode,  in  the 
latter  the  more  developed  mode  of  contemplation. 

In  general,  then,  the  aesthetic  mode  of  sembling  or  contempla 
tion  is  a  complex,  derived  mode  of  feeling  of  value  in  which  the 
presuppositions  are  presentational  content  and  assumptions. 
To  use  again  the  figure  of  the  foreground  and  background  of 
consciousness,  the  foreground  is  taken  up  with  presentational 
content,  the  psychical  energies  involved  in  judgment  are  occupied 

1  For  the  use  of  the  term  "semblant  mode,"  see  Baldwin's  Thought  and  Things ; 
Vol.  I,  especially  chap.  VI.  As  to  his  complete  identification  of  sembling  with  Einfuhl- 
ting,  I  think  there  is  some  doubt,  since  the  latter,  in  at  least  some  of  its  aspects,  is 
earnest,  and  the  feeling  has  presumption  and  judgment — not  merely  assumption — as  its 
presupposition. 


Modes  of  the  Consciousness  of   Value  7 1 

with  the  activities  of  mere  apperception  of  content  in  its  relations, 
with  contemplation,  while  in  the  background  remains  the  as 
sumption  of  existence,  with  its  reference  to  conative  dispositions. 
While  the  object  is  detached  from  immediate  desire,  its  relation 
to  desire  is  not  absolutely  severed. 

An  illustration  will  show  the  situation  with  greater  clearness. 
The  aesthetic  appreciation  of  feminine  beauty  is  a  psychosis 
grafted  immediately  upon  desire  and  desire-dispositions.  The 
process  by  which  the  aesthetic  psychosis  supervenes  upon  that 
of  crude  desire  is  one  of  arrest,  social  and  individual,  and  of  re 
arrangement  of  the  elements  of  the  object,  presented  either  un 
consciously,  or  consciously  as  in  art,  in  such  a  manner  as  to  fill 
the  foreground  of  consciousness  with  presentational  activity, 
and  to  detach  the  object  from  immediacy  of  desire.  An  implicit 
assumption  of  the  existence  of  the  object  for  desire  is,  however, 
a  necessary  presupposition  of  the  aesthetic  appreciation.  Should 
the  conative  disposition  become  explicit  in  actual  desire,  the 
aesthetic  repose  would  cease  and  a  new  adaptation  take  place. 

In  both  these  appreciative  modes,  it  should  finally  be  observed, 
worth  or  affective-volitional  meaning  has  been  acquired.  The 
deepening  of  the  transgredient  or  immanental  reference,  as  the 
case  may  be,  becomes  part  of  the  funded  meaning  of  the  object, 
and  is  imputed  to  the  object.  The  recognition  of  this  fact  is  of 
far-reaching  importance,  for  all  the  meanings  acquired  in  these 
modes  of  appreciation  enter  as  determinants  in  later  judgments 
of  value. 

2.    Acquired,  Meanings  of  Characterisation  and  Participation. 

Simple  appreciation,  with  its  two  primary  modifications  de 
scribed,  is  further  differentiated  into  secondary  acquired  mean 
ings  which  we  may  describe  as  personal  values  of  possession 
and  merit,  instrumental  or  utility  values  of  utilisation,  and 
the  common  values  of  participation.  The  characteristic  of  all 
these  modifications  of  primary  feeling  of  value  is  to  be  found  in 
the  fact  that  they  arise  through  the  establishment  of  relational 
judgments  between  the  object  and  the  disposition  presupposed. 
Otherwise  expressed,  the  merely  felt  transgredient  or  immanental 
reference  of  simple  appreciation  is  now  referred  to  its  explicit 
object  which  is  acknowledged  in  judgment. 

An  analysis  of  the  personal  feelings  makes  this  point  clear. 
The  feeling  of  possession  is  more  than  the  feeling  of  the  worth  of 
the  object,  as  presumed,  judged,  or  assumed  to  exist.  The 


72  Valuation  :   its  Nature  and  Laws 

object  acquires  an  imputed  value  through  the  explicit  acknowl 
edgment  of  the  subject  for  which  it  exists.  Similarly,  the 
feeling  of  personal  obligation  or  merit  arises  on  the  basis  of  a 
reference  of  the  valued  disposition  to  the  personality.  In  general 
we  may  say  that  the  personal  feelings  have  an  additional 
presupposition  of  reality  which  the  primary  feelings  have  not. 
But  the  more  developed  modes  of  these  primary  feelings,  the 
impellent  and  the  semblant,  contain  the  germ  of  these  personal 
values.  They  are  transition  stages  in  which  a  new  feeling  mode 
is  introduced,  through  the  transgredient  or  immanental  reference 
arising  upon  assumptions.  In  the  case  of  the  personal  value  the 
assumption  becomes  an  existential  judgment  of  acknowledgment 
of  the  self.  Of  course  such  a  transition  requires  ideal  construc 
tion  of  the  self,  and  this  involves  an  extension  of  simple  apprecia 
tion  through  sympathetic  Einfiihlung,  a  process  to  be  studied  in 
another  connection. 

The  impersonal  feelings  of  the  participation  values  or  utility- 
values  of  dispositions  and  objects  involve  a  further  extension  of 
this  acquirement  of  common  meaning.  In  addition  to  the  pre 
supposition  of  the  reality  of  the  desired  object,  there  is  an  ad 
ditional  presupposition  of  similar  desires  and  feelings  in  the  minds 
of  others  which  gives  rise  ultimately  to  judgments  and  assump 
tions  of  over-individual  demands.  How  such  presuppositions 
arise  is,  again,  of  course,  a  problem  of  genetic  psychology,  more 
especially  of  the  study  of  the  laws  of  sympathetic  imitation  and 
Einfiihlung.  The  main  point  here  is  that  the  appreciative  differ 
ences  in  the  meaning  of  the  feelings  arises  through  explicit 
acknowledgment  of  references  which  were  previously  merely 
implicit. 

It  should  be  noted  finally  that,  just  as  the  transgredient 
and  immanental  references  acquire  depth  of  meaning  through 
the  obligation  and  aesthetic  modes,  so  in  these  experiences 
primary  feeling  is  further  deepened  and  broadened. 

IV.  THE  QUANTITATIVE  MEANINGS  OF  WORTH  FEELING — DE 
GREE  OF  ACQUIRED  MEANING  AND  DEGREE  OF  INTENSITY — 
THEIR  RELATIONS 

i.    Analysis  of  Concept  of  Degree  of  Worth  Feeling — 

"  Depth  and  Breadth." 

Worth  predicates  have  been  defined  as  the  funded  meanings  of 
objects.  These  predicates  or  meanings  correspond,  we  have 


Modes  of  the  Consciousness  of  Value  73 

seen,  to  certain  qualitative  aspects  of  feeling,  primary  and 
derived.  But  they  have  also  a  quantitative  aspect — of  degree. 
To  what  aspects  of  feeling  do  these  differences  of  degree 
correspond  ? 

It  has  already  been  pointed  out  that  many  psychologists  have 
found  it  necessary  to  distinguish  between  degree  of  feeling  of 
value  and  degree  of  intensity  of  sensation-feeling,  and  some 
have  used  such  terms  as  depth  and  breadth  of  the  feeling  in  the 
personality  to  characterise  quantatitively  the  worth  suggestion  of 
the  feeling.  And  when  we  examine  more  closely  the  appreciative 
distinctions  made  in  the  sphere  of  worth  experience  it  becomes 
clear  that  some  such  distinction  is  necessary.  For,  in  the  first 
place,  it  is  to  be  observed  that,  if  we  make  use  of  those  apprecia 
tive  descriptions  of  feeling  subsumed  under  the  general  terms 
transgredient  and  immanental  references,  we  cannot  properly 
apply  the  quantitative  term,  intensity.  While,  for  instance,  we 
may  speak  of  the  degree,  we  cannot  properly  speak  of  intensity 
of  repose  or  expansion,  and  thus  seem  to  be  driven  to  some  such 
terms  as  depth  and  breadth.  Thus  we  find  Munsterberg1 
accepting  the  ordinary  formula,  that  intensity  of  feeling  de 
creases  with  repetition,  and  at  the  same  time,  in  his  desire 
to  do  justice  to  the  concrete  facts  of  worth  experience, 
insisting  that  repetition  may  increase  the  depth  of  feeling 
tone.  Clearly  depth  and  intensity  are  here  definitely  dis 
tinguished  and  admitted  to  be  independently  variable.  It 
appears,  then,  that  we  must  make  a  distinction  between 
degree,  or  intensity  in  the  broader  Kantian  sense,  and  intensity 
in  the  narrower  sense  of  sensational  intensity,  between  degree  of 
feeling  of  value  and  intensity  of  pleasantness-unpleasantness  as 
feeling-tone  of  sensations.  Intensity  in  this  latter  sense  applies 
to  all  sensation-feelings,  to  "  pleasure-causation,"  as  we  have  de 
scribed  it,  and  probably  to  all  sensation  feelings  which  enter  into 
a  total  feeling  complex,  but  not,  properly  speaking,  to  feeling- 
attitudes,  not  to  the  worth  aspect  of  feeling. 

What,  then,  is  the  relation  between  the  degree  of  acquired 
meaning  or  value  of  a  feeling  attitude  and  intensity  of  pleasant 
ness-unpleasantness  ?  How  are  they  related  for  appreciative 
introspection  and  analysis,  and  how  shall  this  empirical  relation, 
when  determined,  be  connected  with  our  analysis  of  the  con 
ditions,  actual  and  dispositional,  of  these  two  aspects  of  feeling  ? 
This  question  is  of  the  utmost  importance,  because  in  the  solution 

1  Miinsterberg,  Grundziige  der  Psychologies  Leipzig,  1900,  p.  39. 


74  Valuation :   its  Nature  and  Laws 

of  this  problem  is  involved  the  whole  question  of  the  measure 
ment  of  feelings  of  value,  to  which  we  must  presently  turn. 

2.  Degrees  of  "  Worth  "  and  Degrees  of  Intensity 

Independently  Variable — Illustrations. 

We  find,  then,  that  not  only  is  worth  experience  distinguishable 
:rom  pleasure-causation  in  the  aspects  both  of  quality  and  degree 
t  also  that  degree  of  value  varies  independently  of  hedonic 
Two  phenomena  of  our  worth  experience  indicate 
s  relation,     (i)  Feeling  of  positive  worth  may  exist  side  by  side 
with  unpleasant  experiences  and  feeling  of  negative  worth  with 
pleasant.     (2)   Degree  of  worth  feeling  may  increase  with  de- 
rease  of  hedonic  intensity,  and  there  are  numerous  instances 
where  worth  feelings  are  practically  intensity-less.     These  facts 
have  led  to  the  general  conception  of  the  irrelevance  of  the 
hedonic  aspects  of  a  total  attitude  for  worth  judgment  and  to 
the   formulation   of    Brentano's   doctrine   of   "hedonic   redun 
dancies." 

We  shall  examine  the  facts  briefly,  and  then  turn  to  a  con 
sideration  of  the  theories  of  the  relation  of  the  two  distinguish 
able  aspects.  The  first  phenomenon  is  well  illustrated  in  the 

issical  description  of  Lessing.  In  a  letter  to  Mendelssohn  he 
"  In  this  we  are  then  agreed,  my  dear  friend,  that  all 
passions  are  either  strong  likes  or  dislikes.  Also  in  this  that  in 
every  such  feeling  of  like  or  dislike  we  are  conscious  of  a  greater 
sense  of  reality,  and  that  this  consciousness  cannot  be  other 
than  pleasant.  Consequently,  all  passions,  even  the  most  un 
pleasant,  are  as  passions  pleasant."  '  The  paradox  of  calling  that 
which  is  unpleasant  pleasant,  and  the  lack  of  adequate  analysis 
in  this  description,  should  not  blind  us  to  its  essential  truth  While 
the  same  feeling  cannot  at  the  same  time  be  both  pleasant  and 
unpleasant,  it  is  quite  possible  that  we  are  concerned  here  with 
two  actual  feelings  in  certain  definite  relations  to  each  other 

Plausible  explanations  have  been  given  from  the  point  of 
view  of  the  identification  of  worth  feeling  with  a  form  of  simple 
pleasure.  It  is  said  that  we  have  to  do  here  with  an  illusion 
of  judgment,  that  what  was  formerly  unpleasant  has  now  be 
come  pleasant  through  change  in  physiological  disposition,  and 
that  the  unpleasantness  instead  of  being  real  is  merely  a  memory 

former  unpleasantness.     It  seems  hardly  necessary,  however, 

1  Quoted  from  Him,  The  Origins  of  Art,  London,  Macmillan,  1900,  p.  60. 


Modes  of  the  Consciousness  of  Value  75 

to  deny  in  the  interests  of  theory  what  is  a  fairly  constant 
deliverance  of  appreciation,  namely,  that  feeling  of  positive  worth 
may  be  co-existent  with  actual  unpleasantness.  Or  it  might  be 
said  that  we  have  a  simple  case  of  mixed  feeling.  A  pleasant 
and  unpleasant  sensation-feeling  may  exist  side  by  side  in  the 
same  state  of  consciousness,  as  for  instance,  the  pleasant  taste 
of  sugar  and  the  unpleasant  sensations  of  satiety  as  they  are 
just  beginning  to  appear — why  then  should  not  two  worth  feelings 
or  worth  feeling  and  simple  pleasantness  or  unpleasantness  ? 
To  this  we  may  answer  that  the  two  cases  are  not  parallel.  The 
inapplicability  of  the  concept  of  mixture  or  fusion  to  feelings  of 
value  we  have  already  pointed  out,  and  in  this  case  the  figure  is 
especially  misleading. 

If  we  look  at  Lessing's  description  more  closely  we  find  that 
his  paradox  really  arises  from  a  failure  to  analyse — to  distin 
guish  between  two  aspects  of  the  total  psychosis,  the  feeling  of 
value  and  the  irrelevant  hedonic  accompaniments.  The  situa 
tion  he  describes  admits  of  two  interpretations.  On  the  one 
hand,  the  passion,  of  anger  let  us  say,  is  really  a  feeling  of  nega 
tive  worth,  with  certain  cognitive  presuppositions — unpleasant, 
as  Lessing  says.  It  is  quite  possible,  however,  that  the  organic 
disturbance  may  be  pleasantly  toned,  especially  after  long- 
continued  arrest,  with  its  unpleasant  strain  sensations.  We 
should  then  have  pleasant  accompaniments  of  a  feeling  of 
negative  worth.  On  the  other  hand,  it  is  equally  possible 
that  what  Lessing  calls  the  pleasantness  of  the  unpleasant 
passion  may  really  cover  a  gradual  transition  from  one  feeling 
of  value  to  another,  and  that  what  he  calls  the  pleasantness 
of  the  psychosis  may  be  a  feeling  of  value  of  the  personal  type. 
The  object  itself  may  have  negative  worth,  while  the  entire  ex 
perience  of  having  such  a  passion,  or,  in  fact,  the  knowledge  of 
the  capacity  for  such  reaction,  may  give  rise  to  a  feeling  of 
satisfaction,  of  personal  worth.  This  might  even  extend  to 
such  passions  as  have  unpleasant  hedonic  accompaniments. 
Feelings  of  value  would  then  be  accompanied  by  unpleasant 
sensation  feelings. 

The  second  group  of  facts  which  lead  to  this  differentia 
tion  of  degree  of  intensity  of  pleasantness-unpleasantness  from 
degree  of  worth  or  meaning  of  the  feeling,  are  the  so-called 
intensity-less  attitudes  or  acts  of  valuation  and  preference. 
Here,  it  is  maintained,  quasi-logical  modifications  take  the  place 
of  intensity.  If  we  begin  with  the  two  primary  modifications 


7  6  Valuation:   its  Nature  and  Laws 

of  simple  appreciation,  the  ethical  and  aesthetic,  we  find  intensity 
giving  place  to  other  modifications.  A  quiet  sense  of  obligation 
may  reveal  a  degree  of  worth  of  an  ideal  object  which  the  in- 
tensest  passion  or  emotion  does  not  suggest.  Similarly  in  the 
aesthetic,  or  semblant  mode,  a  degree  of  immanental  worth  may 
be  suggested  in  the  depth  and  breadth  of  the  feeling  when  the 
element  of  intensity  is  reduced  to  a  minimum.  But  still  more 
evident  do  these  facts  become  when  we  pass  to  the  secondary, 
derived  feelings,  the  personal  and  the  impersonal  over-individual 
references.  In  a  case  of  preference  between  objects  to  which 
these  feelings  correspond,  a  relatively  intensity-less  feeling  of 
personal  worth  may  have  an  affective-volitional  meaning  which 
the  intensest  passion  connected  with  sense  objects  has  not. 
And  so  with  the  over-individual  feelings.  If,  then,  by  intensity 
we  mean,  not  the  broader  Kantian  conception  of  any  modification 
of  degree  of  inner  experience,  but  that  particular  degree  which 
applies  to  sensation  and  feeling-tone  of  sensation,  there  can  be 
no  question  but  that  worth  feelings,  as  determined  by  judgment 
and  assumption,  may  be  practically  intensity-less.  These  acts 
are,  of  course,  causally  connected  with  sensation  tendencies, 
both  peripheral  and  organic,  and  every  such  act  has  as  its  accom 
paniment  secondary  hedonic  resonances  of  more  or  less  inten 
sity,  but  the  point  is  that  appreciatively  we  can  distinguish  the 
two  factors,  and  are  aware  that  the  latter  do  not  determine  the 
worth  judgment. 

The  facts  upon  which  this  hypothesis  of  independent  varia 
bility  of  the  two  factors  in  a  total  worth  attitude  is  based  are 
now  before  us,  as  well  as  the  subordinate  role  which  the  hedonic 
resonance  plays  in  worth  judgments.  We  are,  however,  as  yet 
wholly  without  any  conception  which  will  enable  us  to  under 
stand  this  relation  functionally. 

3.    Theories  of  this  Relation — Suggestion  of  a  Theory  to 
be  developed  later. 

There  are  two  general  theories  of  this  relation,  which  may 
be  described  as  the  dualistic  and  the  monistic  or  genetic.  The 
dualistic  theory  is  represented  by  Brentano  and  Schwartz.  In 
Brentano's  view,1  as  we  have  seen,  any  concrete  attitude  of 
valuation  can  be  analysed  into  two  aspects,  intensity-less  acts  of 
preference,  acts  of  love  and  hate,  and  the  hedonic  redundancies 

1  Brentano,  Psychologic,  especially  p.  197.    Also  Ur sprung  der  sitt  lichen  Erkentniss, 
especially  p.  86. 


Modes  of  the  Consciousness  of  Value  77 

which  accompany  them.  To  the  latter  alone,  as  sensation-feel 
ings,  belong,  properly  speaking,  degrees  of  intensity.  In 
Schwartz's  view1  feeling  intensity  belongs  to  the  passive  side 
of  consciousness,  degrees  of  worth  to  the  active,  voluntaristic 
side.  The  latter  appear  in  the  form  of  acts  of  analytic  and 
synthetic  preference.  The  essential  point  of  both  conceptions  is 
the  dualism  between  feeling  and  will,  and  the  reference  of  worth 
distinctions  to  modifications  of  will. 

The  facts  which  have  given  rise  to  this  theory  are,  as  we 
have  seen,  true  enough.  So  also  is  the  conception  of  hedonic 
redundancies,  in  so  far  as  it  merely  describes  the  functional 
relation  of  these  two  aspects.  But  it  is  far  from  certain 
that  it  is  necessary  to  draw  the  dualistic  conclusion.  That 
would  follow  only  on  condition  that  feeling  and  will  are 
totally  different  elements,  that  the  distinction  between  them 
as  active  and  passive  is  ultimate,  and  secondly,  that  the  only 
modification  of  feeling  which  can  be  made  the  equivalent  of 
degrees  of  worth  is  hedonic  intensity. 

Whether  these  assumptions  are  necessary  must  be  determined 
ultimately  by  a  consideration  of  the  whole  question  of  the  psy 
chology  of  feeling  and  will  and  their  relations,  which  must  be 
reserved  for  another  chapter.  It  will  be  sufficient  here  to 
deny  the  necessity  of  such  assumptions,  and  in  the  meantime 
to  suggest  a  second  possible  conception,  monistic  and  genetic 
in  character.  Feeling,  according  to  our  analysis,  has  other 
modifications,  other  meanings  than  passive  pleasantness  -  un 
pleasantness,  viz.,  transgredient  and  immanental  references  to 
conative  dispositions.  These  references,  which  arise  only  when 
the  disposition  is  actualised  by  cognitive  acts  of  presumption, 
judgment,  assumption,  are  signs  of  the  affective-volitional 
meaning  of  the  object,  its  relation  to  conation.  Feeling  as 
passive  is  therefore  not  to  be  separated  from  will  as  active. 
But  more  than  this,  these  references  or  aspects  may,  con 
ceivably — with  repeated  actualisation  of  the  dispositions — 
become  differentiated,  as  developed  meanings,  from  the  aspect 
of  hedonic  intensity,  and  increase  in  depth  and  breadth.  If 
this  view  should  prove  tenable,  we  should  have  a  relation 
analogous  to  that  between  the  general  concept  and  the  particular 
presentation.  As  the  meaning  of  the  concept  develops  with 
actualisation  of  the  judgment-disposition  in  successive  cognitive 
acts,  the  particular  presentation  becomes  less  and  less  significant, 

1  Schwartz,  Psychologic,  des  Willens,  chap.  II  ;  also  Appendix  I. 


78  Valuation:   its  Nature  and  Laws 

until  we  have  what  is  practically  imageless  apprehension. 
So  also  with  the  development  of  the  meanings  of  feeling,  the 
hedonic  resonance  may  become  less  and  less  significant  until 
relatively  intensity-less  appreciation  of  the  worth  of  the  object 
appears.  The  substantiation  of  such  a  conception  of  affective 
generalisation  involves  a  more  extended  excursion  into  the  psy 
chology  of  feeling.  Here  we  may  merely  note  the  fact  that  such 
feeling-attitudes  exist,  i.e.,  in  the  case  where  the  presuppositions 
are  assumptions,  either  of  the  explicit  or  implicit  type. 

V.  CONCLUSION— THE  BEARING  OF  THIS  ANALYSIS  ON 
FURTHER  PROBLEMS 

In  concluding  this  chapter  we  may  with  advantage  return  to 
a  consideration  of  the  preliminary  definition  of  worth  and  worth 
predicates  from  which  the  entire  analysis  took  its  start.  This 
analysis,  it  will  be  seen,  has  given  content  to  that  definition.  It 
has  also  given  us  the  ground-work  for  further  researches  into 
the  principles  governing  the  concrete  phenomena  of  valuation 
of  different  types,  economic,  ethical,  aesthetic,  etc.  A  more 
general  view,  both  retrospective  and  prospective,  will  serve  to 
give  unity  to  the  results  thus  far  attained. 

In  general,  we  found  worth  or  value  to  be  the  funded  affective- 
volitional  meaning  of  the  object  for  the  subject.  That  funded 
meaning,  as  expressed  in  terms  of  the  predicates,  goodness, 
utility,  beauty,  obligation,  desert,  etc.,  represents  the  desira 
bility  of  the  object,  although  not  necessarily  the  fact  of  actual 
desire.  The  funded  meaning  is  acquired  through  actualisa- 
tion  of  conative  dispositions  by  acts  of  presumption,  judgment, 
and  assumption,  and  this  actualisation  results  in  feeling  which 
undergoes  certain  modifications,  with  change  in  presuppositions, 
and  with  repetition.  This  feeling,  with  its  modifications,' 
reflects  the  funded  meaning  of  the  object.  Worth  predication,' 
in  the  aspects  both  of  quality  and  degree,  is  determined  by  ap 
preciative  modifications  of  feeling,  which  in  turn  are  determined 
by  changes  in  the  presuppositions  of  the  feeling. 

To  these  funded  meanings,  roughly  classified  as  simple  ap 
preciation  of  objects  (with  its  impellent  and  semblant  modes) 
personal  worths  of  characterisation,  and  common  over-indi 
vidual  values  of  participation  and  utilisation,  correspond  certain 
classes  of  objects,  primary  and  derivative,  perceptual  and  ideal. 
All  these  derived  objects,  with  their  corresponding  attitudes,  are 


Modes  of  the  Consciousness  of  Value  79 

perceptual  and  ideal  constructions  which  emerge  through 
certain  value  movements  from  simple  appreciation.  The  genesis 
of  these  objects,  with  their  corresponding  predicates,  is  one  of 
the  chief  problems  which  now  present  themselves.  This  differ 
entiation  and  fixation  of  objects  and  predicates  of  valuation 
must  be  traced  to  fundamental  laws  of  the  psychical  processes 
by  which  affective-volitional  meaning  is  acquired.  These  laws 
we  may  describe  as  the  Laws  of  Valuation. 

But  worth  predication  has  a  quantitative  as  well  as  quali 
tative  side.  Worth  judgments  express  preferability  of  one  object 
over  another,  as  well  as  degrees  of  preferability  of  different 
amounts  of  the  same  object.  We  are  thus  led  to  the  problem 
of  the  measurement  of  the  worth  or  funded  meaning  of  objects. 
At  this  point  several  questions  arise.  Is  worth  or  value,  as  we 
have  conceived  it,  an  object,  a  function,  to  which  the  concepts 
of  quantity  and  measurement  can  be  applied  ? 

In  answering  this  question  we  must  first  note  the  fact  that 
such    quantitative    judgments    do    exist.     Within    the    various 
regions  of  worth  predication  numerous  empirical   uniformities 
are  discoverable  connecting  quantity  of  object  with  degree  of 
worth  predicated.     Thus  in  the  region  of  economic  "  condition  " 
worths,   there   are   certain   empirical  laws   connecting   changes 
in  the  intrinsic  desirability  or  in  the  utility  (instrumental  de 
sirability)  of  an  object  with  changes  in  its  quantity.     In  the 
region  of  judgments  of  personal  worth  the  obligation  or  merit 
predicated  varies  in  certain  definite  ways  with  changes  in  the 
amount  of  the   object   or   in   this   case   disposition   displayed. 
The  same  is  true  of  those  judgments  upon  dispositions  according 
to   their   over-individual   value  for   participation.     It  is  clear, 
then,  that  empirical  relations  of  a  quantitative  character  may 
be  established  between  objects  and  their  worth  predicates  or 
funded  meanings.     But  such  empirical  laws  would  constitute  no 
explanation,  nor  would  they  enable  us  to  establish  relations  of 
degree   between   objects   of   these    different   types.     While   we 
might  formulate  empirical  statements  of  dependence  of  degree 
of  value  of  the  object  upon  changes  in  the  object  without  for 
mulating  any  theory  of  the  psychological  grounds  for  this  de 
pendence,  such  measurement  must,  if  it  is  to  lead  to  any  insight 
into  the  nature  of  worth  judgments,  involve  the  reduction  of 
these    empirical    uniformities    to    more    ultimate    psychological 
laws.     Thus,   to   take   an   illustration   from   another  region   of 
psychology,    the   significance   of   the   empirical   formulation   of 


8o  Valuation  :   its  Nature  and  Laws 

Weber's  law  for  perception  holds  good,  irrespective  of  any 
theory  of  its  psychological  explanation.  Or,  to  take  another 
illustration  from  a  more  closely  related  region  of  investigation, 
the  law  of  marginal  utility  in  economics  is  an  empirical  law 
which  holds,  within  limits,  irrespective  of  its  interpretation,  and 
is  capable  of  explanation  in  terms  which  do  not  necessitate  the 
hypothesis  of  continuous  change  in  hedonic  intensity.  We  must 
therefore  distinguish  between  the  merely  empirical  formulation 
of  more  and  less  and  a  theory  of  the  psychological  determinants 
of  the  change  in  degree  of  worth  or  affective-volitional  meaning 
of  the  object. 

The  question  whether  worth,  or  funded  meaning  of  an  ob 
ject  as  we  have  denned  it,  is  susceptible  of  measurement  is 
reduced,  then,  to  the  still  more  fundamental  question  whether 
the  psychological  determinants  of  that  meaning  are  objects  of 
measurement.   Into  the  acquired  and  funded  meaning  of  an  object 
enter  various  elements  presupposing  various  processes  and  atti 
tudes.     If  these  can  be  analysed  out,  and  their  contributions  to 
the  total  worth  of  the  object  determined,  such  measurement  is 
possible.     On  the  view  which  we  have  rejected — that  degree 
of  value  is  to  be  equated  with  degree  of  intensity  of  pleasant 
ness-unpleasantness,  or  as  sometimes  formulated,  with  a  func 
tion  of  intensity  and  duration,  the  problem  is,  at  least  theoreti 
cally,  simple.     The  laws  of  habit,   satiety,   contrast,   etc.,   for 
sensation-feelings  might  be  applied  directly  to  feelings  of  value. 
But  such  a  procedure  is  impossible  after  our  analysis.     The 
psychological  determinants  are  for  us  more  complex.     Having 
defined  feelings  of  value  as  feelings  presupposing  dispositions 
actualised    by    presumption,    judgment,    and    assumption,    our 
problem  is  the  determination  of  the  capacity  of  the  object,  as 
presumed,  judged,  or  assumed  to  exist,  to  call  out  feelings  of 
value.     Since  the  worth  of  the  object  is  a  function  of  the  ca 
pacity  of  the  subject  for  feeling,  as  determined  by  these  preceding 
processes  of  accommodation  in  judgment  and  assumption,  we 
must  inquire  into  the  effect  of  these  processes  upon  the 'dis 
positions  presupposed.     The  analysis  and  formulation  of  these 
factors  constitute  the  laws  of  valuation.     Such  laws  are  capable 
of  determination,  and  when  determined  they  enable  us  to  explain 
the  empirical  laws  of  "  more  and  less." 


i 


CHAPTER    IV 

THE  PSYCHOLOGICAL  BASIS   OF  A 
THEORY  OF   VALUATION 

THE  preceding  studies  of  worth  experience — of  "  feelings  of 
value,"  of  their  meanings  and  presuppositions — have  brought 
us  to  a  point  where  new  problems  present  themselves  which 
necessitate  new  methods  of  solution.  Thus  far  we  have  been 
concerned  merely  v/ith  the  analysis  of  the  fundamental  modifi 
cations  of  feelings  of  value  and  of  their  presuppositions.  Our 
task  is  now  to  make  use  of  the  results  of  this  analysis  in  the 
determination  of  the  laws  of  valuation,  as  a  process  of  ac 
quirement  of  affective-volitional  meaning.  Our  former  prob 
lem  of  analysis  and  description  has  now  become  one  of 
explanation. 

The  problem  of  explanation  is  a  genetic  problem,  in  the  larger 
sense  of  that  word.  Now  the  genetic  method  is,  we  have  seen, 
but  an  extension  of  the  presuppositional  method.  Its  task  is, 
in  general,  to  show  how  valuation  may  be  conceived  as  a  sys 
tematic,  progressive,  and  continuous  determination  of  the 
stream  of  conation  and  feeling  in  the  individual's  mind,  to 
show  how  presuppositions  become  actual  in  feeling  and  desire, 
and  how  actual  feelings  and  desires,  and  dispositions  when 
formed,  become  presuppositions  of  new  feelings  and  desires, 
and  of  further  modifications  of  quality  and  degree. 

When  the  nature  of  the  task  is  thus  stated,  it  becomes  evident 
that  its  successful  prosecution  involves  a  further  excursion  into 
the  psychology  of  feeling  and  will.  We  have  as  yet  no  adequate 
conception  of  the  nature  and  relations  of  feeling  and  will,  such 
as  would  enable  us  thus  to  conceive  valuation  as  a  systematic, 
progressive,  and  continuous  determination  of  the  stream  of 
conation  and  feeling.  We  have  not  yet  developed  such  abstract 
conceptions  of  feeling  and  conation,  as  would  make  possible 
an  account  of  the  manner  in  which  actual  experiences  of  feeling 
G  81 


Valuation :   its  Nature  and  Laws 

and  desire  create  the  presuppositions  of  succeeding  valuations. 
We  have  no  developed  concept  of  affective-volitional  deter 
mination.  To  this  general  problem  we  must  now  turn  our 
attention. 

For  the  purposes  of  this  study  the  general  question  of  the 
nature  of  feeling  and  will  and  their  relations  may  be  reduced  to 
three  special  problems. 

i.  In  the  first  place,  what  is  the  nature  of  feeling  and  of  its 
relation  to  conation,  actual  and  dispositional,  as  required  by 
our  definition  and  analysis  of  worth  experience  ?  We  dis 
tinguished  between  feeling  and  desire,  and  although  the  dis 
tinction  appeared  to  be  not  very  clear  at  the  limits,  we  found 
grounds  for  defining  experience  of  worth  as  feeling.  But  we 
found  feeling  determined  by  conative  dispositions,  and  these 
in  turn  determined  by  feeling.  How  shall  we  understand  this 
relation  ?  Moreover,  in  our  analysis  of  the  fundamental  modi 
fications  of  worth  feeling  we  used  the  concept  of  feeling  in  a  broad 
sense,  which  included  references  to  conation  in  a  sense  quite 
different  from  the  definition  of  feeling  as  passive  affection. 
Clearly,  a  psychological  theory  of  feeling  and  will  is  involved, 
which  must  now  be  more  fully  developed. 

2.  In  the  second  place,  the  appreciative  analysis  of  the  differ 
ent  modifications  of  worth  feeling  calls  for  some  further  theory  of 
the  nature  of  feeling  itself.     This  demand  appears  especially  in 
connection  with  our  distinction  between  degree  of  intensity  of 
feeling  and  degree  of  worth  or  affective-volitional  meaning,  but 
is  also  involved  in  our  analysis  of  the  transgredient  and  imma- 
nental  references  and    their   distinction    from  pleasantness-un 
pleasantness.     These  aspects  or  meanings  of   feeling  -  attitude 
must  be  founded  upon  some  actual  content  or  process  of  con 
sciousness,   must    have   equivalents   in   terms   of    content   and 
function. 

3.  Finally,  we  have  a  third  problem,  which  may  be  described 
as  the  problem  of  the  continuity  of  worth  process.     This  is  closely 
connected  with   the  preceding  question— how  an   object   may 
acquire  funded  meaning  while  the  intensity  of  feeling  reaction 
diminishes.     The  problem  presents  itself  in  the  following  way. 
The  consciousness  of  value  is  at  any  given  moment  an  emo 
tional  consciousness.     Such  emotional  states  are,  however,  when 
viewed   as  content,   discrete   and    fugitive.      While,   therefore, 
worth  judgment,  viewed  in  the  light  of  its  meaning,  is  continu 
ous—that  is  worth  experiences  of  the  present  are  conserved  in 


Psychological  Basis  of  a   Theory  of  Valuation         83 

succeeding  judgments  of  value— the  emotional  experiences  which 
find  expression  in  these  judgments,  seem  at  first  sight  discon 
tinuous.  Are  feeling  and  emotion  thus  fugitive  and  discontinu 
ous,  so  that  their  meaning  survives  merely  in  the  physiological 
dispositions  created  ?  Or  is  there  such  a  thing  as  a  felt  con 
tinuity,  wherein  feeling  acquires  recognitive  and  generic  meanings, 
which  are  then  taken  up  into  permanent  sentiments,  and  upon 
the  basis  of  which  as  presupposition  new  feelings  and  volitions 
are  formed?  Is  there  such  a  thing  as  an  "affective  logic"  or 
quasi-logical  continuities  of  affective- volitional  meaning  ?  It  is 
in  connection  with  this  problem  that  a  satisfactory  conception 
of  the  nature  of  feeling  as  content  of  experience,  and  as  functional 
in  determining  succeeding  experiences  of  feeling  and  will,  is 
most  important.  Its  solution  is  possible  only  after  an  investi 
gation  of  the  first  two  problems  concerning  the  nature  of  feeling 
and  will.  To  these  the  present  chapter  will  be  devoted,  the 
problem  of  the  nature  of  this  continuity  being  reserved  for  a 
special  investigation. 

I.  THE  NATURE  OF  FEELING  AND  WILL  AND 
THEIR  RELATIONS 

i.  The  Problem. 

The  problem  of  feeling  and  will  and  the  nature  of  their 
relations  is  perhaps  the  most  difficult  within  the  entire  field  of 
psychological  analysis.  The  reason  for  this  is  not  far  to  seek, 
for  nowhere  is  it  more  important  that  the  distinction  between 
appreciative  and  non-appreciative  description  should  be  realised 
and  a  true  theory  of  their  relations  formed,  and  nowhere  is  there 
such  confusion  on  these  points  as  in  this  sphere.1 

To  illustrate  my  point  in  detail,  the  distinction  between  feel 
ing  as  passive  and  will  as  active  is  an  appreciative  distinction. 
One  concrete  attitude  is  relatively  more  passive  than  another 

1  The  consequence  has  been  the  widely  divergent  analyses  with  which  psychologists 
have  been  scandalised.  The  original  distinctions  within  this  sphere  were  made  from  the 
appreciative  point  of  view  because  analysis  of  feeling  and  will  first  began  S  the  worth 
problem,  e.g.,  Plato  and  Aristotle  and  in  modern  times  the  English  Utilitarians  As  the 
original  interest  became  secondary  to  that  of  non-appreciative  description,  the  distinctions 
developed  in  appreciative  description  were  applied  without  reflection  to  a  hypothetical 
feeling  abstracted  from  Us  presuppositions.  Tradition  was  all  powerful  here  tor  we  are 
natural  y  conservative  m  all  that  concerns  the  feeling  and  worth  side  of  exper  enc^and 
when  a  last  independence  of  analysis  appeared,  the  question  of  the  retention  or  elimina 
tion  of  these  distinctions  seems  to  have  been  determined  largely  by  personal  ind™  ion 
rather  than  by  considerations  of  scientific  method. 


84  Valuation  :   its  Nature  and  Laws 

with  reference  to  its  meaning  in  a  series  of  attitudes,  i.e.  with 
reference  to  what  succeeds  or  precedes  ;  but  when  we  abstract 
from  the  meaning  of  the  attitude  and  apply  the  distinction  to 
hypothetical  content,  it  involves  us,  we  shall  find,  if  it  is  made 
absolute,  in  contradictions,  and  is  far  from  representing  the  facts. 
The  distinctions  between  feeling,  emotion,  impulse,  desire,  wish, 
and  will,  are  primarily  appreciative,  made  with  reference  to  the 
meanings  of  the  attitude  and,  as  we  shall  see  later,  go  back  to 
certain  differences  in  cognitive  presuppositions.  Similarly,  the 
distinction  pleasantness-unpleasantness,  and  its  selection  as 
the  dominant  in  the  feeling  complex  or  attitude  to  the  exclusion 
of  other  aspects,  is  one  which  has  been  determined  largely  by 
appreciative  purposes,  i.e.,  it  is  the  abstract  aspect  which  appears 
emphasised  when  the  subjective  attitude  is  transformed  into  a 
state  and  itself  becomes  an  object.  Now  when  these  apprecia 
tive  distinctions,  which  are  largely  concerned  with  the  intent  of 
an  attitude  rather  than  with  the  content  of  a  state,  are  taken  to 
apply  to  content  from  which  meaning  has  been  abstracted,  in 
teresting  difficulties  and  contradictions  arise.  When  the  dis 
tinctions  between  passive  and  active,  feeling  and  conation,  are 
taken  as  non-appreciative  ultimate  distinctions,  we  have  a 
dualism  in  affective-volitional  meaning  which  the  several  different 
dualistic  theories  seek  to  bridge  by  establishing  relations  of 
causal  determinism  between  the  two  aspects.  One  finds  feeling, 
as  a  distinct  element  (passive  pleasantness  or  unpleasantness), 
the  necessary  antecedent  of  all  conation  ;  another,  giving  the 
primacy  to  conation,  finds  in  the  passive  feeling  the  sign  of  the 
satisfaction  or  arrest  of  some  antecedent  active  impulse  or  desire  ; 
or,  finally,  the  dualism  may  be  pressed  so  far  (as  in  the  recent 
work  of  Schwartz)  as  to  admit  the  existence  of  volition  without 
fee)ing. 

The  extent  to  which  these  fundamental  differences  colour  all 
worth  analysis  and  theory  is  obvious.  Psychological  hedonism, 
with  its  incapacity  to  explain  a  great  part  of  worth  experience, 
is  the  result  of  the  first.  A  theory  which  is  unable  to  include 
the  aesthetic  in  the  sphere  of  worths  is  the  result  of  the  second. 
From  the  third  we  get  the  strained  formalism  of  Kant  and 
Schwartz.  In  view  of  these  difficulties,  no  theory  of  feeling 
and  will  and  of  their  relations  (and  some  theory  is  necessary)  is 
of  any  value  unless  it  is  formed  with  a  clear  consciousness  of 
the  problem  involved  in  the  relation  of  the  appreciative  to  the 
scientific  description  of  the  psychical. 


Psychological  Basis  of  a  Theory  of  Valuation         85 

There  are  two  views  which  have  been  formed  with  this  clear 
consciousness  of  the  methodological  presuppositions  involved. 
On  the  one  hand,  Meinong  tells  us,  the  relation  of  feeling  and 
will  can  be  determined  only  from  the  worth  standpoint,  while 
Wundt  looks  upon  the  distinctions  introduced  from  the  point  of 
view  of  worth  analysis,  such  as  the  distinctions  between  feeling, 
desire,  and  will,  as  "  pure  logical  artifacts,  not  in  the  least 
psychical  ultimates  distinct  from  each  other."  As  a  conse 
quence,  the  distinction  between  feeling  and  will  is  for  the  former 
ultimate,  while  for  Wundt's  monistic  theory  there  is  a  funda 
mental  identity  of  feeling  elements  underlying  all  these  artificial 
distinctions. 

Between  two  such  divergent  views,  with  such  different 
methodological  presuppositions,  there  appears  to  be  no  middle 
ground,  and  yet  to  my  mind  each  has  a  relative  validity,  and  is 
susceptible  of  reconciliation  with  the  other.  More  than  this,  I 
am  inclined  to  think  that  the  Identity  theory,  developed  from  the 
standpoint  of  analysis  of  content,  is  the  only  one  which  will 
harmonise  with  the  distinctions  in  affective-volitional  meaning, 
developed  from  the  standpoint  of  functional  intent. 

2.    Dualistic  Theories  of  Feeling  and  Will — Criticism. 

We  may  begin  our  study,  then,  with  a  brief  critical  exam 
ination  of  those  views  which,  upon  the  assumption  of  an  absolute 
distinction  between  feeling  as  passive  pleasantness  or  unpleasant 
ness  and  conation  as  active,  seek  to  establish  a  relation  of  causal 
psychical  determination  between  them.  If  the  distinction  is 
one  of  content  viewed  apart  from  its  intent  or  meaning,  then  it  is 
necessary  that  experience  shall  show  us,  either  that  passive  feeling 
is  the  necessary  antecedent  of  all  active  states  which  are  called 
conative,  or,  on  the  other  hand,  that  all  passive  states  of  feeling 
have  as  their  necessary  antecedents  arrest  or  accommodation 
of  conscious  impulse  or  desire,  in  its  very  nature,  as  content, 
different  from  feeling. 

The  first  of  these  dualistic  views,  in  its  original  form  of 
psychological  hedonism,  was  beautiful  in  its  simplicity.  Feeling, 
as  a  passive  state,  is  always  an  effect  of  content,  sensation  and 
idea,  and  their  relations.  The  aspects,  quality  and  intensity, 
vary  with  the  changes  in  sensational  and  ideal  content,  and  the 
intensity  and  quality  determine  impulse,  desire,  etc.,  the  active 
side  of  consciousness. 


86  Valuation :   its  Nature  and  Laws 

A  very  superficial  examination  of  the  facts  suffices  to  show 
us  that,  if  by  feeling  we  mean  simple  passive  pleasantness  or 
unpleasantness  with  certain  intensities,  it  is  by  no  means  the 
necessary  antecedent  of  any  given  impulse  or  desire.     On  the  one 
hand  we  have  simple  impulses  for  which  there  is  no  such  con 
scious  hedonic  antecedent.     When  the  impulse  to  take  exercise 
comes  over  me  at  a  given  time,  introspection  will  show  me  that 
it  is  necessarily  preceded  neither  by  a  conscious  feeling  of  un 
pleasantness  nor  by  an  anticipation  of  pleasantness,  although 
either  may  be  the  antecedent.    On  the  other  hand  there  are  phen 
omena  of    a  more   developed    conation  which   we    have  seen 
described  as  "  intensity-less  "  acts  of  preference  in  which  affective 
disturbance  is  at  a  minimum,  and  which,  if  feeling  be  described 
as  passive  hedonic  intensity,  certainly  show  no  such  feeling  ante 
cedent.     Impulses  with  the  note  of  obligation  in  them  are  fre 
quently  of  this  character. 

That  there  are  changes  in  affective-volitional  meaning  (Ge- 
muthsbewegungen,  in  the  broadest  sense),  described  as  impulse 
and  desire,  which  do  not  presuppose  an  antecedent  passive 
hedonic  consciousness  or  consciousness  of  hedonic  difference,  is 
clear.  If  we  include  in  feeling  other  qualities  such  as  tension- 
•  relaxation,  restlessness-quiescence,  it  is  merely  a  verbal  quibble 
to  raise  any  question  of  antecedent  and  consequent.  We 
have  already  attributed  to  the  concrete  feeling  the  essential 
character  of  the  conative  side,  a  virtual  acceptance  of  the  Identity 
theory. 

This  fact,  that  there  are  numerous  impulses  and  desires  which 
follow  immediately  upon  presentation  and  judgment  without 
appreciable  hedonic  consciousness  intervening,  is,  moreover, 
admitted  by  the  upholders  of  the  theory  of  dependence  them 
selves.  Thus  Kreibig  speaks  of  dispositional  feelings  below  the 
threshold  as  determining  impulse  and  desire,  while  Ehrenfels 
speaks  of  desire  as  determined  by  feeling  or  feeling-dispositions. 
And  even  when  it  is  actual  feeling  which  is  conceived  as  causally 
determinative,  it  is  not,  as  we  have  seen  in  our  previous  analysis 
of  Ehrenfels's  worth  definition,  feeling  as  a  separate  antecedent 
state,  but  the  feeling-difference  as  determined  by  the  object  as 
existing  or  not  existing  and  the  feeling-disposition  of  the  subject. 
In  the  case  of  the  impulse  to  exercise  it  would  be,  not  necessarily 
the  unpleasantness  of  the  present  state  nor  the  anticipated 
pleasure,  but  the  difference  between  the  two,  which  constitutes 
the  necessary  presupposition  of  the  impulse  or  desire. 


Psychological  Basis  of  a   Theory  of  Valuation         87 

It  is  in  these  admissions,  and  consequent  modifications,  that 
we  see  the  failure  of  this  entire  theory  of  dependence  growing 
out  of  the  separation  of  feeling  from  conation.  A  feeling 
which  does  not  rise  above  the  threshold  is  a  pure  conceptual 
construction.  So  also  is  the  feeling-difference  when  made  the 
presupposition  of  desire.  For  a  feeling-difference  can  be  an 
actual  psychical  determinant  in  only  two  ways  :  either  it  is  an 
ideal  construction,  the  resultant  of  reflection  upon  feelings, 
and  then  we  have  an  idea  as  the  presupposition  of  the  desire 
or  else  this  difference  is  felt  as  tension  or  restlessness,  as  an  ex 
pectancy  generated  by  the  hypothetical  disposition,  the  active 
conative  moment  supposed  to  be  determined  by  the  feeling. 
Where  feeling- difference  is  conceived  to  be  the  presupposition  of 
conation,  it  is  either  not  distinct  from  conation  or  else  it  is  a 
purely  conceptual  construction. 

The  second  theory  of  dependence,  which  has  been  de 
veloped  upon  the  assumption  that  feeling  and  conation  are 
ultimates  from  the  point  of  view  of  content,  is  that  all  feelings 
have  as  their  necessary  antecedent  some  phase  of  conscious  con 
ation,  and  that  feeling  is  the  sign  of  arrest  or  satisfaction  of 
desire.  Here,  again — if  conation  is  conceived  to  be  an  aspect  of 
consciousness  which,  as  content  for  non-appreciative  description/ 
is  distinct  from  feeling,  it  is  difficult  to  establish  a  thorough-going 
relation  of  dependence.  It  is  true  that  affective  attitudes  on  the 
plane  of  worth  suggestion  presuppose  the  activities  of  accept 
ance  or  rejection,  but  even  here  it  cannot  be  said  that  the 
relation  is  one  of  antecedent  and  consequent,  nor  can  it  be  said 
that  the  worth  feelings  are  passive  pleasantness  and  unpleasant 
ness.  But  it  is  by  no  means  easy  to  include  in  such  a  general 
isation  all  the  phenomena  of  feeling.  There  are,  in  the  first 
place,  the  feelings  which  accompany  simple  sensations,  the 
agreeable  or  disagreeable  affective  tone  of  an  odour  or  colour. 
There  are  also  the  sudden  emotions  of  -surprise  and  fear,  and 
finally  the  instinctive  emotions,  inherited  and  appearing  at  first 
without  any  conative  experience  as  their  antecedent. 

As  to  the  first  group  of  phenomena,  those  who  hold  the  view 
that  feeling  has  its  rise  in  arrested  conation  insist  that  even  these 
phenomena  fall  under  the  general  law.  So  also  does  the  func 
tional  theory  in  general  when  it  is  consistent  and  sharply  dis 
tinguishes  feeling  and  conation.  Thus,  in  a  recent  article  written 
from  this  point  of  view,  unpleasantness  is  conceived  to  follow 
upon  arrested  conation,  while  pleasantness  appears  only  when 


Valuation :   its  Nature  and  Laws 

conation  is  accommodating  itself  after  arrest.     States  which  do 
not  contain  conative  moments  are  neutral. 

Nevertheless,  the  difficulties  in  the  way  of  such  an  answer 
are  not  to  be  minimised.     If  we  examine  the  reasons  given  for 
this  inclusion  we  find  that  they  are  of  two  kinds— the  first  being 
analytical  and  introspective,  the  second  functional.     The  first  is 
to  the  effect  that  it  is  impossible  to  get  the  feeling-tone  of  a 
simple   sensation   uncomplicated  with   the   aspects   of  tension- 
taxation    or  restlessness-quiescence,  with  their  suggestion  of 
onative  presuppositions.     The  second  or  functional  argument 
is  to  the  effect  that  the  law  of  decrease  of  affective  tone  through 
t  and  repetition  of  stimulus  is  primarily  a  law  of  adaptation 
tendency  to  stimulus,  and  that,  when  an  odour  or  tone  loses 
5  affective  tone  through  repetition,  it   does   so   because  the 
sense-tendency,  or  need  of  excitation  of  the  physical  organism 
•reduced   by   arrest,   has   been   satisfied.     Here    again    as   in 
the  preceding  theory,  the  relation  can  be  made  universal  only 
by  going  beyond  immediate  experience  and  supplementing  it 
with   hypothetical   conceptual   constructions.     The    aspects    of 
tension-relaxation  or  restlessness-quiescence,  if   they  appear  in 
the  simple  feeling-tone  of  sensation,  are  analytically  separable 
:rom  the  feeling  as  antecedent  content  and  intrinsically  different 
rom  feeling;    impulse  and   desire   are  not  conscious  "presuppo 
sitions  of  the  feelings.     Nor  when  the  intensity  of  feeling-tone 
imimshes  with  repetition  does  it  necessarily  mean  that  actual 
impulse  or  desire  gradually  disappears,  but  merely  that  some 
Imposition  or  tendency  diminishes  in  strength  with  repetition  of 
stimulus.     The    proposition    that    all    feeling    presupposes 
conation  holds  only  when  modified  to  read,  "  or  conative  dis 
position  and  tendency." 

The  same  reflections  hold  good  for  the  other  phenomena  of 
Jling,  the   sudden   emotions   of   surprise    and    fear,   and   the 
inherited  instinctive  emotions.     When,  upon  walking  through 
the  woods,  I  am  surprised  with  the  odour  of  flowers,  this  emotion 
has  as  its  presupposition  no  specific  experience  of  impulse  or 
Such  surprise  is  possible  with  relative  passivity  of  con 
sciousness,  although,  were  there  complete  passivity,  even  sur- 
i  would  be  impossible.     The  situation  seems  to  be  that  at 
least  some  general  conative  tendency  toward  objects  other  than 
the  flower  must  be  arrested  in  order  that  surprise  shall  arise, 
surprise  is  not  occasioned  by  the  odour  directly,  but  by  the 
arrest  of  some  other  conative  interest  or  tendency.     It  does  not, 


Psychological  Basis  of  a   Theory  of  Valuation        89 

however,  presuppose  actual  desire.  The  same  may  be  said  of 
the  instinctive  emotions.  They  presuppose  dispositional  or  i 
stinctive  conative  tendency,  not  actual  conation  ;  they  are  them 
selves  experiences  which  may  with  equal  right  be  described  as 
feeling  or  arrested  impulse.  Finally,  there  is  the  aesthetic 
feeling  in  the  case  of  which,  while  conation  is  presupposed  dis- 
positionally,  certainly  no  conscious  impulse  or  desire  necessarily 
precedes.  Analysis  shows  the  aspects  \vith  conative  connota 
tion,  relaxation  and  repose,  as  well  as  the  merely  hedonic,  but 
these  are  aspects  of  the  total  attitude,  not  different  states,  except 
for  retrospective  analysis. 

The  conclusion  of  these  reflections  is,  then,  that  a  thorough 
going   dependence   of  feeling,   as   distinguished  from  conation, 
upon  conation,  can  be  established  only  when  we  modify  our 
proposition  to  read  conation  or  conative  disposition  or  tendency. 
This  is  practically  the  conclusion  reached  in  the  examination  of 
the  theory  which  makes  conation  determined  by  feeling.     But 
when  we  have  introduced  the  concept  of  disposition,  that  is  when 
we  have  gone  beyond  the  distinctions  of  immediate  experience, 
and  supplemented  them  with  conceptual  constructions,  it  does 
not  matter  greatly  whether  these  dispositions  are  described  as 
belonging  to  feeling  or  desire.    As  Ehrenfels  wisely  recognises,  for 
worth  theory— which  is  concerned  with  the  changes  in  valuation 
and  their  laws,  as  determined  by  changes  in  dispositional  pre 
suppositions,  it  does  not  matter  whether  these  dispositions  are 
described  as  affective  or  conative  :    the  laws  of  valuation  will 
hold  on  either  assumption.     The  conclusion  of  real  importance, 
however,  is  that  the  distinction  between  feeling  and  will  is  not 
one  of  psychical  content,  but  of  intent  or  meaning. 

3.  Monistic  and.  Genetic  Theory  of  Feeling  and  Witt. 

The  chief  outcome  of  our  consideration  of  these  two  theories 
of  the  relation  of  feeling  to  will  is  that  no  thoroughgoing  relation 
of  dependence  can  be  established  either  way  except  by  leaving 
the  sphere  of  psychological  fact  and  supplementing  it  with  the 
conceptual  constructions  of  dispositions.  If,  however,  in  order 
that  we  may  fill  out  this  relation  of  dependence,  we  include  among 
the  attributes  of  feeling  restlessness-quiescence,  which  have  the 
conative  connotation  in  them,  it  is  doubtful  whether  anything  is 
gained  by  this  complete  separation  of  the  two  aspects  of  experience. 
The  "  Identity  "  theory  denies  that  this  distinction  is  fundamental, 


90  Vacation :    its  Nature  and  Laws 

and  asserts  that  it  arises  only  from  the  difference  in  point  of  view 
rom  which  we  look  at  one  primary  content  of  consciousness 
My  own  view  is  that  this  theory,  rightly  understood,  affords 
the  most  satisfactory  basis  for  a  true  theory  of  values  while 
doing  most  complete  justice  to  the  facts  of  analysis.  We  shall 
now  turn  our  attention  to  the  development  of  this  theory. 

In  its  most  general  form,  it  has  been  well  stated  by  Wundt 
in  the  psychological  part  of  his  Principles  of  Morality.1    There 
we  are  told  that  these  distinctions  are  purely  conceptual    de 
termined  by  the  point  of  view  from  which  we  observe  a  series 
inner  events,  the  flow  of  consciousness  itself  being  not  con 
cerned  with  them.     "  Every  act  of  will  presupposes  a  feeling 
with  a  definite  and  peculiar  tone  :   it  is  so  closely  bound  up  with 
elmg  that,  apart  from  it,  the  act  of  will  has  no  reality  at 
On  the  other  hand,  all  feeling  presupposes  an  act  of  will  ; 
the  quality  of  the  feeling  indicates  the  direction  in  which  the 
will  is  stimulated  by  the  object  with  which  the  feeling  is  con 
nected." 

This  view  is  developed  in  more  detail  from  the  standpoint  of 
psychological  analysis  of  content  in  the  last  edition  of  his  Psy 
chology.    Here  the  emotion  ("Affekt "  or  Gefiihlsverlauf]  is  taken  as 
e  ultimate  of  concrete  affective-volitional  meaning  or  intent  and 
this  emotion,  which,  as  content,  is  a  complex  of  feeling  elements 
may  be  called  emotion,  impulse,  desire,  or  will,  according  to  the 
nature  of  the  movement  or  complex.      "  The  question  is  no 
longer  what  specific  conscious  content  the  will  is,   but  what 
aspect  a  feeling  must  assume  to  become  volition."     This  specific 
difference  he  finds  (i)  in  the  character  of  the  "  end-feelings  "  of  the 
emotion  and  (2)  in  a  certain  meaning  or  intent  of  the  total  emotion 
which  can  be  formulated  only  in  retrospective  logical  terms.     As 
to  the  first  point,  conation  or  will-process  is  an  emotion  which 
through  its  movement  produces  a  final  feeling  which  in  turn 
destroys  the  emotion.     It  is  the  final  feeling  of  relaxation  which 
distinguishes   the   conative   process   from   emotion.     Again,    in 
the  entire  Gefiihlsverlauf,  when  experienced  as  conation,  there 
dwells  a  Zweck-richtung  which  is  realised  in  the  end-feeling  of 
relaxation.     Primary  conative   processes,   such  as  impulse    are 
emotions  with  this  meaning  ;   secondary  derived  conation,  such 
as  desire  and  will,  are  emotions  in  which  certain  single  feelings 
and  presentations,  elements  in  the  total  emotion,  are  singled 
out    as    the    motive    for    the    final   feeling  of  relaxation.     So 

1  Wundt,  Ethics,  Vol.  I,  «« The  Principles  of  Morality,"  pp.  6  and  7. 


Psychological  Basis  of  a   Theory  of  Valuation         9 1 

that  "  desire  is  not  so  much  the  preparatory  stage  of  an  actual, 
as  the  feeling  basis  of  an  arrested  conation."  The  experience 
which  constitutes  desire  may  be  viewed  as  feeling  or  conation 
according  to  the  point  of  view  from  which  it  is  observed.  All 
these  concepts  are  finally  logical  abstractions,  and  not  funda 
mental  distinctions  of  content.1 

A  similar  view  was,  in  all  its  essentials,  developed  by  Bren- 
tano,2  from  the  point  of  view  of  worth  analysis,  before  Wundt's 
present  formulation,  in  his  well-known  claim  that  in  a  given 
series  of  affective-volitional  meanings,  a  vita]  series  of  adaptation 
passing  from  feeling  to  will  (as,  for  instance,  sadness,  longing 
for  an  absent  good,  passing  into  desire  to  secure  it,  courage  to 
undertake  to  secure  it,  decision  to  act),  it  is  impossible  at  any 
point  to  make  an  absolute  distinction  between  feeling  and  will. 
They  constitute  rather  a  continuous  series  of  meanings  in  which 
these  two  aspects  can  be  distinguished  only  relatively  and  con 
ceptually. 

The  criticisms  passed  upon  this  conception  by  the  upholders 
of  the  dualistic  views  are  instructive  as  showing  the  contradic 
tions  involved  in  the  theories  which  make  these  distinctions 
ultimate  differences  of  content.  The  upholder  of  such  a  dualism 
must  put  his  finger  on  the  point  in  the  series  where  feeling  ends 
and  conation  begins.  Ehrenfels  finds  it  immediately  after  the 
first  stage  of  the  series.  Sadness  alone  is  pure  passive  unpleasant 
ness.  All  the  others  have  in  them  the  active  principle  of  desire. 
But  both  the  superficiality  and  the  contradictions  in  such  an 
analysis  become  immediately  evident.  For  what  is  involved  ? 
Clearly,  to  make  the  distinction  at  this  point  necessitates  the 
throwing  of  the  emotions  of  hope  and  courage  from  the  feeling 
to  the  desire  side  of  the  distinction,  as  indeed  Ehrenfels  does, 
and  the  logic  of  such  procedure  would  be  to  confine  feeling  to 
pleasantness  and  unpleasantness  as  passive  and  unspecified  states. 

1  Physiologisclie  Psychologic  ($th  edition),  Vol.  Ill,  chaps,  xvi  and  XVII.  Affekt  is 
here  translated  emotion  in  accordance  with  the  broad  use  of  that  term  in  chapter  in, 
p.  64,  and  as  recommended  by  the  Dictionary  of  Philosophy  and  Psychology,  pp.  310, 
317,  article  on  Emotion. 

»  The  considerations  which  were  influential  in  this  analysis  of  Brentano  were  pre 
cisely  those  of  which  we  have  already  taken  cognisance.    If  feeling  be  taken  as  identical 
with   passive  pleasantness  and  unpleasantness,  valuation  cannot   be  reduced  to  d 
mination  of  conation  by  feeling,   to   pleasure-causation.      Feeling,  it  is  true,  viewed 
merely  as  pleasantness  and  unpleasantness,  is  present  throughout  the  entire  accommod 
ative  or  vital  series,  such  as  that  described  above,  but  it  becomes  less  and  less  signifi 
cant  in  the  latter  stages  where   the   dynamic    tension  becomes  dominant.     Hedonic 
intensities  become  irrelevant  redundancies  and  we  have  practically  intensity-less  conation. 
The    absolute    dualism    between    the   worth   and   hedonic   element,    as   described   in 
chapter  III,  p.  76,  is  unnecessary  if  these  distinctions  are  interpreted  genetically. 


92  Valuation:   its  Nature  and  Laws 

But  when  this  violence  is  done,  the  superficiality  of  the  analysis 
immediately  appears.  Can  we  say  that  sadness  is  pure  passive 
unpleasantness  ?  Certainly  not.  Already  in  the  relatively  passive 
state  of  sadness  we  have  the  preliminary  stage  of  the  accommod 
ative  reaction,  the  vital  series.  This  is  to  be  found  in  the 
expansiveness  of  the  feeling.  The  concentration  of  images  in 
this  phase  of  brooding  sadness,  the  tendency  of  the  feeling  to 
expand,  contains  already  an  immanent  activity,  differing  only  in 
degree  from  succeeding  phases  of  more  explicit  conation.  The 
fact  of  the  matter  appears  to  be  that  feeling  seems  to  be  mere 
feeling,  and  passive,  only  when  we  separate  it  retrospectively 
from  the  functional  whole,  the  vital  series  of  which  it  is  the  first 
phase.  Prospectively,  in  the  first  phase  of  expanding  feeling,  is 
already  contained  a  sense  of  the  strength  and  extent  of  the  con- 
ative  system  arrested,  and  this  feeling  passes  without  a  break  over 
into  the  relatively  more  active  emotions,  desire  and  will,  acts 
which  follow  as  the  arrest  increases  in  strength  and  duration. 
From  the  standpoint  of  the  later  stages,  the  initial  feeling,  viewed 
as  a  cause,  seems  relatively  passive. 

If,  on  the  other  hand,  we  seek,  as  some  do,  to  find  the  point 
of  distinction  between  the  more  active  feelings  and  decision  at 
the  end  of  the  series,  the  only  point  of  difference  that  we  can 
find  is  again  an  end-feeling  of  relaxation.  The  origin  of  this 
end-feeling,  and  of  the  characteristic  sensations  which  go  with  it, 
is  to  be  found  in  the  simple  fact  that  the  general  disturbance, 
displayed  in  the  series  of  emotions  preceding  the  moment  of  de 
cision,  has  found  a  definite  motor  channel  in  some  specific  bodily 
movement  or  word-formation.  But  to  separate  this  final  phase, 
this  end-feeling,  from  the  feelings  which  precede  it,  is  again  to 
give  us  a  mere  torso,  an  unreal  abstraction.  The  entire  vital 
or  worth  series  is  one,  with  a  continuity  of  affective-volitional 
meaning.  Each  phase  may  be  interpreted  as  conation  or  feeling 
according  to  the  point  of  view  from  which  it  is  observed. 

The  consideration  of  these  two  attempts  to  mark  off  the 
active  and  passive  aspects  of  experience — to  differentiate,  in 
terms  of  elementary  content,  the  affective  and  conative  phases 
of  a  total  vital  worth  series,  shows  that  such  an  effort  must 
prove  unsuccessful.  If  we  abstract  from  the  meaning  which 
the  attitude  has  by  virtue  of  its  place  in  such  a  series,  the  dis 
tinction  between  active  and  passive,  and  with  it  that  between 
affection  and  conation,  lapses.  We  have  in  these  conclusions 
therefore,  without  further  analysis,  the  grounds  for  our  negative 


Psychological  Basis  of  a   Theory  of  Valuation         93 

position  with  regard  to  the  dualistic  theories  of  feeling  and  will 
which  find  the  worth  element  in  feeling  conceived  as  passive 
pleasantness-unpleasantness  or  in  desire,  and  for  our  criticism 
of  any  conception  of  causal  determination  between  them.  They 
afford  positive  grounds,  moreover,  for  our  definition  of  worth  as 
"  affective- volitional  meaning,"  and  for  the  view  that  the  worth 
experience  is  a  concrete  feeling-attitude,  in  which  conation  is 
always  present  and  conative  dispositions  always  presupposed.1 

4.   Interpretation   of  the   Monistic    Theory — Us    Relation    to    the 
Definition  and  Analysis  of  the  Consciousness  of  Value. 

Nevertheless,  while  this  duality,  this  distinction  between 
feeling  and  will,  is  not  one  of  elementary  content,  it  is  still  a 
duality  of  meaning  which  becomes  fundamental  from  the  appre 
ciative  point  of  view.  Feeling  and  will  are  two  meanings  of  the 
same  content ;  but  what  determines  the  difference  in  meaning  ? 
How  is  this  differentiation  to  be  understood  ?  Our  answer  to 
this  question  must  be  in  the  general  terms  of  the  "  Identity 
theory,"  that  is,  that  the  difference  can  be  described  only  in  con 
ceptual,  logical,  retrospective  terms.  By  this  it  is  meant — to 
make  the  general  statement  more  specific — that  this  duality 
or  distinction  is  one  of  recognitive  and  selective  meaning. 
The  passive  or  active  meaning  is  one  which  the  attitude  gets  by 
reason  of  its  place  in  the  vital  series,  and  one  which  becomes  ex 
plicit  only  when  the  attitude  is  viewed  in  relation  to  preceding 
or  succeeding  phases  of  the  series.  They  are  differences  of  genetic 
mode. 

If  we  seek  to  describe  retrospectively  these  two  modes, 
if,  in  other  words,  we  seek  to  convey  their  internal  meaning, 
after  the  fact,  we  find  that  we  can  do  so  only  in  terms  of  cog 
nition,  by  description  of  the  cognitive  presuppositions  of  the 
attitudes.  According  to  Wundt,  the  special  aspect  which  an 
emotion  must  assume  to  become  volition,  is  an  immanental  Zweck- 

1  It  is  interesting  to  note  that  in  a  recent  article,  "The  Nature  of  Conation  and 
Mental  Activity"  (The  British  Journal  of  Psychology,  Vol.  II,  Part  I),  Stout,  while 
defining  conation  c<  as  a  complex  experience  "  which,  however,  contains  as  one  of  its 
elements  "a  simple  and  unanalysable  element  uniquely  characteristic  of  it — an  element 
from  which  the  whole  derives  its  distinctively  conative  character"  (which  he  describes  as 
felt  tendency,  and  which  is  not  identical  either  with  motor  sensations  or  affection),  never 
theless  admits  that  this  felt  tendency  and  affection,  though  distinguishable,  do  not  occur 
separately,  and  he  proposes  to  use  the  term  "interest"  to  express  the  unity  of  conative 
and  affective  characters  in  the  same  process.  I  cannot  see  that  this  view  differs  essen 
tially  from  the  one  developed  here.  As  analysed  by  Stout,  these  two  aspects  are  retro 
spective  abstractions. 


94  Valuation:    its  Nature  and  Laws 

richtung,  and  this  aspect  can  be  understood  only  as  change  in 
cognitive  attitude,  not  in  content.  The  attempt  of  Miinsterberg 
to  characterise  the  distinction  is  also  instructive  in  this  connection. 
"  In  feeling,"  he  says,  "  an  object,  independent  of  us,  is  inter 
preted  through  conation  (Trieb).  .  .  .  This  Trieb  remains,  however, 
as  overtone  and  as  a  help  in  apperception  of  the  object,  thought  of 
as  independent,  which  we  judge  in  feeling.  If  we  make  the  object 
dependent  upon  us,  so  that  we  perceive  it  as  retained  or  excluded, 
then  we  experience  conation  and  impulse,  but  not,  properly 
speaking,  a  feeling."1  Now,  to  make  the  object  dependent 
upon  us  is  to  assume  its  existence  or  non-existence,  as  the  case 
may  be — that  assumption  being  motived  by  a  subjective 
disposition  presupposed.  To  think  it  as  independent  of  us, 
which,  according  to  Miinsterberg's  analysis,  we  do  when  we 
feel  rather  than  desire,  is  again  to  judge  or  assume  its  existence 
or  non-existence,  but  the  motivation  of  the  cognitive  act  is  in 
this  case  a  control  of  a  more  objective  origin  and  character. 
The  significance  of  this  analysis  is  to  be  found  in  the  fact  that 
the  distinction  between  feeling  and  conation  is  one  which, 
in  the  last  analysis,  is  reducible  to  a  difference  in  the  immediate 
functional  meaning  of  elementary  content,  and  that,  when  this 
meaning  is  retrospectively  described,  such  description  involves 
recourse  to  cognitive  presuppositions. 

There  can  be  no  doubt,  however,  that  this  difference  in  im 
mediate  functional  meaning,  though  retrospectively  describable 
only  in  terms  of  cognitive  attitude,  is  really  implicitly  present 
prior  to  explicit  cognitive  acts  of  judgment  and  assumption, 
and  that  below  the  level  of  worth  experience  this  duality  has  its 
germs  in  the  simplest  types  of  organic  accommodation  and  habit. 
The  "  dependence  upon  "  or  "  independence  of "  subjective 
control,  which  on  the  higher  level  is  explicitly  cognised  in  acts 
of  judgment  and  assumption,  is  implicitly  felt  in  the  funda 
mental  attitudes  of  habit  and  accommodation  after  disturbance  of 
habit.  If  then  we  view  in  this  more  external  way  such  a  vital  ac 
commodative  series  as  that  described  by  Brentano,  we  find  that 
what  distinguishes  the  phases  which  are  predominantly  affective 
from  those  predominantly  conative  is  the  degree  of  inhibition 
of  a  presupposed  disposition  or  tendency.  Whether  we  call  the 
phase  in  question  feeling  or  will  depends  upon  the  point  in  the 
process  of  accommodation  in  which  we,  so  to  speak,  catch  the 
experience.  In  Brentano's  series  the  first  stages  are  characterised 

1  Grundzuge  der  Psychologie,  p.  366. 


Psychological  Basis  of  a   Theory  of  Valuation         95 

by  the  apprehension  of  the  object  as  relatively  independent  of 
the  subject  (in  this  case  the  apprehension  is  judgmental) — and 
in  introspection  they  are  interpreted  as  feeling;  in  the  later 
stages  the  object  is  apprehended  as  more  and  more  dependent, 
until  in  the  last  phases  the  belief  or  judgment  that  the  desire  will 
be  accomplished  enters  and  voluntary  decision  has  been  reached. 
Likewise,  when  Wundt  describes  the  relation  in  the  statement 
that  "  feeling  may  just  as  well  be  looked  upon  as  the  beginning 
of  a  conative  process,  as  on  the  other  hand  will  may  be  con 
ceived  as  a  complex  feeling-process,  and  that  the  emotion  is  a 
transition  between  both/'  he  is  distinguishing  different  phases 
of  one  accommodative  process. 

With  this  conception  of  the  nature  of  the  duality  in  meaning 
of  feeling  as  passive,  and  desire  or  volition  as  active,  we  are 
in  a  position  to  justify  our  definition  and  analysis  of  worth-ex 
perience.  Feeling  and  desire  are  differences  of  genetic  mode, 
relative  differences  of  functional  meaning,  not  of  content.  The 
worth  of  an  object  is  therefore  its  affective- volitional  meaning, 
and  is  given  in  feeling- attitudes  in  which  there  is  always  an  ele 
ment,  transgredient  or  immanental,  of  conation.  We  describe  the 
worth-fundamental  as  feeling  or  concrete  emotion,  because  pure 
passive  affection  and  pure  active  conation  are  limiting  terms  in 
the  series,  and  really  exist  merely  as  abstractions.  But  the 
affective-volitional  meaning,  or  worth,  of  an  object,  its  rela 
tion  to  desire  and  conative  disposition  as  interpreted  through 
feeling,  becomes  explicit  only  on  the  cognitive  level  where  ac 
commodation  is  in  the  form  of  cognitive  acts  of  presumption, 
assumption,  and  judgment.  It  is  the  actualisation  of  the  dis- 
positional  tendency,  either  in  feeling  or  desire,  through  these 
cognitive  acts,  which  gives  to  the  feeling  or  desire  that  meaning 
described  as  worth.- 

In  conclusion  it  may  be  pointed  out  that  in  this  conception 
of  the  nature  of  feeling  and  will  and  of  their  relations  we  have 
a  psychological  basis  for  the  study  of  the  laws  of  valuation. 
The  concrete  laws  of  valuation  are  not  reducible  to  general  laws 
of  feeling,  abstracted  from  conation,  nor  of  desire  abstracted 
from  feeling,  but  rather  of  affective-volitional  process  conceived 
as  a  whole.  If  we  apply  the  term  interest,  employed  by  Stout 
in  the  connection  already  referred  to,  to  designate  the  conative 
process  in  its  twofold  aspect,  we  may  quite  properly  speak 
of  these  laws  of  acquirement  of  affective- volitional  meaning  as 
laws  of  interest. 


96  Valuation  :    its  Natiire  and  Laws 

II.    FURTHER  ANALYSIS  OF  FEELING — THEORY  OF  THE 
NATURE  AND  RELATION  OF  ITS  DIFFERENT  ASPECTS 

i.  Structural  Analysis. 

The  theory  of  the  relation  of  feelings  and  will  already  de 
veloped  has  given  us  a  psychological  basis  for  the  definition  of 
worth  experience  as  feeling  with  certain  presuppositions.  Our 
second  problem,  the  further  analysis  of  feeling  itself,  arises 
from  the  fact  that  the  appreciative  distinctions  already  made — 
between  the  different  modes  of  feeling,  between  the  feeling-tone 
of  sensation  and  the  meanings  of  feeling-attitudes,  which  are 
alone  suggestive  of  worth,  between  intensity  of  feeling-tone  and 
depth  and  breadth  of  feeling,  all  demand  some  further  theory 
of  the  nature  of  feeling  itself.  In  our  previous  analysis  of  these 
appreciative  descriptions  we  have  already  suggested  certain 
elements  of  such  a  theory  which  must  now  be  developed  in 
more  detail. 

(a)  Feeling  as  a  Kind  of  Sensitivity. 

We  have  said  that  the  concrete  psychoses  which  are  subsumed 
under  the  general  terms  feeling  and  will — emotion,  passion, 
sentiment  and  mood,  impulse  and  desire — are  fundamentally 
different  meanings  of  the  same  general  content.  Can  we  specify 
more  completely  the  nature  of  this  content  ?  A  consideration 
of  the  attempts  to  answer  this  question  may  well  be  preceded 
by  an  explicit  recognition  of  the  fact  that,  just  as  the  concepts 
of  feeling  and  will  are  abstractions  from  a  concrete  whole  of 
meaning  which,  as  such,  can  be  only  appreciatively  described, 
so  any  attempt  to  reconstruct  the  appreciative  difference  of 
feeling  in  terms  of  content,  i.e.,  in  terms  of  non-appreciative 
description,  must  involve  a  process  of  abstraction  which  makes 
one  aspect  of  the  total  complex  do  duty  for  the  whole.  Simmel 
has  well  said  of  the  use  of  pleasure  as  synonymous  with  feeling 
that :  it  rises  as  a  concept,  as  a  separate  content  of  conscious 
ness,  only  after  its  manifold  real  characteristics  (Ausgestal- 
tungen)  have  determined  our  actions,  and  after  precisely  these 
differences  have  had  their  effect.1  But  when  this  fact  is  fully 
recognised,  it  still  remains  true  that  one  abstraction,  one  equiva- 

1  Simmel,  Einleitung  in  die  Moralwissenschafl,  Vol.  I,  p.  307  :  "  Lust  taucht  als 
Begriff,  als  gesomlerter  Bewustseinsinhalt  erst  auf,  nachdem  ihre  realen  Ausge- 
staltungen  tausendfach  das  Zweckhandeln  beherrscht  und  nachdem  gerade  die  Fiille 
und  Verschiedenheit  dieser  dazu  angeregt  hat,  unter  gegenseitiger  Verdunklung  jenes 
Verschiedenen  dem  Gemeinsamen  davon  eine  besondere  Beleuchtung  zu  verleihen." 


Psychological  Basis  of  a   Theory  of  Valuation        97 

lent  may  be  more  useful  for  its  purposes  than  another.  The  chief 
purpose  of  a  theory  of  feeling  stated  in  terms  of  mental  elements 
is,  to  my  mind,  and  according  to  the  general  principles  already 
laid  down,  its  use  as  an  instrument  in  the  interpretation  of  the 
functional  relation  of  the  general  aspect  of  experience  which  we 
describe  as  feeling  to  that  other  aspect  which  we  call  cognition  ; 
and  while  from  the  point  of  view  of  our  special  study  there  are 
many  questions  of  analysis  which  may  be  left  unconsidered, 
nevertheless,  we  cannot  avoid  altogether  the  formulation  of 
some  working  conception  which  will  serve  this  special  purpose. 

All  theories  are  agreed  that  feeling  is  the  peculiarly 
subjective  aspect  of  experience,  in  contrast  to  the  objective, 
cognitive,  and  it  has  been  proposed  that  the  term  feeling  be 
applied  in  a  broad  sense  to  the  subjective  fringe  of  all  cognitive 
experience.  But  the  question  arises  immediately  just  how 
subjective — and  we  have  answers  varying  all  the  way  between 
the  extremes  of  those  who  hold  that  feeling  is  entirely  unpre 
sentable  and  is  never  found  on  the  objective  side  of  the  equation, 
and  those  who  find  in  it  quasi-cognitive  functions  such  as  memory 
and  generalisation.  Closely  connected  with  this  is  the  question 
of  the  degree  of  independence  of  feeling  of  the  cognitive  aspects 
of  experience,  of  sensation  and  image  content. 

Without  going  too  fully  into  minor  distinctions,  we  may 
in  general  distinguish  three  main  theories  of  the  structural 
nature  of  feeling.  The  first  of  these  is  that  feeling  is  not  content, 
but  merely  the  affective  tone  of  content,  sensational  and  re 
presentational.  But  on  this  view  it  is  necessary  to  distinguish, 
as  Stout  has  done,  the  affective  tone  of  content,  sensational  and 
ideal,  from  the  affective  tone  of  process,  perceptual  and  idea- 
tional.  The  fundamental  aspect  of  emotion,  sentiment,  and 
mood  is,  then,  the  feeling-tone  of  the  process,  positive  or  negative, 
according  as  it  is  facilitated  or  arrested  (the  feeling-tone  of 
sensations  and  images,  the  more  objective  peripheral,  and  the 
more  subjective  organic,  entering  as  secondary  qualifying 
elements,  through  surplus  excitation).  The  second  theory, 
Wundt's,  is  like  the  preceding  in  distinguishing  feeling  completely 
from  sensitivity,  but  differs  in  the  fact  that  it  conceives  feeling  to 
be  a  special  mental  element,  on  the  analogy  of  sensation.  These 
elements  have  three  attributes,  pleasantness-unpleasantness, 
tension-relaxation,  and  restlessness-quiescence,  distinguished  in 
the  preceding  chapter.  Total  feelings  are  then  complexes  of 
these  hypothetical  elements.  The  third  theory,  of  which 


98  Valuation :   its  Nature  and  Laws 

Miinsterberg  may  be  taken  as  representative,  holds  that  these 
are  appreciative  descriptions,  and  that  when  we  turn  the  feeling- 
attitude  into  an  object  of  non-appreciative  description,  only 
sensational  elements  can  be  discovered.  A  psychosis  called 
feeling  is  then  but  a  complex  of  sensations  more  difficult  of 
analysis  than  other  more  objective  complexes. 

Leaving  specific  criticisms  of  the  first  two  theories  to  be 
developed  in  the  course  of  the  discussion,  I  find  certain  general 
considerations  which  lead  to  a  modified  form  of  the  last  theory. 
In  the  first  place  the  non-sensational  aspects  of  any  experience 
are,  in  the  very  nature  of  the  case,  only  appreciatively 
describable  in  functional  terms.  In  the  second  place  the  dis 
tinction  between  affectivity  and  sensitivity  cannot,  I  think, 
be  as  completely  drawn  as  is  presupposed  in  the  first  two  con 
ceptions.  And,  finally,  properly  combined  with  the  functional 
genetic  conception  of  feeling  and  will  already  developed,  it 
seems  adequate  to  account  for  all  the  modifications  of  feeling- 
attitude  which  we  have  found  significant  for  a  theory  of  feeling 
and  of  worth  experience. 

The  breadth  of  the  use  of  the  word  feeling  in  ordinary  speech 
has  been  frequently  commented  upon  in  recent  discussions. 
We  feel  a  sensation,  an  emotion,  a  mood,  or  sentiment.  We 
feel  darkness  and  distance,  the  remoteness  or  nearness  of  things 
in  memory.  We  feel  an  impulse,  a  desire,  and  we  feel  determined 
to  do  a  thing.  If  we  start  with  the  last  use  of  the  term,  we 
find  that,  apart  from  the  functional  place  of  the  attitude  in 
a  vital  accommodative  series,  and  the  nature  of  its  cognitive  pre 
suppositions,  there  is  nothing  to  distinguish  the  feeling  of  desire  or 
will  (i.e.,  of  being  determined)  from  the  feeling  of  an  emotion  or 
mood,  except  the  character  of  the  sensational  content  in  the  two 
experiences.  In  the  case  of  the  desire  or  feeling  of  determination 
there  are  certain  kinaesthetic  or  motor  sensations  which  are  quite 
distinct  from  the  organic  and  systemic  sensations  which  qualify 
the  attitude  which  we  describe  as  emotion,  sentiment,  or 
mood.  The  difference  in  the  "  feel  "  of  the  two  kinds  of  atti 
tudes  is  adequately  describable  in  terms  of  these  sensations  and 
their  combinations.  In  the  vital  series  some  distinguishable 
feeling-attitude  has  normally  preceded  (not  always,  however, 
as  we  have  seen)  the  desire  or  act  of  will,  and  its  meaning  is 
taken  up  into  the  more  explicit  volition  ;  but  the  added  content 
which  comes  with  the  volitional  attitude  gives  a  distinguishing 
quality  to  the  psychosis.  Moreover,  in  the  case  of  explicit 


Psychological  Basis  of  a   Theory  of  Valuation        99 

desire  or  volition  the  subjective  sensational  content  tends  to 
usurp  consciousness,  and  the  perception  of  the  object  is  a  second 
ary  element  in  the  total  complex. 

In  the  coarser  emotions,  so  called,  the  sensational  elements 
stand  out  almost  as  clearly  as  in  the  case  of  desire.  In  these, 
in  very  truth,  the  "  body  "  of  the  emotion  is  the  mass  or 
series  of  sensations,  mostly  organic,  but  also  partly  motor, 
since  the  emotion  can  never  be  completely  distinguished  from 
the  desire.  But  the  "  soul  "  of  the  emotion,  to  use  a  correlative 
term,  its  meaning,  is  found,  as  in  the  case  of  the  volition,  in 
its  place  in  the  vital  series,  in  its  cognitive  presuppositions. 
When,  however,  we  seek  to  describe  this  meaning,  we  can  do  so 
only  by  having  recourse  to  appreciative  terms — we  describe  its 
positive  or  negative  directions,  its  dynamic  and  expansive 
suggestions,  etc. 

The  preceding  attitudes  are  analysable  as  psychoses,  because 
their  content  usurps  consciousness  even  to  the  partial  exclusion 
of  the  objects  toward  which  they  are  directed,  and  because  they 
represent  the  result  of  marked  inhibition  of  conative  tendency. 
When  we  turn  to  those  "finer"  phases  of  feeling,  the  sentiments 
and  moods  which  attach  to  ideal  objects  and  the  "feeling- 
tone  "  of  sensational  and  perceptual  objects,  we  find  the  object 
upon  which  they  are  directed,  or  to  which  they  are  attached, 
more  in  the  ascendent.  Those  sensational  experiences  which  in 
the  coarser  emotions  stood  out  in  all  their  discreteness,  now  fall 
into  the  background,  and  are  recoverable  only  in  case  the  feeling 
flares  up  again  through  arrest,  into  a  coarser,  more  fully  embodied 
meaning.  What  shall  we  say  of  the  structure  of  these  finer 
states  ?  I  cannot  see  that  we  have  any  other  alternative  than 
to  say  that,  as  content,  the  feeling  is  the  same  sensation  mass, 
but  so  reduced  in  intensity  as  to  be  for  practical  purposes  in 
separable  from  the  objective  cognitive  content,  or  so  completely 
fused  as  to  forbid  analysis  of  the  separate  elements.  These 
finer  feelings  seem,  in  contrast  to  the  coarser  forms,  to  lead  a 
disembodied  existence,  or,  better  still  perhaps,  they  become 
embodied  in,  or  fuse  with,  other  presentational  material  such  as 
the  motor  or  auditory  resonance  of  a  word,  visual  or  tonal  com 
plexes,  in  nature,  human  expression,  and  art.  We  speak  then 
of  the  feeling-tone  of  the  object,  or  describe  the  object  as  suffused 
with  feeling. 

Finally,  there  is  a  group  of  experiences,  of  which  we  shall  have 
more  to  say  presently,  in  which  the  affective  state  is  scarcely 


ioo  Valuation :   its  Nature  and  Laws 

distinguishable  from  the  cognitive.  The  affective  aspect,  the 
subjective  reference,  is  so  sublimated  that  it  is  little  more  than 
an  accompaniment  of  the  cognitive  process.  Speaking  figuratively, 
we  may  perhaps  describe  them  as  subtle  tones  which,  in  ordinary 
unimpeded  mental  process,  take  the  place  of  explicit  fuller  em 
bodiment  of  desire,  emotion,  and  sentiment,  and  call  them 
"  affective  signs."  The  so-called  feelings  of  relation  are  of  this 
character— the  feeling-tones  which  attach  to  conjunctions,  ad 
verbs,  etc.,  or  to  the  moods  of  the  verb,  I  shall,  I  can,  I  will,  I 
ought,  I  must.  These  affective  tones  are  all  sublimated  forms 
of  explicit  emotion  and  conation,  and,  with  sufficient  arrest  in 
the  ideational  process  in  which  they  occur,  may  flare  up  into  the 
explicit  embodied  feeling  of  which  they  are  the  signs,  with  all 
its  characteristic  sensational  content.  They  have  consequently 
been  described  as  vestigial  phenomena,  survivals  of  former 
motor  and  organic  attitudes.  It  is  to  be  noted  too  that,  while 
normally  they  are  the  affective  tones  of  words,  they  may,  never 
theless,  appear  prior  to  and  independently  of  word  formation. 
For  the  sake  of  completeness  it  should  be  added  that  the  affective 
tone  of  abstract  and  general  terms,  the  affective-volitional 
meaning  acquired  in  processes  of  ideal  construction,  and  accom 
panied  by  explicit  desires  and  emotions,  are  of  the  same  general 
character.  But  of  this  more  later. 

(b)    The  Appreciative  Distinctions  in  Feeling  as  "Forms  of 
Combination"  of  the  Elements. 

To  this  theory — that  feeling,  when  viewed  as  content,  is  a 
form  of  sensitivity,  serious  objections  have  been  raised.  Es 
pecially  these  finer  forms  of  affectivity,  which  seem  to  be  with 
out  embodiment  in  analysable  muscular  or  organic  sensations, 
raise  a  doubt  as  to  the  completeness  of  our  analysis.  And, 
indeed,  we  may  well  question  its  adequacy,  as  it  stands,  without 
any  further  modification,  to  account  for  the  different  qualifica 
tions  of  feeling  as  differentiated  by  appreciative  description.  Is 
it  not  necessary,  after  all,  as  for  instance  in  Wundt's  theory,  to 
assume  special  feeling  elements  to  account  for  these  differences  ? 

Both  Wundt  and  Stout  have  criticised  this  theory  of  emo 
tions  on  the  ground  of  the  difficulty  of  accounting  for  the 
qualitative  differences  in  emotion  merely  as  differences  in 
sensation  quality  of  the  muscular  and  organic  reflexes,  and 
because  of  its  being  viewed  as  an  effect  rather  than  as  a  cause 


Psychological  Basis  of  a   Theory  of  Valuation       101 

of  the  organic  resonance.  The  differences  in  motor  and 
organic  reaction,  and  therefore  in  sensation  complexes,  con 
sistent  with  the  realisation  of  the  same  subjective  emotional 
state,  necessitate,  it  is  said,  our  looking  elsewhere  for  the  dis 
tinctive  content  of  emotional  psychoses.1  In  addition,  it  is 
held  that  such  a  theory  wholly  ignores  in  its  explanation  of  the 
emotions  the  conative  tendencies  which  they  presuppose.  "  An 
emotion  involves  a  certain  trend  or  direction  of  activity  which 
particularises  itself  in  any  way  it  can,  according  to  circumstances." 
These  criticisms,  it  may  be  admitted,  are  in  the  main  sound, 
but  the  modification  which  the  theory  requires  to  make  it  adequate 
is  one  which  presents  itself  almost  immediately  when  one  views  an 
emotion  in  its  aspect  as  content.  A  closer  analysis  indicates  that 
what  gives  such  a  psychosis  its  specific  quale  is  not  the  separate 
sensation  qualities,  nor  yet  their  fusion  in  an  indistinguishable 
mass,  but  rather  the  structural  relations  among  these  elements. 
In  addition  to  the  qualities  of  the  sensations,  there  are  certain 
temporal  and  intensity  relations  among  them.  Every  emotion  has, 
in  addition  to  the  object  toward  which  it  is  directed,  the  pre 
suppositions  which  give  it  its  meaning,  and  its  positive  and 
negative  direction,  its  own  peculiar  organic  resonance  with 
its  own  specific  character  due  to  the  "form  of  combination"  of 
the  temporal  and  intensity  aspects  of  the  elements. 

This  concept  of  a  new  specific  quale,  or  form  of  combination, 
giving  rise  to  new  qualities  which  may  be  appreciatively  dis 
tinguished,  seems  to  have  become  a  permanent  feature  of  theories 
of  emotional  complexes,  irrespective  of  their  conception  of  the 
nature  of  the  elements  in  those  complexes.  Wundt,  who  holds 
that  these  elements  are  hypothetical  feelings,  differing  in  toto 
from  sensitivity,  reconstructs  the  different  complexes  as  different 
forms  of  Gefuhlsverlauf  among  the  elements.  Witasek,  who 
holds  that  these  "  movement  -  forms "  are  forms  of  com 
bination  of  sensational  elements,  conceives  that  the  Gestalt- 
qualitdt  thus  formed  becomes  the  presentable  aspect  of  the 
total  affective  complex.  It  is  not  necessary  to  reopen  the  dis 
cussion  between  the  sensation-element  and  affective-element 
theories.  For  the  point  we  are  concerned  with  here,  that  ques- 

1  Moreover,  difficulties  arise  from  the  genetic  point  of  view.  As  has  been  pointed 
out  by  Stout,  the  same  emotion— for  instance,  fear — may  arise  in  connection  with  very 
different  motor  expressions,  even  in  the  same  animal,  showing  clearly  that  the  specific 
quale  of  an  emotional  psychosis  is  something  which  may  remain  constant  notwithstand 
ing  considerable  variation  in  the  qualities  of  organic  and  muscular  sensations  following 
upon  motor  expression. 


IO2  Valuation:   its  Nature  and  Laws 

tion  is  relatively  unimportant.  The  point  of  real  importance 
is  that  such  form-qualities  exist,  and  that  they  are  founded  on 
the  temporal  and  intensity  relations  of  the  elements.1 

These  form-qualities,  moreover,  as  has  been  clearly  shown 
by  Witasek,  may  be  transposed  from  the  organic  sensitivity 
in  which  they  first  inhere  to  presentational  and  ideal  content. 
Many  of  the  "finer"  sentiments  are  without  embodiment  in 
sensational  content,  and  therefore  without  this  emotional 
intensity.  They  consist  wholly  in  the  "  form  of  combination  " 
of  cognitive  content,  which,  however,  still  retains  the  meaning 
of  former  fully  embodied  emotions.  Of  this  fact  we  shall 
make  further  use  in  our  study  of  the  processes  of  sym 
pathetic  Einfithlung  :and  the  new  meanings  and  values  which 
arise  in  those  processes.  Sympathetic  projection  of  a  feel 
ing-attitude  involves  presentation  of  the  psychical,  in  this 
case  the  peculiarly  subjective  aspect  of  experience  described 
as  feeling.  Now,  of  course,  the  hedonic  intensity  and 
the  cognitive  presuppositions  of  the  feeling-attitude  can  be 
projected  only  conceptually,  as  objects  of  judgment,  assump 
tion,  etc.  But  this  does  not  exclude  the  possibility  of  the 
intuitive  presentation  of  the  form  of  the  psychosis,  and,  there 
fore,  of  its  specific  quale.  For,  if  the  form  of  an  emotion  is  the 
system  of  temporal  and  intensity  relations  among  the  elements 
of  the  complex,  this  aspect  can  be  abstracted  from  the  elements 
precisely  as  any  other  form  of  combination  (e.g.,  rhythm),  and 
transposed  to  other  elements.  This  is  precisely  what  happens, 
and,  when  we  come  to  the  study  of  sympathetic  projection 
(Einfuklung)  in  detail,  we  shall  find  that  the  form-qualities  of 
objects,  persons,  and  things  constitute  the  inducing  conditions 
for  this  intuitive  realisation  of  the  emotion,  as  when  the  sighing 
of  the  wind  or  the  bodily  or  vocal  expression  of  a  person  expresses 
for  our  intuitive  perception  the  particular  emotion  in  .-  iestion. 

Moreover,  on  this  hypothesis  it  is  possible  to  understand  how 
the  peculiar  resonance  of  an  emotion  may  remain  unmodified, 
notwithstanding  great  changes  in  the  qualities  of  the  separate 
sensations  and  in  the  absolute  intensity  of  the  emotion  viewed 
as  a  total  complex.  As^on  the^more  objective  presentational 

1  Wundt — we  may  merely  note  in  passing — has  made  some  progress  in  the  direction 
of  classifying  these  various  movement  -  forms :  (i)  The  rapidly  rising  and  slowly 
falling ;  (2)  the  slowly  rising  and  rapidly  falling ;  (3)  the  intermittent  ;  and  (4) 
the  oscillating,  and  in  subsuming  the  affects  appreciatively  distinguishable  under  these 
rubrics.  The  most  fruitful  principle  of  classification  is,  I  believe,  to  be  found  in  this 
Direction,  but  with  the  actual  classification  we  are  here  only  remotely  concerned. 


Psychological  Basis  of  a   Theory  of  Valuation       103 

side,  the  form  of  combination  of  the  temporal  and  intensity 
aspects  of  sensational  elements,  as,  for  instance,  a  rhythm,  may 
be  transposed  from  one  set  of  elements  to  another,  and,  provided 
the  relative  intensities  and  temporal  relations  are  unimpaired, 
remain  entirely  clear  and  distinct  with  dampening  of  the  in 
tensity  of  the  total  complex — so  here,  even  on  the  assumption 
that  affectivity  is  a  form  of  sensitivity  with  a  peculiarly  sub 
jective  meaning,  we  may  see  how  the  specific  quale  of  the  emotion 
may  still  be  accounted  for. 

In  conclusion,  then,  we  may  say  that  a  complete  analysis 
of  any  total  affective  attitude  must  differentiate  four  aspects  : 
(i)  its  positive  and  negative  direction  ;  (2)  its  presuppositions, 
dispositional  and  actual ;  (3)  its  sensation  content ;  and  (4) 
the  form  of  combination  of  that  content,  or  this  quale  transposed 
to  more  objective  cognitive  content,  of  which  it  becomes  then 
merely  the  affective  over- tone. 


2.  Correlation  of  Structural  and  Functional  Analysis 

of  Feeling. 

» 

The  outcome  of  this  structural  analysis  of  the  different 
phases  of  experience  which  we  group  under  the  general  term 
feeling,  is  the  view  that  the  different  modifications  of  feeling 
are  meanings  of  a  certain  type  of  sensitivity.  The  significance 
of  this  conception  lies  in  the  fact  that,  while  the  essence  of  feeling 
is  a  functional  meaning,  it  is  an  embodied  meaning,  and  we  may 
therefore  expect  different  modifications  of  that  meaning  to  be  cor 
related  with  significant  changes  in  this  sensitivity.  These  changes 
in  meaning,  with  their  correlated  changes  in  sensitivity,  will, 
moreover,  in  all  probability  be  connected  with  changes  in  func 
tional  presuppositions.  The  working  out  of  these  relations 
would  enable  us  to  correlate  our  structural  analysis  and  classi 
fication  with  the  earlier  genetic  analysis  of  the  accommodative 
series. 

(a)  Changes  in  Functional  Meaning  and  Changes  in  Sensitivity- 
Passion  and  Emotion-Sentiment  and  Mood. 

The  duality  of  affection  and  conation  has  been  interpreted 
as  a  differentiation  of  meaning  determined  by  the  place  of 
the  psychosis  in  the  accommodative  series.  Structurally, 
the  emotion  is  seen  to  be  a  shortened  form  of  desire  and  volition, 


IO4  Valuation:   its  Nature  and  Laws 

in  which  organic  largely  takes  the  place  of  motor  sensitivity. 
In  other  words,  from  the  standpoint  of  structural  analysis, 
the  nearer  a  psychosis  approaches  to  explicit  desire  and  vo 
lition,  the  more  pronounced  becomes  that  special  phase  of 
sensational  content  which  we  call  strain  and  effort.  It  is  in 
these  phases  that  we  find  inhibition  of  conative  disposition 
or  tendency  at  a  maximum. 

But  there  are  other  modes  of  feeling,  distinguished  by  psy 
chological  analysis,  which  represent  various  stages  in  accommod 
ation,  and  which  show  corresponding  typical  changes  in  the 
sensitivity  of  which  they  are  the  meanings.  Certain  modes  of 
feeling,  finer  forms  of  affectivity,  such  as  sentiment  and  mood, 
are  characterised  by  the  fact  that  the  organic  sensations  with 
their  hedonic  intensity  become  less  and  less  important  while 
the  functional  meaning  remains  unimpaired.  There  are  even 
certain  phases  of  affective-volitional  meaning  in  which  the 
meaning  is  without  embodiment  in  analysable  organic  sensitivity, 
but  in  which  it  appears  merely  as  an  overtone  of  perception 
or  cognitive  activity  in  general.  Can  we  understand  these 
phenomena  in  terms  of  our  genetic  theory  ? 

(b)  Their  Genetic  Relations. 

In  a  general  way  it  may  be  immediately  seen  that  the  terms 
passion  and  emotion,  sentiment  and  mood,  stand  for  appreciable 
differences  in  meaning  of  affective  attitude.  One  love  is  best 
described  as  a  passion,  another  as  a  sentiment ;  one  fear  as 
an  emotion,  another  as  a  mood.  It  is  also  apparent  that  there 
are  genetic  relations  among  them.  Emotion  may  become 
fixed  as  a  mood,  passion  may  pass  over  into  a  sentiment.  Senti 
ment  may  flare  up  into  passion  and  mood  be  stirred  into  emotion. 
The  significance  of  these  appreciative  distinctions  is  to  be  found 
in  the  fact  that  they  represent  different  attitudes,  different 
modes  of  affective- volitional  meaning  of  the  object  for  the  sub 
ject  which,  as  in  the  case  of  Brentano's  series,  with  its  dis 
tinctions  between  feeling,  desire,  and  will,  go  back  ultimately 
to  differences  in  presuppositions.  In  fact,  the  transition  from 
passion  to  sentiment,  and  its  limiting  term  "  affective  sign," 
or  from  emotion  to  mood  and  "  affective  sign  "  constitutes  a 
vital  series  which  may  be  appreciatively  segregated  for  analysis 
and  description  in  terms  of  function  and  structure. 

A  singularly   superficial  view  of   the   nature   of   these   dis- 


Psychological  Basis  of  a   Theory  of  Valuation       105 

tinctions  reduces  them  to  differences  merely  in  degree  of  in 
tensity.  This  insufficiency  of  the  analysis  has  been  supplemented 
by  Paulhan's  study,  Les  phenomenes  affectifs  et  le  lois  de  leur 
apparition,  the  general  import  of  which  is  so  thoroughly  in 
harmony  with  our  study  that  it  may  be  stated  in  his  own 

terms. 

His  general  theory  is  that  affective  states  presuppose  conative 
tendencies  of  varying  degrees  of  strength  and  systematisation  ; 
that  the  actual  appearance  in  consciousness  of  an  affective  atti 
tude  is  due  to  incoordination  and  arrest  of  these  tendencies, 
and  that  in  these  affective  states  there  are  certain  aspects 
of  content  which  disclose  the  degree  of  systematisation  and 
arrest  presupposed  ;  that  the  modifications  of  attitude,  passion, 
sentiment,  emotion,  mood,  and  affective  sign  may  be  reduced 
to  differences  of  degree  in  which  these  aspects  are  present, 
and  therefore  to  differences  in  arrest  of  tendencies  presup 
posed.  Those  aspects  of  affective  disturbance  upon  the  basis  of 
which  a  genetic  and  functional  classification  is  thus  possible  are : 
(i)  the  intensity  or  vehemence  of  the  disturbance,  the  force  and 
persistence  of  the  arrested  tendency;  (2)  the  multiplicity  of 
phenomena  in  the  affective  state  (mass  of  sensational  content) 
disclosing  the  complexity  of  the  tendencies  presupposed;  (3) 
the  degree  of  the  tendency  of  the  disturbance  to  absorb  conscious 
ness  (which  we  may  interpret  as  indicating  the  degree  of  con 
centration  of  subsidiary  tendencies  about  the  fundamental). 
All  these  aspects  are  interpreted  as  functions  of  the  two  factors 
systematisation  and  arrest  of  conative  tendency. 

Having  established  the  general  fact  that  all  affective  attitude 
involves  at  least  a  minimum  of  arrest,  he  examines  the  different 
attitudes,  passion,  emotion,  sentiment,  mood,  etc.,  in  the  light 
of  the  preceding  analysis.  Passions  and  emotions  disclose  these 
aspects  at  a  maximum  and  represent  the  extreme  of  arrest.  On 
the  other  hand,  sentiments  and  moods,  derived  respectively 
from  the  passions  and  the  emotions,  show  these  aspects  in  a  less 
marked  degree,  and  represent,  therefore,  the  beginning  of  re- 
adaptation  after  arrest  or  the  reduction  of  the  moment  of  arrest. 
All  these  aspects  are,  however,  again  reproducible  in  passion 
and  emotion  when  the  conative  disposition  is  again  subject  to 
arrest.  On  the  functional  side,  then,  this  diminution  of  in 
tensity,  multiplicity,  and  absorption  of  consciousness  represents 
the  habit  which  comes  with  repetition. 

In  the  "affective  sign"  (a  new  descriptive  term  introduced 


io6  Valuation:   its  Nature  and  Laws 

by  Paulhan  for  affective  phenomena  not  ordinarily  observed  in 
superficial  introspection),  we  have  a  still  further  reduction  of 
these  aspects  of  the  affective  experience.  It  represents  a  limiting 
term  in  the  affective  or  vital  series,  hard  to  distinguish  from 
terms  in  an  intellectual  series.  Paulhan  describes  the  role  of 
these  affective  signs  in  the  following  way  :  "  We  may  describe 
more  clearly  perhaps  the  nature  of  this  class  of  facts  by  com 
paring  them  with  the  operations  of  the  cognitive  conscious 
ness.  We  know  that  cognitive  phenomena  are  often  substituted 
the  one  for  the  other.  Thus  the  image  replaces  the  sensation, 
ideas  or  words  are  able  to  replace  images.  The  substitu 
tion  takes  place  so  easily  that  one  is  not  as  a  rule  led  to  recog 
nise  it,  and  the  more  ultimate,  abstract,  and  feeble  substitutes 
have  consequently  been  but  rarely  studied  by  psychologists. 
These  substitutes  are  pure  abstracts,  produced  perhaps  by  the 
partial  excitation,  feeble  but  systematised,  of  a  large  number 
of  tendencies.  We  find  in  the  affective  sphere  facts  of  substitu 
tion  analogous  to  those  which  we  recognise  in  the  functioning 
of  intelligence.  Passion  and  sentiment  are  replaced  often  by 
other  states  of  consciousness  of  an  affective  nature,  which  be 
come  substitutes  for  them  and  fill  their  role."  x 

This  greater  capacity  of  the  affective-sign  for  substitution 
or  transference  from  one  fundamental  to  another  is  functionally 
its  most  significant  aspect,  and,  as  Paulhan  points  out,  this 
capacity  goes  with  the  fact  that  it  shows  intensity,  multiplicity 
of  content,  and  absorption  of  consciousness,  at  a  minimum. 
On  the  side  of  content  it  is  the  relative  absence  of  these 
aspects  which  marks  the  affective  sign  as  the  limiting  term  in 
the  vital  series  we  are  considering.  But  to  this  negative  aspect 
must  be  added  a  positive.  This  Paulhan  recognises  in  the  pe 
culiar  tone  which  the  sign,  although  relatively  intensity-less, 
gives  to  consciousness.  The  decrease  in  these  contentual  aspects 
does  not,  however,  mean  loss  in  worth  suggestion  or  affective- 
volitional  meaning.  The  affective  sign  has  a  functional  meaning 
which  does  not  lie  in  its  intensity. 

1  Les  phinomtnes  affcctifs,  etc.,  Paris,  1901,  p.  72.  He  thus  further  describes  their  func 
tion  in  consciousness  :  "  If  we  recall  an  affective  impression  with  strength  and  persistence, 
we  may  be  able  to  free  ourselves  from  it  partially,  as  we  would  not  be  able  in  the  case 

the  original  affective  disturbance,  but  we  feel  always  within  us,  not  the  original  im 
pression,  but  an  element  which  replaces  it  momentarily  and  which  gives  to  the  state  of 
consciousness  a  peculiar  tone."  This  tone  he  then  goes  on  to  describe  as  both  generic 
and  recognitive  in  its  suggestions.  This  is  a  true  piece  of  psychological  analysis  and 
introspection,  examples  of  which  will  be  given  and  discussed  in  more  detail  in  the 
following  chapter  on  affective  memory  and  affective  generalisation. 


Psychological  Basis  of  a   Theory  of  Valuation       107 

In  this  application  of  the  genetic  view  to  the  explanation 
i  of  the  relation  of  the  functional  meaning  of  feeling  attitudes  to 
!  their  sensitivity,  we  have  not  only  an  important  extension  of 
the  general  principle  that  the  differences  in  affective-volitional 
attitudes  are  reducible  to  differences  of  genetic  mode  in  a  vital 
accommodative  series,  but  also  the  basis  for  psychological  ex 
planation  of  certain  phenomena  of  valuation.  We  have  found 
that,  on  the  level  of  worth  experience,  accommodation  and 
habit  are  to  be  co-ordinated  with  cognitive  acts,  feeling  and 
desire  being  actualisations  of  conative  dispositions  through 
presumption,  judgment,  and  assumption.  An  obvious  inference 
from  this  conclusion  would  be  that  the  differences  in  meaning 
of  these  different  modes,  passion,  sentiment,  affective  sign, 
emotion,  mood,  affective  sign,  may,  in  so  far  as  they  are  signifi 
cant  for  worth  experience,  be  conceived  in  terms  of  modification 
of  cognitive  presuppositions. 

This   inference   is,  I    think,  justified   by  the   facts.     Senti 
ments,   moods,   and,   still    more,   affective   signs,   representing, 
as  they  do,  habituation   in  varying  degrees,  are  significant   in 
worth  experience  precisely  because  they  are  modes  of  feeling 
which  go  with  assumptions  of  the  two  types,  explicit  and  im 
plicit.     In  the  preceding  chapter  we  had  occasion  to  point  out 
that,  even  in  the  case  of  those  moods  apparently  physiological 
in  origin,  objectless  feelings  in  general,  there  is  really  a  vague 
presumption   of   reality.     All    forms    of    affectivity  which    are 
worth-suggestive   have    cognitive   presuppositions,   and  in   the 
case   of   the  modes    of    feeling    which    act    as   substitutes    for 
passion    and    emotion    the    presuppositions    are    assumptions, 
explicit  or  implicit.     In  the  first  case,  as  we  shall  see  presently, 
we  are  concerned  with  feelings  of  the  imagination,  in  the  second, 
with  feeling-abstracts.     Accordingly,  the  different  modes  of  the 
vital  series,  passion,  sentiment,  affective  sign  or  emotion,  mood, 
affective  sign,  represent  on  the  side  of  presuppositions,  and  in  their 
aspect  of  worth  suggestion,  a  gradual,  although  not  always  recog 
nised,  change  in  presuppositions.     Passion  and  emotion  represent 
readjustment  after  arrest,  and  this  readjustment  in  so  far  as  it 
is  a  worth  experience  implies  judgment.     But  sentiment  and 
mood,  and  still  more  the  affective  sign,  stand  for  the  adjustment 
as  it  approaches  accomplishment.     These  forms  of  affectivity 
represent  the  affective-volitional  meaning  of  habit,  and  as  such 
are  the  psychical  correlates  of  dispositions. 


1 08  Valuation  :   its  Nature  and  Laws 

3.   Corollaries  from  the  Preceding  Theory  of  Feeling. 

In  the  light  of  the  preceding  analysis  and  theory  it  is  now 
possible  to  understand  psychologically  certain  facts  of  worth 
experience  which  the  studies  of  the  preceding  chapters  have 
brought  to  light.  We  have  seen  that  the  appreciative  dis 
tinctions  and  descriptions  of  worth  experience  involve  the 
possibility  of  differentiation  and  ultimately  independent  vari 
ability  of  different  aspects  of  the  total  feeling-attitude.  This 
appeared  most  emphatically  in  our  introductory  distinction 
between  feeling  of  worth  (with  its  meanings,  references)  and 
pleasantness-unpleasantness,  between  degree  of  feeling  of  value 
and  degree  of  intensity  of  pleasantness-unpleasantness.  From 
the  differentiation  of  these  aspects,  and  from  the  very  fact  that 
they  can  be  presented  and  described,  it  was  also  inferred  that 
the  psychical  as  such,  in  this  case  the  feeling,  must  in  some 
way  be  the  object  of  recognition,  presentation,  and  judgment, 
and  thus  approach  to  the  cognitive  side  of  experience.  The 
application  of  these  conceptions  constitutes  the  special  problem 
of  the  following  chapter.  Here  we  shall  content  ourselves  with 
showing  how  our  theory  of  feeling  makes  these  conceptions 
possible. 

With  respect  to  the  first  of  these  problems — how  acquire 
ment  of  funded  meaning  may  go  hand  in  hand  with  diminution 
of  hedonic  intensity,  or  how  the  aspect  of  hedonic  intensity 
may  become  redundant  in  any  given  worth  experience — our 
concept  of  feelings  as  meanings  embodied  in  a  certain  type  of 
sensitivity  is  seen  to  be  enlightening.  The  two  directions  of 
feeling  attitude,  positive  and  negative,  which  belong  to  it  by 
virtue  of  its  relation  to  conative  disposition,  become  pleasant 
ness-unpleasantness  when  the  attitude  is  viewed  retrospectively, 
as  predominantly  passive.  When  thus  viewed,  moreover, 
passive  affectivity  becomes  a  form  of  sensitivity  and  may 
properly  be  said  to  have  degrees  of  intensity.  But  since 
intensity,  in  its  narrower  sense,  is  also,  properly  speaking, 
an  attribute  of  sensitivity  alone,  the  meanings  of  the  attitude 
in  its  prospective  reference,  the  transgredient  and  immanental 
references  which  it  has  by  virtue  of  its  cognitive  presuppositions, 
have  degree,  but  not  intensity.  As  an  analogy  we  may  take  the 
meaning  of  an  idea  and  its  content  or  imaged  substrate.  The 
latter  has  intensity,  but  the  former  intension  and  extension. 

This  fact  being  recognised,  we  may  see  the  meaning  of  the 


Psychological  Basis  of  a   Theory  of  Valuation       109 

statement  previously  made  that  "  pleasantness  rises  as  a  separate 
content  of  consciousness  only  after  the  real  manifold  character 
istics  have  determined  our  actions,"  i.e.,  only  when  it  is  selectively 
differentiated,  as  passivity,  from  the  other  meanings,  the  trans- 
gredient  and  immanental  references.  When  the  other  aspects, 
or  meanings,  of  the  total  attitude  are  uppermost,  i.e.,  in  the 
prospective  reference  of  the  attitude,  the  passive  hedonic  aspect 
is  in  abeyance,  and  for  the  worth  aspect  irrelevant  and  re 
dundant.  Thus  it  comes  about,  as  we  have  repeatedly  seen, 
that  the  worth  suggestions  of  the  feeling-attitude  may  be  un 
affected  by  the  dampening  of  the  absolute  intensity  of  the 
organic  resonance  or  sensitivity.  Intensity  is  a  function  of 
structural  modification,  of  degree  of  arrest  of  organic  tendency, 
and,  while  habit,  which  comes  with  adaptation,  involves  dimin 
ution  of  this  intensity,  it  does  not  necessarily  involve  modi 
fication  of  the  other  meanings  of  the  content.  This  we  have 
seen  in  our  study  of  sentiments,  moods,  and  "  affective  signs." 

Thus  it  comes  about  also  that  in  any  given  total  attitude 
(of  any  duration)  there  may  be  variations  of  hedonic  tone, 
both  qualitative  and  quantitative,  which  do  not  in  the  least 
affect  the  worth  suggestion  or  meaning  of  the  total  attitude. 
Sadness  is  a  negative  worth  attitude  and  hope  a  positive,  the  one 
is  predominantly  pleasant  and  the  other  predominantly  un 
pleasant  ;  but  the  sadness  as  a  total  attitude  may  be  now  pleas 
antly,  now  unpleasantly  tinged,  while  hope,  though  predomi 
nantly  pleasant,  may  contain  unpleasant  moments.  Yet  this 
does  not,  as  Stout  says,  affect  in  the  least  degree  the  strength  of 
the  conative  aspect  of  the  attitude  or,  as  I  should  prefer  to  say, 
the  variations  of  hedonic  tone  of  the  passive  aspects  of  the  feeling 
attitude  are  irrelevant  to  the  worth  suggestion  or  meaning  of  the 
total  attitude. 

With  this  conception  of  the  different  aspects  or  meanings  of 
the  feeling  attitude,  differentiated  in  the  process  of  accommod 
ation,  there  arises  the  further  question  of  the  possibility  of 
feeling  acquiring  that  objectivity  which  goes  with  recognition 
and  presentation.  In  our  structural  study  we  had  occasion  to 
note  different  degrees  in  which  a  feeling  may  be  independent  of 
its  obj  ect.  The  feeling-tone  of  the  sensation  is  inseparable  from  the 
content  to  which  it  attaches,  while  the  emotion,  sentiment,  or 
mood,  in  other  words,  the  "  disposition-feeling,"  spreads  over 
consciousness.  The  question  arises  whether  this  independence, 
this  segregation  or  detachment  of  feeling  from  its  object  can 


I  IO 


Valuation :   its  Nature  and  Laws 


reach  such  a  point  that  the  feeling  itself  may  acquire 
and  genenc  meaning.     There  are  seyera]  ^   1 

enology  of  valuation  where  such  an  hypothesis  is  required 

vhicht"?'  ,'"  the  Pr°CeSSeS  °f  Bathetic  participa  on  to 
which  the  feeling  as  projected  object  acquires  recognitive  mean" 
mg  for  the  subject,  but  with  these  special  applicataTwT^e 
not  now  concerned.  Our  only  interest  here  is  to  point  ouTthat 
m  this  conception  of  feeling  as  a  meaning,  as  a  fom  of  comb  n- 

penden   o6f  tmentS  v  ^"^  S6nSitiV"y' itsdf  -latively  ,ndt 
lent  of  this  sens.tivity,  we  have  a  basis  for  this  hypothesis 


CHAPTER  V 

THE   CONTINUITY   OF  AFFECTIVE-VOLITIONAL 
MEANING 

THE  ACQUIREMENT  OF  RECOGNITIVE  AND  GENERIC  AFFECTIVE 
MEANINGS — AFFECTIVE  MEMORY  AND  GENERALISATION 

i.  The  Problem. 

A  THIRD  problem  of  the  psychology  of  valuation,  requiring 
a  special  analysis  of  feeling,  is  connected  with  the  question 
of  the  nature  of  the  continuity  of  worth  experience  and  judgment. 
This  question,  briefly  stated  in  the  preceding  chapter  and  re 
served  for  a  separate  discussion,  now  becomes  answerable  in 
the  light  of  the  results  of  the  analysis  of  that  chapter.  The 
question,  as  there  stated,  is  this  :  How  does  the  meaning  of 
previous  emotional  experience,  having  found  expression  in  a 
judgment  of  value,  function  in  new  judgments,  in  new  accommod 
ations?  Otherwise  expressed,  How  shall  we  understand  psy 
chologically  the  funded  meaning  which  an  object  acquires  through 
the  formation  of  dispositions  ? 

The  view  generally  held  is  that  feeling  is  a  discontinuous 
accompaniment  of  cognitive  experience,  that  it  becomes  func 
tional  merely  by  the  creation  of  physiological  dispositions  and 
tendencies,  or  by  reflective  interpretation  of  the  past  experience 
in  a  new  situation.  One  writer  states  the  problem  thus  :  "  Now, 
if  it  were  in  a  direct  way,  as  immediately  felt  emotion,  that  the 
consciousness  of  value  must  be  functional  if  functional  at  all,  then 
the  problem  might  well  be  given  up ;  but  it  would  be  a  serious 
blunder  to  conceive  the  problem  in  this  strictly  psychological  way. 
A  logical  statement  of  the  problem  would  raise  a  different  issue- 
not  the  question  whether  emotion,  as  emotion,  can  in  any  sense 
be  functional  in  experience,  but  whether  the  consciousness  of 
value,  and  emotion  in  general,  may  not  receive  reflective  in 
terpretation,  and  thereby,  becoming  objective,  play  a  part  as  a 


1 1 2  Valuation  :   its  Nature  and  Laws 

factor  in  subsequent  valuation-processes."1  Now,  it  is  quite 
true  that  from  the  logical  point  of  view  it  is  "  simply  a  matter 
of  fact  "  that  a  present  consciousness  of  value  does  become  a 
factor  in  subsequent  valuation-process.  It  is,  however,  not 
clear  that  this  dispenses  us  from  the  necessity  of  understanding 
how  this  is  psychologically  possible.  It  is  also  clear,  on  the 
other  hand,  that  if  this  continuity  is  conceived  purely  physio 
logically,  if  the  disposition  is  conceived  to  be  functional  without 
any  psychical  correlate  of  conscious  meaning,  the  valuation- 
process  is  psychologically  discontinuous  just  as  truly  as  in  the 
logical  statement  of  the  situation. 

Neither  of  these  views  is  an  adequate  account  of  the  con 
tinuity  of  valuation.  Immediately  felt  emotion  does  in 
deed  become  functional  through  reflective  interpretation,  i.e.,  by 
becoming  the  object  of  new  judgments  and  assumptions,  but 
there  are  also  felt  continuities  in  which  the  acquired  meaning 
of  past  feelings  determines  directly  the  meaning  of  the  present 
feeling.  For  the  explanation  of  such  continuities  we  must 
assume  an  emotional  "  logic,"  in  which  feeling  is  directly  de 
terminative  through  its  acquirement  of  recognitive  and  generic 
meanings.  In  a  preceding  chapter,  where  the  difference  be 
tween  hedonic  intensity  and  "  depth  and  breadth  "  of  funded 
meaning  was  appreciatively  determined,  it  was  suggested  that 
a  theory  of  "  affective  generalisation,"  of  development  of  generic 
feeling-attitudes,  afforded  a  basis  for  the  understanding  of  these 
facts.  To  the  consideration  of  this  theory  we  must  now  turn. 

Certain  phenomena  already  examined  afford  the  starting- 
point  for  this  theory.  Our  combined  structural  and  functional 
analysis  of  feeling  has  disclosed  certain  forms  of  feeling,  senti 
ments,  moods,  and,  more  particularly,  "  affective  signs,"  which 
differ  in  certain  definite  ways  from  passional  and  emotional 
reactions  to  specific  situations.  Their  lack  of  sensational 
intensity,  their  character  as  residual  feelings,  gives  them  a 
representative  function,  and  enables  them  to  play  a  functional 
role  in  worth  determination,  irrespective  of  their  intensity,  a 
role  analogous  to  that  of  the  general  concept  in  cognition.  As 
we  may  speak  of  a  relatively  imageless  apprehension  of  cognitive 
meaning,  so  we  may  speak  of  a  relatively  intensity-less  apprecia 
tion  of  affective-volitional  meaning  or  value. 

Now  the  fact  itself — of  the  existence  of  such  residual  feelings, 

1  H.  W.  Stuart,  Valuation  as  a  Logical  Process,  in  Dewey's  Studies  in  Logical 
Theory,  p.  336. 


The  Continuity  of  Affective-  Volitional  Meaning     1 1 3 

such  relatively  intensity-less  phases  of  feeling  (which  take  the  place 
of  explicit  and  more  concrete  emotions  and  passions) — is,  I  think, 
beyond  doubt.  From  various  quarters  the  facts  themselves 
are  being  recognised.  The  only  question  is  as  to  the  usefulness 
of  this  conception  of  memory  and  generalisation,  or  of  recognitive 
and  generic  meanings,  as  a  formula  under  which  to  group  the 
facts.  Of  its  usefulness  as  an  hypothesis  for  explaining  the 
unities  and  continuities  of  valuation  the  present  writer  is  per 
suaded,  despite  recent  criticisms  of  the  conception.  Nevertheless, 
it  is  important  to  recognise  that  it  is  merely  an  hypothesis  of  an 
analogical  character,  and  that  it  is  valuable  only  in  so  far  as  it 
enables  us  to  get  a  deeper  insight  into  the  facts  of  valuation ;  and 
that,  further,  there  is  no  objection  to  the  substitution  of  another 
conception  if  it  fulfils  this  function  better.  This  position 
I  have  already  maintained  in  a  critical  study  of  Ribot's  theory.1 
The  reason  for  the  retention  of  the  concept  of  an  "  affective 
logic  "  is  to  be  found  in  the  fact  that  the  phases  of  affectivity 
under  discussion  do  seem  to  have  a  representative  role  analogous 
to  that  of  generals  and  abstracts  in  the  sphere  of  cognition, 
that  they  do  determine  actual  feeling  and  worth  judgments,  as 
opposed  to  the  view  that  they  are  totally  different  from  real 
feelings  and  do  not  affect  them.  This  is  the  significant  point, 
and  a  point  to  which  we  shall  devote  a  fuller  discussion  later. 

With  this  view  of  the  nature  of  the  hypothesis  in  mind,  we 
have  the  following  problems  before  us  :  (i)  What  are  the  phe 
nomena  of  worth  determination,  of  affective-volitional  process, 
which  give  rise  to  this  hypothesis  ?  (2)  What  is  the  psychological 
criterion  in  terms  of  content  and  function,  of  recognitive  and 
generic  meaning  in  the  sphere  of  cognition  ?  What  are  the 
similarities  between  these  and  the  phases  of  affectivity  under 
discussion  which  give  rise  to  this  analogical  hypothesis  ?  In 
other  words,  What  is  the  criterion  of  affective  memory  and 
generalisation  ?  (3)  What  is  the  origin  of  these  phases  and 
their  function  in  the  unities  and  continuities  of  valuation  ? 

II.  THE  ACQUIREMENT  OF  RECOGNITIVE  MEANING  BY  FEELING 
THE  PROBLEM  OF  "AFFECTIVE  MEMORY" 

The  problem  of  "  affective  memory  "  has  recently  received 
considerable  attention.2  In  my  own  earlier  studies  in  this 

1  Review  of  Ribot's  La  logique  des  sentiments.  Psych.  Bulletin,  Vol.  II,  No.  9. 

2  The  fullest  and  most  recent  study  of  Affective  Memory  is  found  in  Paulhan's  La 
Jonction  de  la  memoirc  et  le  souvenir  affectif,  Paris,  1904. 


ii4  Valuation:   its  Nature  and  Laws 

subject.1  it  was  pointed  out  that  a  distinction  must  be  made 
between  the  fact  of  recognition  of  feeling  as  past  and  our 
explanation  of  the  phenomenon,  and  that,  moreover,  the  accept 
ance  of  the  facts  has  been  hindered  by  certain  prejudices  of 
theory,  both  as  to  the  nature  of  feeling  and  of  memory.  If  we 
identify  feeling  with  one  of  its  aspects,  pleasantness-unpleasant 
ness,  and  then  ask  whether  an  hedonic  tone  may  be  recalled,  there 
is  indeed  none  but  a  negative  answer  to  that  question.  There  is 
no  restoration  of  a  past  hedonic  tone,  except  in  conceptual  terms, 
for  every  experience  of  pleasantness-unpleasantness  is  an  actual 
present  feeling.  If,  however,  we  put  the  question  as  it  should  be 
put — does  a  present  feeling  ever  acquire  recognitive  meanings  ? — 
the  answer  is  in  the  affirmative.  And,  when  we  further  ask: 
How  does  it  acquire  this  meaning,  and  what  aspects  are  recog 
nised  ?  there  is  some  hope  of  answering  the  question.  To  this 
question,  in  so  far  as  it  concerns  our  psychological  theory  of 
valuation,  we  must  give  some  attention,  but  we  may  at  the 
outset  disclaim  any  intention  of  exhaustive  treatment  and  refer 
for  details  to  the  studies  already  cited. 

i.   Types  of  Affective  Memory. 

The  facts  of  affective  memory  have  been  collected  in  con 
siderable  number.  The  restatement  of  them  here  would  re 
quire  more  space  than  can  be  devoted  to  the  subject.  Our 
concern  is  primarily  with  the  interpretation  of  the  phenomena 
and  the  determination  of  their  place  in  our  genetic  and  functional 
theory  of  acquirement  of  affective-volitional  meaning. 

We  may  recall,  however,  that  there  are  two  forms  which 
have  been  generally  distinguished  :  (i)  what  may  be  called 
voluntary  revival,  or,  more  carefully  expressed,  perhaps,  volun 
tary  re-instatement  of  a  feeling  attitude  with  recognitive  meaning, 
and  (2)  involuntary  spontaneous  recurrence  of  an  attitude, 
without  its  accompanying  object,  but  with  recognitive  meaning. 

Of  the  first  type  of  re-instatement  there  are  numerous  illus 
trations  in  the  literature  referred  to,  and,  in  fact,  in  every 
day  experience.  I  shall  note  only  one,  which  is  considered  by 
Pillon  to  be  distinct  disproof  of  the  view  that  the  so-called 
revived  emotions  are  really  new  ones.  It  is  a  passage  in  the 
Nouvelle  Heloise  (Part  IV,  Letter  17)  where  Saint-Preux  de- 

"  The  Problem  of  a  Logic  of  the  Emotions  and  Affective  Memory,"  Psychological 
Review,  Vol.  VIII,  Nos.  3  and  4. 


The  Contimiity  of  Affective-  Volitional  Meaning     1 1 5 

scribes  himself  as  reviving  in  the  presence  of  the  old  scenes  of 
love  the  same  emotions,  but  as  falling  into  rage  and  despair 
upon  recognising  their  futility.  Pillon  considers  this  evidence 
that  there  is  a  recognised  difference  between  the  revived  emotion 
and  the  new  in  the  mind  of  Rousseau. 

If  we  examine  this  illustration  of  the  first  type  more  closely, 
it  appears  that  we  have  to  do  with  what  has  recently  been 
described  as  "  feelings  of  the  imagination."  Feelings  of  imagina 
tion  are  those  which  follow  upon  the  representation  of  image 
content,  where  the  content  is  not  presumed  or  judged,  but  merely 
assumed  to  exist.  Such  a  feeling  may  be  evoked,  either  by 
recalling  images  of  the  past  with  assumption  of  existence  in 
the  past,  or  by  calling  them  up  with  assumption  of  present  or 
future  existence.  In  the  first  case  the  feelings  have  recognitive 
meaning,  and  we  speak  of  affective  memory  ;  in  the  latter  case 
we  have  imagined  feelings  in  the  more  limited  sense  of  the  word. 
In  both  cases  the  "  feeling  of  the  imagination  "  differs  from 
the  actual  present  feeling  in  its  coefficient  of  reality  and  in 
certain  aspects  of  structure. 

Some  feelings  of  the  imagination  (i.e.,  assumption-feelings) 
have,  then,  recognitive  meaning,  and  it  is  important  to  note 
that  these  feelings  with  recognitive  meaning  may  be  reinstated, 
either  by  recalling  images  of  the  past  with  assumption  of  ex 
istence,  or  by  a  form  of  auto-suggestion,  motor  or  verbal,  in 
which  the  subject  induces  sentiments  or  moods  by  assuming 
a  motor  attitude,  or  by  use  of  verbal  symbols  having  emotional 
overtones  from  past  experiences  of  feeling.  Toward  this  re 
instated  feeling  a  new  feeling-attitude,  positive  or  negative, 
may  be  taken,  as  in  the  case  of  Saint-Preux  already  described. 
This  feeling  is  the  actual  present  feeling,  and  presupposes  the 
passage  of  assumption  into  explicit  judgment  of  existence  or 
non-existence. 

The  second  type  of  reinstatement  is  quite  different.  In 
this  case  the  feeling-attitude  may  recur  with  recognitive  meaning, 
even  without  any  recognised  presentational  content  or  any 
explicit  assumption  as  presupposition.  The  feeling  itself  is  first 
recognised,  and  only  later  emerge  the  presentations  to  which 
it  refers.  It  is  at  first  of  the  nature  of  an  objectless  feeling, 
but  has,  nevertheless,  recognitive  meaning.  A  classical  illus 
tration  of  this  type  is  the  experience  of  M.  Littre,  given  by 
Ribot,  where  the  feeling  connected  with  the  death  of  his  sister, 
which  took  place  in  his  youth,  comes  over  him  unexpectedly 


n6  Valuation:   its  Nature  and  Laws 

in  his  old  age,  after  the  memory  of  the  death  had  for  years 
ceased  to  be  accompanied  by  feeling.1  In  such  cases  the  recog 
nition  of  the  feeling  seems  to  be  independent  of  the  recognition 
of  presentational  content,  and  is  reinstated  indirectly  through 
association. 

The  two  types  thus  distinguished  have  been  called  respec 
tively  false  and  true  affective  memory,  and,  as  Paulhan  recog 
nises,  there  seems  to  be  a  real  distinction  involved,  although 
not  one  so  ultimate  as  these  terms  would  indicate.  The  dis 
tinction  lies  in  the  fact  that,  while  both  are  feelings  of  the  imagina 
tion  with  certain  definite  marks  which  distinguish  them  from 
real  present  feelings,  the  evocation  of  the  one  is  dependent  upon 
revival  of  image  content,  while  the  other,  the  so-called  true 
memory  of  feelings,  is  itself  the  condition  of  revival  of  memory 
images.  To  the  functional  basis  of  this  distinction  we  shall 
return,  but  we  must  first  determine  the  criterion  which  dis 
tinguishes  the  feeling  of  the  imagination  (in  both  forms)  with 
its  recognitive  meaning,  from  the  "  present  "  actual  feeling. 

2.  The  Criterion  of  Recognitive  Meaning. 

The  criterion  of  recognitive  meaning  (of  any  experience, 
whether  cognitive  or  affective)  is  a  certain  mark  of  pastness 
inherent  in  a  present  experience  or  content.  This  mark  of 
pastness  is  reducible  to  certain  equivalents  in  terms  both 
of  function  and  content.  On  the  side  of  function  the  differ 
ence  between  an  experience  with,  and  one  without,  recognitive 
meaning  seems  to  be  primarily  one  of  control,  of  reality  co 
efficient.  In  the  case  of  the  content  which  has  the  mark  of 
the  present  without  any  reference  to  the  past,  the  content  is 
an  object  cognised  as  immediately  given.  Where,  on  the  other 
hand,  it  has  the  mark  of  pastness,  the  content  is  felt  not  to 

1  Reference  may  also  be  made  to  a  case  cited  by  Mauxion  in  an  article,  "  La  vraie 
memoire  affective,"  Revue  philosophique,  February,  1901,  and  I  may  also  quote  a  case 
from  my  own  experience  described  in  the  article  already  referred  to  :— 

"A  few  years  ago,  while  living  abroad,  there  came  into  my  consciousness,  entirely 
without  associational  conditions  that  were  recognisable,  a  peculiar  emotional  tone  which 
I  recognised  as  having  been  experienced  before  with  peculiar  intensity.  I  located  it 
finally  as  the  emotional  overtone  of  a  peculiarly  desolate  bit  of  anthracite  coal  region. 
So  strong  and  marked  was  it  that  it  developed  into  particular  emotions  of  great  vivid 
ness,  and  sufficient  to  lead  immediately  to  a  bit  of  descriptive  writing.  The  point  of 
psychological  interest  is  that  with  the  closest  search  no  ideal  content  could  be  found 
which  would  account  for  its  revival.  It  is  probably  explained  by  the  fact  that  a  some 
what  similar  feeling  had  that  day  been  generated  by  wholly  different  content— the 
squalor  of  a  certain  quarter  in  a  foreign  city,  and  that  there  had  been  direct  emotional 
recall  through  emotion." 


The  Continuity  of  Affective-  Volitional  Meaning     1 1 7 

be  immediately  given,  but  as  so  connected  with  the  immediately 
given  that  the  immediacy  can  be  restored. 

This  difference  is  reducible,  in  its  functional  aspects,  to 
difference  in  cognitive  acts.  All  content  with  the  mark  of 
presentness  is  presumed  or  judged  to  exist.  The  presumption 
of  existence  characterises,  we  have  seen,  all  acceptance  of  con 
tent  which  is  objectively  determined,  while  judgment  is  but 
the  explicit  acknowledgment  of  that  determination.  The 
content  with  the  mark  of  pastness,  on  the  other  hand,  is  felt 
to  be  connected  with  immediate  content  the  existence  of  which 
is  assumed.  It  is  this  connection  with  assumed  reality  which 
gives  it  its  recognitive  meaning. 

To  this  functional  difference  between  the  content  with  past 
and  present  meaning  there  corresponds  a  difference  in  structure. 
Analysis  discloses  the  fact  that  memory  images  differ  from  im 
mediate  perception  in  that  they  are  schematic,  and  lack  the 
peculiar  "  sensational  intensity  "  which  characterises  immediate 
perception.  The  schematic  character  lies  in  the  fact  that 
there  has  been  a  "  wearing  away  "  of  certain  elements  in  the 
total  complex,  those  elements  being  retained  which  through 
complication  and  other  forms  of  association  are  connected 
with  previous  experiences.  The  reduction  in  intensity  is  prob 
ably  to  be  correlated  with  the  reduction  of  tendency  to  motor 
expression  in  the  case  of  memory  images.  However  we  explain 
these  structural  differences  between  immediate  perception  and 
memory  image,  they  are  closely  connected  with  the  functional 
criterion  of  recognitive  meaning.  The  content  with  these 
characteristics  is  distinguished  from  immediate  perception, 
is  inhibited  by  perception,  is  thrown  back  and  localised  in  the 
past. 

3.  The  Criterion  of  Recognitive  Meaning  as  Applied 
to  Feelings. 

From  these  reflections  we  should  naturally  be  led  to  infer 
that,  among  the  classes  of  feelings  distinguished  by  our  analysis, 
feelings  with  presumptions  and  judgments  as  presuppositions 
do  not  have  recognitive  meaning,  and  that  assumption  feelings 
alone  have  the  qualification  of  pastness.  And  this  inference 
is  justified  by  the  facts.  For,  in  the  first  place,  the  two  types 
of  feelings  which  have  the  mark  of  pastness,  where  a  present 
feeling  seems  to  be  a  re-instatement  of  a  past  experience  (the 
so-called  false  and  true  affective  memory),  are  either  "  feelings 


1 1 8  Valuation :   its  Nature  and  Laws 

of  the  imagination,"  with  assumptions  as  presuppositions 
or  objectless  feelings  which,  although  at  first  without  objects 
and  explicit  presuppositions,  develop  into  feelings  of  the  imagina 
tion.  And  what  gives  them  their  recognitive  meaning,  it  is 
easy  to  see,  is  the  belief  that  they  are  reducible  to  or  convertible 
into,  actual  immediate  feelings,  */  the  content  which  is  their 
object  should  again  acquire  the  coefficient  of  reality  which  goes 
with  perception  and  its  presumption  of  existence,  or  if  the 
content  should  again  become  the  object  of  existential  judgment. 
On  the  other  hand,  it  is  equally  easy  to  see  why  feelings  with 
presumptions  and  judgments  as  presuppositions  have  no  mark 
of  pastness.^  Each  act  of  apprehension  or  judgment  is  itself 
a  new  accommodation  and  involves  a  new,  unique  feeling. 

Both  these  types  of  feeling  have,  then,  this  qualification  of  • 
pastness  or   recognitive  meaning,  because  of  their  reference  to 
an  actual  feeling  which  it  is  believed  would  be  reinstated  with 
the  reinstatement  of  perception  or  judgment.     But  this  con 
version  is  not   always    possible,   and,   when    attempted,   it  is 
frequently  followed   by  a  change   in  feeling— the  new   feeling 
being  neutral,  or,  indeed,  often  opposite  in  character.1     New  con 
ditions  of  the  present,  or  changes  in  the  disposition  of  the  subject, 
often  inhibit  the  judgment-feeling  toward  which  the  assumption- 
feeling  points.     In  the  case  of  Saint-Preux  the  feelings  of  the 
imagination  with  recognitive  meaning  refuse   to  take   on   the 
actuality  of  a  present  feeling,  and  are  felt  to  be  futile.     When 
one  returns  to  the  scenes  of  his  childhood,  feeling  has  often 
become   neutral   or   opposite   in   character.     The   feelings   con 
nected  with  a  shipwreck  or  a  death,  as  long  as  they  are  merely 
remembered  or  imagined,  retain  all  the  aspects  of  the  early 
experience,  except  that  which  gives  it  its  mark  of  presentness  ; 
but  the  present  actual  affective  attitude  toward  the  past  event 
may  become  neutral,  or  at  least  blunted.     This  appears  notably 
in  the  case  of  the  experience  of  M.  Littre,  where,  although  the 

1  It  is  interesting  to  note  that  Angell  (Psychology,  p.  266)  admits  the  existence  of 

ctive  memory  in  the  practical  sense  that  we  feel  that  we  could,  if  necessary,  recall 

events  with  their  former  feelings,  although,  he  goes  on  to  say,  "if  we  actually  attempt 

o  recall  the  event  we  find  that  sometimes  the  recollection  itself  is  affectively  colourless, 

jmetimes  it  has  the  affective  character  of  the  original  event  and  sometimes  an  opposite 

character.       It  is  this  fact  that  we  feel  that  we  could  recall  that  is  significant.     In  order 

that  we  may  have  this  reference  to  a  real  feeling  of  the  past  at  all,  there  must  be  an 

affective  sign     representing  it.     Whether  that  affective  sign  or  feeling  of  the  imagina- 

lon  with  recognitive  meaning,  can  be  converted  into  a  real  feeling  or  not,  whether, 

owing  to  change  in  feeling-dispositions,  the  present  actual  feeling  differs  from  the  past 

feeling,  is  beside  the  mark.     Angell's  analysis  is  here  inadequate  for  the  reason  that  he 

tails  to  distinguish  between  the  revived  feeling  and  the  present  actual  feeling,  the  latter 

emg,  of  course,  a  new  feeling  without  the  mark  of  pastness. 


The  Continuity  of  Affective-Volitional  Meaning     119 

actual  feeling  attitude  toward  the  past  event,  the  death  of 
his  sister,  had  long  since  become  neutral,  nevertheless  the  feeling 
of  the  past,  through  some  chance  association,  returned  in  its 
former  character  and  meaning. 

What,  then,  is  this  mark  of  the  actual  present  feeling  which 
distinguishes  it  from  the  feeling  of  the  past  ?     Those  who  have 
analysed  these  feelings  of  the  imagination  with  the  mark  of 
pastness  agree  that  the  "  revived  "  feeling,  whether  imaginatively 
revived  or   spontaneously  recurring,  differs  from  an  "  actual ' 
present  feeling  in  certain  respects  analogous  to  those  differences 
discoverable  in  the  sphere  of  cognitive  images.     Like  the  revived 
image,    the  imagined   feeling    or    emotion    is    schematic.     But 
schematic  in  this  sphere  means  the  absence  of  the  full,  rich 
organic    sensations    in    which    the   "present"    feeling    is    em 
bodied.     With  this  reduction  in  sensitivity  comes  a  corresponding 
reduction  in  intensity.     The  revived  feeling  lacks  that  "  sen 
sational  intensity  "  which  actual  feeling  has.     It  has  its  own 
intensity,  which,  as  has  been  pointed  out  by  Paulhan  and  the 
present  writer,  and  more  recently  by  Saxinger,  is  unaffected 
by  repetition,  and  in  that  respect  stands  out  in  marked  contrast 
with   the   present   actual   feeling.     This   so-called   intensity   of 
the  feeling  of  the  imagination  is  better  described,  however,  as 
the  meaning  of  the  feeling,  to  distinguish  it  from  the  sensational 
intensity  of  the  actual  feeling. 

The   feelings   of   imagination,   following   upon   assumptions, 
have  undergone  certain  modifications  which  distinguish  them 
from  "  actual,"  present  feelings.     It  remains  now  merely  to  de 
termine  what  aspect  of   the   total  feeling-complex  remains  in 
the  feeling  of  the  imagination  and  acquires  recognitive  meaning. 
Upon  the  theory  of  feeling  developed  it  is  not  difficult  to  answer 
this  question.     We  have  already  seen  that,  while  feeling  is  the 
embodied  meaning  of  a  certain  type  of  sensitivity,  that  meaning, 
as  a  form  of  combination  of  the  content,  is  relatively  independent 
of  the  specific  elements  of  the  content  and  of  their  intensity.   It  is 
this  "  form-quality,"  I  think  we  may  safely  say,  that  is  recognised, 
and  it  has  recognitive  meaning  because  it  feels  as  though  it 
could,  were  the  necessary  conditions  given,  be  converted  into 
that  fully  embodied  meaning,  with  all  the  reality  feeling  of 
immediate  perception  with  its  primary  presumption  of  reality 
or  of  explicit  existential  judgment. 

This  view,  much  more  fully  developed  in  the  paper  already 
referred  to,  receives  confirmation  in  the  introspective  accounts 


1 20  Valuation :   its  Nature  and  Laws 

of  "revived"  feelings.  Notwithstanding  the  fact  that  the 
actual  feeling  attitude  toward  a  past  experience  may  have 
changed,  the  subject  may  still  recall  in  imagination  the  precise 
nuance  of  emotion,  the  thrill,  the  expansiveness,  etc.,  of  the 
feeling  from  which  the  mark  of  present  reality  has  vanished, 
provided,  of  course,  he  is  of  the  affective-memory  type.1  It  is 
a  view,  moreover,  which,  if  valid,  will  enable  us  to  understand 
how  feeling  may  become  objective,  may  be  projected,  in  processes 
of  imaginative  Einfuhlung,  into  another  person,  recognised  and 
read  back  into  the  self. 


III.   THE  ACQUIREMENT  OF  GENERIC  MEANINGS  ON  THE  PART 
OF  FEELING — AFFECTIVE  GENERALISATION 

With  this  understanding  of  the  nature  of  the  acquirement 
of  recognitive  meaning  on  the  part  of  feeling-attitudes,  we 
may  turn  to  the  consideration  of  the  further  hypothesis  de 
veloped  in  the  effort  to  account  for  the  continuity  of  affective- 
volitional  meaning— namely,  that  certain  phases  of  feeling 
acquire  a  generic  meaning,  relating  them  to  the  judgment  of 
value  in  a  manner  analogous  to  the  relation  of  the  general  con 
cept  to  the  judgment. 

The  common  characteristic  of  the  theories  of  affective- 
continuity  criticised  is,  it  will  be  remembered,  that  there  is 
no  felt  continuity— that  habit  has  no  felt  meaning.  Feeling 
of  worth  being  identified  with  hedonic  intensity,  as  feel 
ing-attitude  approaches  habit  actual  worth  feeling  decreases. 
In  the  case  of  the  genetic  modes  of  feeling-attitude  analysed, 
passion,  emotion,  sentiment,  mood,  and  affective  sign,  we 
have,  with  approach  to  habit,  a  progressive  diminution  of 
intensity,  and  with  it  a  decrease  of  actual  worth  feeling. 
In  opposition  to  this  view  we  have  maintained  that  these 
higher  genetic  modes  of  feeling-attitude  are  related  to  passion 
and  emotion  as  the  general  concept  is  related  to  the  particular 
percept.  As  the  general  concept  represents  the  acquired  mean- 

)  A  passage  from  Taine's  De  ?  intelligence,  quoted  by  Paulhan  with  reference  to  the 
revival  of  emotions,  shows  the  aspects  of  feeling  recognised  on  revival.  "  La  seule  chose 
qui  en  moi  se  reproduise  intacte  et  entiere,  c'est  la  nuance  precise  demotion,  apre, 
tendre,  etrange,  douce  ou  triste,  qui  jadis  a  suivi  ou  accompagne  la  sensation  exte- 
rieure  et  corporelle;  je  puis  renouveler  ainsi  mes  peines  et  mesplaisirs  les  plus  compliques 
et  les  plus  delicats,  avec  une  exactitude  extreme  et  a  de  tres  grandes  distances  ;  a  cet 
egard  le  chuchotement  incomplet  et  defaillant  a  presque  le  meme  eftet  que  la  voix." 


The  Continuity  of  Affective-  Volitional  Meaning     121 

ing  of  a  judgment  disposition,  so  these  generic  forms  of  affectivity 
represent  the  acquired  or  funded  affective-volitional  meaning  of 
particular  emotional  reactions,  and  have  a  functional  role  in 
worth  determination,  independent  of  their  intensity,  analogous 
to  the  role  of  the  general  concept  in  cognitive  judgment. 


i.    The  Phenomena  of  Affective  Continuity  :  Substitution, 
Subsumption,  Transition. 

This  analogy  between  the  function  of  the  general  concept 
and  the  so-called  generalised  emotion  appears  more  specifically 
in  three  types  of  felt  continuity — affective  substitutions,  affective 
subsumptions,  and  affective  transitions.  Just  as  an  image  with 
generic  meaning  may,  because  of  its  reference  to  past  situations 
and  judgments,  take  the  place  of  varied  image  content  in  our 
judgments  and  reactions,  so  certain  affective  signs  may  take 
the  place  of  specific  emotional  reactions.  As  the  subsumption 
of  a  particular  image  under  a  general  concept  gives  to  the 
particular  the  cognitive  predicates  of  the  general,  so  the  colour 
ing  of  a  specific  emotion  by  a  sentiment  or  mood  gives  to  the 
former  the  worth  connotation  of  the  latter.  Finally,  as  a  par 
ticular  image  with  generic  meaning  affords  the  basis  of  transi 
tion  from  particular  to  particular,  so  the  generic  meanings  of 
affective  states  make  possible  affective  transitions  analogous  to 
logical  judgment,  giving  rise  to  an  "  affective  logic." 

The  phenomena  of  affective  substitution  have  already  been 
described  under  the  head  of  "  affective  signs."  There  it  was 
seen  that  relatively  contentless  and  intensity-less  phases  of 
affectivity  could,  as  schematic  meanings,  take  the  place  of 
particular  passions  and  emotions.  On  closer  analysis  it  further 
appeared  that  these  substitutes  might  be  of  two  kinds,  either 
a  feeling  following  upon  explicit  assumption  of  an  object  (where 
the  assumption  is  a  substitute  for  judgment)  or  the  emotional 
connotation  of  a  general  term,  the  feeling  in  this  case  following 
upon  implicit  assumption,  but  connected  either  with  judgmental 
habit  or  with  a  mere  verbal  image. 

The  second  group  of  phenomena,  affective  subsumptions, 
has  been  presented  in  detail  in  two  papers  in  the  Psychological 
Review."  A  brief  restatement  of  the  chief  points  of  this  discus- 

1  Cf.  pp.  105  ff. 

2  "The  Problem  of  a  Logic  of  the  Emotions  and  Affective  Memory,"  Psychological 
Review,  Vol.  VIII,  Nos.  3  and  4. 


122  Valuation  :   its  Nature  and  Laws 

sion  will  suffice.  Two  forms  of  this  emotional  unity  or  affective 
subsumption  were  distinguished:  (i)  ethical  sanction,  and  (2) 
aesthetic  appreciation.  The  characteristic  of  both  of  these  phe 
nomena  is  that  certain  particular  emotional  reactions  are  sub 
sumed  directly  under  sentiments,  moods,  or  affective  signs 
without  the  mediation  of  intellectual  relational  judgments, 
thus  giving  rise  to  wholly  emotional  unities.  To  these  may 
be  added  certain  mystical  religious  states  which  represent  the 
extreme  of  emotional  unity,  with  a  minimum  of  presentational 
content  and  judgment. 

Ethical  sanction  of  the  emotional  type  is  characterised  by 
the  fact  that  some  generic  dominant  emotional  attitude  having 
acquired  the  transgredient  reference  of  obligation,  already 
described,  is  "  felt "  to  include  more  particular  desires  and 
emotions,  and  to  sanction  them.  Illustrations  of  this  inclu 
sion  appear  in  such  expressions  as  "  holy  "  anger,  "  reverent " 
fear  or  mirth.  In  such  cases  the  immediate  emotional  re 
action,  the  anger  or  mirth,  is  coloured  by  a  presupposed  senti 
ment  or  mood.  As  will  be  shown  in  detail  later,  an  actual 
judgment- feeling  is  determined  by  a  generic  feeling- attitude, 
which  accompanies  an  ever-present  implicit  assumption.  Of 
interest  in  this  connection  is  also  the  gradual  assimilation  of 
marital  under  the  maternal  sentiment,  described  by  Ribot, 
for  it  frequently  has  the  character  of  ethical  sanction.  These 
phenomena  have  generally  been  described  as  mixtures  of  feeling, 
resulting  from  the  presence  in  the  same  unity  of  consciousness  of 
different  images,  but  such  a  conception  ignores,  as  we  have 
already  seen,  the  difference  in  the  functional  character  and 
presuppositions  of  the  two  feelings.  The  justification  of  calling 
them  inclusions,  rather  than  mixtures  of  feeling,  is  to  be  found 
in  the  fact  that  the  generic  aspect  of  the  total  attitude  gives  its 
meaning  to  the  other  aspect — sanctions  it,  in  that  it  determines 
its  place  in  a  system  of  worths.  The  generic  feeling  is,  more 
over,  often  without  explicit  image  content. 

The  experiences  in  which  this  affective  subsumption  is 
seen  in  its  most  complete  form  are  certain  cases  of  aesthetic  ap 
preciation  and  creation.  An  intuitive  "unity  in  variety"  has 
always  been  recognised  as  a  characteristic  of  the  aesthetic  mode, 
but  recently  it  has  become  apparent  that  such  unity  may  be 
almost  exclusively  emotional.  The  unity  for  appreciation 
may  be  created  and  retained  by  inducing  certain  assumptions, 
with  their  corresponding  generic  sentiments  and  moods,  and  by 


The  Continuity  of  Affective-Volitional  Meaning     123 

the  arrangement  of  details  of  imagery  in  such  a  manner  that  the 
feelings  and  emotions  connected  with  this  imagery  may  be 
easily  assimilated  to  the  primary  sentiment  or  mood.  They 
do  not  disturb  the  assumption,  are  not,  in  other  words,  "  illusion- 
disturbing."  As  extreme  cases  of  this  we  may  cite  certain 
impressionistic  or  symbolic  styles  where  the  general  mood  or 
Stimmung  is  almost  palpable,  and  can  be  easily  segregated  from 
the  emotional  or  feeling  tones  of  the  elements — whether  words, 
visual  images,  tones,  or  what  not.1  Not  only  may  the  generic 
feelings  be  distinguished  from  the  feeling- tones  of  the  elements, 
but,  like  the  sentiments  of  places,  they  may  often  be  recalled, 
and  recognised  without  revival  of  the  particulars. 

On  the  other  hand,  such  generic  emotions,  sentiments,  or 
moods,  may  be  germinal  to  aesthetic  creation,  and  the  pro 
cess  of  creation  becomes,  in  such  cases,  a  subordination  or  sub- 
sumption  of  particular  images  under  the  general  mood,  accord 
ing  to  their  emotional  values.  On  the  side  of  active  creative 
expansion  of  sentiment  and  mood  over  particulars  we  have 
the  interesting  account  by  Poe  of  the  construction  of  the  Raven. 
He  tells  us  that  his  starting-point  was  the  purpose  to  express 
the  mood  of  melancholy.  For  this  mood  he  found  the  charac 
teristic  refrain,  Nevermore.  His  art  then  consisted  in  finding 
particular  images,  such  as  the  Raven,  the  locality  of  the  poem, 
etc.,  which  had  emotional  tones  that  could  be  subsumed  under 

1  A  striking  illustration  of  this  is  that  exquisite  mood  poem  of  Tennyson's  The 
Lotos-Eaters.  All  his  technique  of  imagery  and  rhythm  is  expended,  not  only  to 
arouse  the  grey  mood  of  forgetfulness  and  indifference,  but  to  carry  it  on  and  on, 
intensified  and  solidified,  until  it  becomes  the  mood  through  which  the  very  gods  see  the 
world.  That  world  includes  many  things  which  do  not  fit  this  mood,  yet,  like  gods, 

They  smile  in  secret,  looking  over  wasted  lands. 

Blight  and  famine,  plague  and  earthquake,  roaring  deeps  and  fiery  sands, 

Clanging  fights,  and  flaming  towns,  and  sinking  ships  and  praying  hands. 

And,  although  these  rapid  pictures  suggest  incipient  emotional  responses  of  another  sort 
than  the  dominant  mood,  they  succeed  each  other  so  rapidly  that  it  is  all 

Like  a  tale  of  little  meaning,  though  the  words  are  strong. 

The  general  mood  may  become  so  strong  that  it  will  spread  over  all  particular  emotional 
tendencies,  provided  the  technique  of  expression  is  such  as  not  to  allow  these  particular 
motor  tendencies  to  get  above  a  certain  strength.  In  this  case  the  technique  consists  in 
the  rapid  piling  up  of  the  pictures,  thus  preventing  the  particular  emotional  suggestions 
from  getting  in  their  full  motor  value.  Experiments  with  people  in  the  reading  of  this 
poem  have  led  me  to  the  conclusion  that  if  the  mood  of  the  poem  is  fully  appreciated 
before  this  passage  comes,  it  may  be  subsumed  under  the  dominant  mood.  Otherwise 
the  contrast  is  too  great  and  the  unity  of  the  poem  seems  to  be  broken. 

The  technique  of  the  Lotos-Eaters,  when  closely  examined,  shows  that  this  subsump- 
tion  is  furthered  by  the  imagery  and  sounds  employed  in  the  poem.  The  former  is 
always  in  vague  general  terms,  keeping  out  the  particularised  images  which  with  their 
intenser  emotions  would  overcome  the  languor  of  the  general  mood.  The  resonances 
are  further  damped  by  the  careful  use  of  vowel  sounds  :  "  Here  are  cool  mosses  deep  "- 
"  the  ivies  creep" — "  the  long  leaved  flowers  weep." 


124  Valuation:   its  Nature  and  Laws 

this  mood.  And  he  did  not  scruple,  he  tells  us,  deliberately  to 
tone  down  or  exclude  all  imagery,  the  emotional  intensity  of 
which  was  too  great  for  the  mood.1 

The  third  group  of  phenomena,  described  as  affective  tran 
sitions,  comprises  certain  continuities  in  which  transition  is  made 
from  one  affective  attitude  to  another  through  the  medium  of  an 
affective,  rather  than  ideal  or  conceptual  mean  term.     In  such 
cases  the  same  phase  of  affectivity,  i.e.,  the  affective  sign,  which 
we  have  seen  in  the  role  of  substitute  for  particular  emotion, 
or  as  giving  colour,  through  a  kind  of  subsumption,  to  particular 
emotions,  becomes  a  mean  term,  connecting  in  one  continuity 
of  meaning,  particular  emotions  otherwise  different  in  character. 
The  most  characteristic  forms  of  such  emotional  signs  are  the 
residual  feelings  or  abstracts  that  inhere  in  general  terms  and 
their  corresponding  words,  such  as  love,  duty,  God,  etc.     These 
may  function  as  transitional  mean  terms  when  there  is  no  con 
crete  object  of  judgment  toward  which  the  feeling  represented 
by  these  words  is  directed.     The  objects  of  such  feelings  are 
not  definite  presentations,  but  as  in  the  case  of  objectless  feelings, 
already   considered,   vague   universals — not   the   object   of   ex 
plicit  judgment,  but  of  assumption.     But  while  the  emotional 
suggestion  of  words  is  the  typical  form  of  transition,  it  is  possible 
that  even  the  word  embodiment  may  be  lacking,  and  that  the 
vague   mood  or  affective   sign  may   determine  worth  attitude 
without  embodying  itself  in  any  presentation  that  analysis  can 
discover  to  be  relevant.     Full  illustrations  of  such  transitions 
are  given  in  the  later  paragraphs  of  this  chapter. 

2.    The  Psychological  Theory  of  Generic  Meanings- 
Structural  and  Functional  Analysis 

The  facts  which  give  rise  to  this  hypothesis  of  the  existence 
of  the  phenomena  variously  described  as  affective  generals, 
abstracts,  or  affective  signs,  and  of  a  corresponding  "  affective 
logic,"  are  now  before  us.  These  facts  are,  in  general,  in  favour 
of  the  view  that  there  are  phases  of  affectivity  in  some  way 
different  from  the  usual  emotional  reaction  to  a  particular  definite 
situation,  therefore  (in  our  terminology),  from  the  particular 
feeling.  They  indicate  also  that  these  phases  acquire  recognitive 

1  Poe     The  Philosophy  of  Composition ;    also    Paulhan,   V Invention,   p.   8l  ;    also 
Ribot,  U  Imagination  crtatricc,  where  Poe  is  classed  as  of  the  diffluente  type  of  imagm: 
tion,  in  which  the  unity  is  the  emotional  abstract. 


The  Continuity  of  Affective-Volitional  Meaning     125 

and  generic  meanings,  which  enable  them  to  function  as  substi 
tutes  for  particular  feeling  reactions.  What,  now,  is  the  ground 
for  describing  these  phenomena  as  affective  generals  or  abstracts  ? 
To  answer  this  question  it  is  necessary  to  develop  a  criterion 
of  abstraction  and  generalisation,  and  to  show  that  these  phe 
nomena  fall  under  such  descriptive  terms. 

(a)   The  Nature  of  Generic  Meanings  of  Ideas: 
Imageless  Apprehension. 

Upon  this  problem,  as  it  applies  to  the  cognitive  side  of 
experience,  we  are  not  without  some  definite  conceptions. 
We  may,  therefore,  take  our  departure  from  the  psychology 
of  general  ideas  in  which  the  problems  are  fairly  clear,  and 
in  which  some  conclusive  results  have  been  obtained.  With  these 
before  us,  we  may  then  turn  with  some  hope  of  success  to  the 
similar  problem  in  the  sphere  of  feeling.  The  most  important 
analytical  problem  in  the  psychology  of  general  concepts  may 
be  stated  in  the  following  way.  The  general  concept  stands  for 
a  meaning  which  has  been  acquired  in  the  process  of  judgment ; 
its  functional  correlate  is,  therefore,  habit  or  judgmental  dis 
position.  But  this  meaning  is  always  the  meaning  of  some 
psychical  content.  What  content  can  introspection  discover 
corresponding  to  this  meaning  ?  If  this  content  be  looked  for 
m  terms  of  images,  the  answer  is  given— there  is  no  content,  or 
t  is  inadequate  and  irrelevant.  The  final  question  then  is,'  in 
the  words  of  Stout,1  how  is  "  imageless  apprehension  "  possible  ? 

Considering  the  phases  of  this  problem  in  detail,  we  find 
that  for  the  most  abstract  concepts  there  is  no  image  equivalent. 
Words  such  as  liberty,  truth,  force,  may  pass  through  our  minds! 
be  intrinsically  apprehended,  and  leave  a  trail  of  meaning,  without 
calling  up  specific  images,  and  even  when  there  are  images,  these 
are  often  wholly  irrelevant  to  the  meaning.  Moreover,  in  the  case 
of  the  most  abstract  terms,  there  may  be  judgments  and  subsump- 
lons,  which  approximate  to  the  habitual  and  automatic,  in  which 
the  subject  finds  no  image  content  corresponding  to  the  term. 
Thus^to  sum  up  the  results  of  Ribot's  investigation  of  general 
ideas,  he  finds  no  characteristic  content  distinguishing  the 
formation  of  general  ideas,  and  for  the  more  abstract  notions 
no  analysable  content  whatever.  The  conclusion  is  that  the 

*  Stout,  Analytical  Psychology,  Vol.  I,  chap.  IV. 
Ribot,  The  Evolution  of  General  Ideas,  chap.  IV. 


126  Valuation:   its  Nature  and  Laws 

concept  can  be  formulated  only  in  functional  terms  and  must 
be  relegated  to  the  unconscious,  which,  on  his  principles,  means 
physiological  disposition. 

It  is  true  that  the  criticism  of  Royce  upon  these  conclusions 
of  Ribot — that  the  latter's  investigations  were  concerned  merely 
with  the  most  abstract  notions,  and  that  the  poverty  of  his 
results  is  due  rather  to  the  extreme  simplicity  and  automatic 
character  of  the  judgments  and  reactions  which  the  experimenter 
used  than  to  the  absence  of  a  psychical  correlate,  is  well  founded. 
It  is  true,  as  he  maintains,  that  when  judgment  takes  place  with 
effort,  after  inhibition,  in  the  attempt  to  assimilate  some  novel 
particular,  conscious  experience  of  the  process  is  present.  Never 
theless,  even  here,  in  the  instrumental  use  of  the  concept,  the 
particular  images  called  up  are  not  adequate  equivalents  for 
this  consciousness  of  meaning,  they  are  often  irrelevant — and 
besides,  there  still  remains  the  intrinsic  meaning  of  the  term, 
already  described,  which  is  present,  when  there  is  no  inhibition, 
and  which,  nevertheless,  has  no  image  equivalent.  The  problem 
still  remains  to  find  some  correlate  for  this  consciousness — the 
question  still  is :  How  is  imageless  apprehension  possible  ? 

The  approach  to  the  solution  of  this  problem  has  been  but 
gradual.  As  long  as  psychology  failed  to  go  beyond  the 
separate  presentations  and  their  external  associations  for  prin 
ciples  of  description  and  explanation,  it  could  by  no  possibility 
find  any  terms  of  description  for  this  additional  meaning.  Judg 
ments  appear  to  be  mere  associations  after  disjunction,  and  the 
general  concept  has  no  reality  except  as  physiological  dis 
position.  The  nearest  approach  along  the  old  lines  was  when  the 
equivalent  of  this  "  meaning  "  was  found  in  a  sort  of  composite 
photograph  of  particulars,  a  blurred  reproduction  which  is 
susceptible  of  recall  by  a  large  number  of  presentations,  and 
which,  in  the  process  of  recall,  receives  special  emphasis  upon 
that  aspect  with  which  the  new  particular  assimilates.  But  the 
irrelevance  of  these  images  in  some  cases,  and  their  total  ab 
sence  in  others,  mark  the  limits  of  this  conception.  The  attempt 
of  James  to  bring  this  "  meaning  "  within  the  ken  of  psychology 
by  use  of  the  metaphor  of  the  psychic  "  fringe  "  or  overtone 
to  describe  our  sense  of  the  halo  of  relations  about  an 
image,  has  been  of  great  historical  importance,  but  it  was  not 
until  there  appeared  a  much  wider  extension  of  the  concept 
of  content  beyond  the  limits  of  an  atomistic  psychology,  that 
justice  could  be  done  to  this  suggestion.  The  condition  of 


The  Continuity  of  Affective-Volitional  Meaning     127 

this  extension  was  the  recognition  of  the  fact  that  between 
content  and  function  there  is  no  fixed  gulf,  that  what  is  function 
on  one  level  becomes  conserved  as  content  on  a  higher  level — the 
recognition,  in  short,  of  the  genetic  point  of  view.  Now,  func 
tionally,  we  have  seen,  the  general  concept  is  a  judgment-dis 
position,  an  acquired  meaning  developed  in  the  processes  of 
judgment,  and,  judgment  being  a  developed  form  of  conation, 
the  positive  significance  of  the  general  concept  is  its  instrumental 
use  as  a  means  of  conative  unity  and  continuity. 

The  meaning  or  intent  of  a  general  notion  is  therefore 
not  the  object  of  presentation  at  all.  It  is  the  object  of  judgment. 
By  becoming  the  object  of  judgment,  however,  it  also  becomes 
a  content  of  a  higher  order.  This  meaning,  become  content,  is 
therefore,  as  we  have  already  seen,  relatively  independent  of 
particular  presentations  ;  indeed,  the  particular  presentations 
may  be  wholly  irrelevant,  and  the  meaning  may  inhere  in  mere 
words  which,  although  they  may,  in  their  origin,  have  been 
accompanied  by  associated  images,  are  now  wholly  independent 
of  them. 


Distinction  between  the  Instrumental  and  Intrinsic  Meaning 
of  the  General  Notion. 

It  is,  however,  quite  possible  to  over-emphasise  the  instru 
mental  character  of  the  general  notion.  This  has  been  a  notice 
able  characteristic  of  recent  discussions.  As  a  matter  of  fact, 
psychological  analysis  must  distinguish  two  types  of  meaning 
of  the  general  notion,  genetically  related  to  each  other,  an  in 
trinsic  as  well  as  an  instrumental.  Under  the  influence  of  sub 
jective  interests  and  dispositions,  an  aspect,  which  has  already 
recognitive  meaning,  is  separated,  abstracted  from  the  object, 
and  in  the  interest  of  conative  continuity  is  assumed  to  exist, 
is  given  a  quasi-existence.  This  assumption  of  the  first  explicit 
type  (arising  in  the  semblant  mode,  and  having  schematic 
meaning,  to  use  Baldwin's  terms)  passes  into  judgment  when 
the  assumed  existence  is  acknowledged  in  a  judgment  relating 
the  quasi-real  with  the  reals  of  immediate  perception.  But 
there  is  also  a  function  of  the  general  notion  in  which  its  meaning 
is  intrinsic,  as  in  the  cases  of  imageless  apprehension  already 
discussed,  in  which  words  such  as  truth,  liberty,  may  leave  a 
trail  of  meaning  in  our  minds  without  the  presence  of  images. 
The  point  of  difference  lies  in  the  fact  that  the  third  stage  of 


128  Valuation:   its  Nature  and  Laws 

the  accommodative  process  has  appeared.  The  judgment  dis 
position  created  by  repeated  judgments  of  an  instrumental 
character  has  given  rise  to  the  implicit  assumption  of  the 
second  type.  When  the  general  concept  contains  this  im 
plicit  assumption  of  existence,  as  in  the  case  of  the  belief  in 
abstract  truth  as  a  reality,  intrinsic  meaning  is  hypostatised  into 
a  reality.  The  fulfilment  of  such  an  assumption  lies  wholly  in 
the  continuity  of  the  intrinsic  meaning,  and  not  in  its  instru 
mental  reference  to  a  particular  situation.  This  meaning,  for 
which  an  "  objective"  is  assumed  to  exist — at  least  until  the  im 
plicit  assumption  is  disturbed  by  new  existential  or  truth  judg 
ments,  is  that  which  characterises  contemplation.  This  is 
the  origin  of  all  belief  in  the  reality  of  general  concepts.  We 
are  not  concerned  here  with  the  metaphysical  validity  of  such 
a  belief,  but  merely  with  the  psychological  origin  of  this  second 
intrinsic  meaning  of  the  general  concept,  for  we  shall  find  a 
similar  distinction  in  the  meanings  of  affective  generals.1 


(b)   The  Nature  of  Generic  Meanings  of  Feelings. 
Intensity-less  Appreciation. 

The  answer  to  the  question — how  is  imageless  apprehension 
possible  ? — has  led  to  a  development  of  a  positive  functional 
criterion  of  generalisation  which  supplements  the  negative 
criterion  of  relatively  imageless  apprehension.  The  problem 
of  the  psychical  correlate  of  the  meaning  of  the  affective  general 
or  abstract  presents  itself  in  a  somewhat  similar  way.  Here 
the  negative  criterion  is  intensity-less  appreciation,  and  the  cor 
responding  question  is,  how  is  this  intensity-less  appreciation 
possible,  or  if  we  use  emotion  in  its  limited  sense  of  concrete 
affective  disturbance — how  is  emotionless  appreciation  possible  ? 

In  the  first  place,  it  is  important  to  emphasise  the  fact  that 
this  lack  of  intensity  is  the  characteristic  attributed  by  those  who 
have  made  a  study  of  the  phenomena  to  the  phases  of  affectivity 

1  Thus,  as  Stout  says  of  the  conceptual  "now,"  the  word  "now"  may  stand  for  the 
moment  of  the  specious  present  with  its  sensation  content,  for  the  now  of  an  historical 
epoch,  of  a  year  or  a  day  with  their  varied  ideal  content.  These  are  instrumental 
functions  involving  judgment,  and  the  relative  constancy  of  the  meaning  of  the  con 
ceptual  "now"  is  not  a  function  of  the  sameness  of  its  content,  but  of  the  constancy  of 
the  judgmental  processes  of  which  it  is  the  objective  meaning.  The  word  "  now  "  has  a 
meaning  (as  in  the  case  of  the  other  words  studied)  even  when  none  of  this  varied  con 
tent  is  present,  when  it  is  founded  merely  upon  the  auditory  or  motor  sensations  which 
make  up  the  word.  This  is  its  intrinsic  meaning,  and  here  we  have  merely  the  vague 
assumption  of  an  objective  for  it,  of  an  existence  in  which  it  inheres. 


The  Continuity  of  Affective-Volitional  Meaning     129 

variously  described  as  affective  sign,  affective  abstract  or 
general.  When  subjected  to  structural  analysis,  these  feelings 
are  seen  to  differ  from  particular  feelings  in  the  fact  that 
they  lack  the  intensity  and  multiplicity  of  organic  sensations 
which  characterise  the  particular  emotion.  In  the  second 
place,  these  phases  of  affectivity  have  a  funded  meaning  which 
is  independent  of  increase  or  diminution  of  intensity.  In 
fact,  the  one  criterion  of  these  forms,  however  they  be  desig 
nated  (as  recognised  by  Paulhan,  Elsenhans,  and  Saxinger), 
is  that  they  are  not  subject  to  the  law  of  diminution  of  in 
tensity  with  repetition.  In  these  descriptions,  it  is  true,  no  clear 
distinction  is  made  between  intensity  and  degree  of  meaning, 
but  a  careful  scrutiny  shows  that  it  is  degree  of  funded  meaning 
which  they  find  unmodified  by  repetition.  On  our  view,  which 
identifies  intensity  with  intensity  of  organic  sensitivity,  the  term 
intensity-less  appreciation  means  that  the  degree  of  intensity  of 
the  sensitivity  in  which  the  feeling  is  embodied  is  irrelevant  for 
the  worth  suggestion  of  the  feeling.  The  question  then  is,  how 
can  the  meaning  of  emotional  reactions  remain  after  the  particu 
lar  organic  sensitivity,  with  its  intensity,  and  the  particular 
percepts  which  called  out  the  emotion  have  disappeared  ? 

Let  us  first  examine  these  generic  phases  of  affectivity  more 
closely.  The  existence  of  words,  with  cognitive  meaning  in 
dependent  of  particular  images,  was  the  starting-point  of  the 
analysis  of  the  psychology  of  the  general  concept.  Words  with 
emotional  connotation,  when  there  is  no  specific  image  or  emotional 
content,  may  properly  be  the  starting-point  for  our  study  of 
the  affective  abstract.  With  the  emotional  connotation  of 
words  in  its  most  patent  aspects  we  are  tolerably  familiar. 
Burke  tells  us  that  he  found  it  "  hard  to  persuade  several  that 
their  passions  (a  general  term  for  feeling)  are  affected  by  words 
for  which  they  have  no  ideas,"  but  that  was  at  a  time  when  a 
purely  intellectualistic  psychology  as  well  as  art  made  im 
possible  a  true  perception  of  the  facts.  Certain  forms  of  modern 
art,  as  well  as  a  more  subtle  and  sophisticated  analysis  of  our 
own  experience,  make  us  now  thoroughly  cognisant  of  the  general 
fact.  Ribot  has  taken  as  the  most  primitive  type  of  emotional 
logic  (that  is  mental  movement  which  has  as  its  mean  term 
the  emotional  rather  than  the  cognitive  connotation  of  a  term), 
what  he  describes  as  the  logic  of  persuasion  or  appeal.1  It  is 
"  most  primitive  "  because  it  diverges  least  from  the  conceptual 

1  La  logique  des  sentiments,  Paris,  Felix  Alcan,  1905. 


I3o  Valuation:    its  Nature  and  Laws 

logic-  yet  the  transitions,  while  giving  rise  to  the  illusion  of  cog 
nitive  mean  terms,  are  for  the  most  part  emotional. 

But  still  more  marked  is  the  purely  emotional  connotati- 
of  words  in  their  intrinsic  appreciation,  where  they  pass  through 
the  mind,  leaving  a  trail  of  appreciation  without  raising  ideas 
or  definite  emotional  responses.     Such  use  of  words  for  tl 
purely   emotional  connotation,   as  in  symbolist  poetry,  Kit 
has  taken  as   a  type  of   affective  logic  almost  pure    and 
these   intrinsic   emotional    appreciations   of   words    1 
tones,  so  to  speak,  which  Elsenhans  takes  as  the  basis  f 
conception  of  generalised  emotions. 

Words   then   may  have  an  emotional  connotation  for  whi 
there   is   no    adequate    emotional    content-equivalent    and   n< 
adequate  presentational  presuppositions.     But  there  are  oth 
generic  phases  of  emotionalism  which  do  not  have  even  this 
embodiment    in    words.     Sentiments    are    almost    wholly    con 
nected  with  words,  but  moods  and  affective  signs  may  lac 
even  this  embodiment.     Of  chief  importance  in  this  connect] 
are  the  so-called  sentiments  or  moods  of  places.     Here  a  certain 
feeling  attitude,  with  definite  recognitive  meaning,  includes  in 
it  the  funded  meaning  of  past  emotional  reactions  which,  H 
felt    could  be  individually  revived  with  the  revival  of  the  par 
ticular  presentations.     This  is  also  characteristic  of   the  moc 
we  carry  away  from  the  reading  of  a  poem,  as  illustrate 
the  Lotos-Eaters  already  referred  to.1 

From   these   illustrations   it   appears   that,   as   in    the 
of   imageless   apprehension   of   cognitive   meaning,   so 
intensity-less  appreciation  of  affective-volitional  meaning   ther 
are  two  types,  the  instrumental  and  the  intrinsic.    And  this  fact 
underlies    I   think,   the   distinction   frequently   made   between 
affective  generals  and  affective  abstracts.     In  the  case 
the  feeling  with  generic  meaning   functions  as  a  substit 
representative  (either  as  recognitive  of  former  particular  emotion 
or  as  anticipatory  of  concrete  feeling  situations),  it  may 

outlines  have  long  since, disappeared  from  °2£^^O?dS£Sto  the  whole  of 
impression,"  we  make     combination  and  tusic      01 :  m>  ,         j  t     its  constituent 

a  supersensible  intuition,  which  we  but  re luctan ^*&™      me  !Lm  have  the  writers 

of  its  significance  for  the  psychology  of  feeling. 


The  Continuity  of  Affective-  Volitional  Meaning     131 

spoken  of  as  abstract  or  instrumental.  On  the  other  hand,  in 
the  case  of  sentiments  of  places,  or  sentiments  attached  to  words, 
the  meaning  is  generic  but  intrinsic. 

In  all  these  phenomena  what  we  have  described  as  "  intensity- 
less  appreciation  "  is  the  negative  characteristic,  and  may  be 
said  to  be  the  negative  condition  of  their  generic  character  and 
function.  In  the  several  cases  we  have  considered — the  emo 
tional  connotation  of  a  word,  representing  repeated  particular 
judgment  feelings  of  the  past,  the  "  affective  sign  "  as  mean 
term  of  affective  transitions,  the  generic  and  schematic  sentiment 
of  places — all  owe  their  generic  and  representative  capacity  to  the 
fact  that,  because  of  some  specific  difference  in  functional  pre 
supposition,  they  do  not  pass  over  into  the  full  emotional  res 
onance  which  characterises  the  particular  emotional  reaction 
to  a  definite  situation.  They  are  in  the  meaning  of  our  terms 
intensity-less  appreciations.  To  account  for  this  difference 
in  functional  presuppositions  is  to  account  for  these  phenomena. 

3.    The  Process  of  Affective  Generalisation — Acquirement 
of  Generic  Feeling. 

With  this  analysis  of  the  facts  before  us,  it  is  now  possible 
to  understand  the  process  by  which  a  concrete  feeling-attitude 
acquires  generic  meaning.  Evidently  the  process  has  two 
aspects,  one  describable  in  terms  of  change  of  function,  and  the 
other  in  terms  of  modification  of  structure.  On  the  one  hand, 
how  does  the  meaning  become  independent  of  specific  image 
content  and  cognitive  reaction  ?  On  the  other  hand,  how  does 
the  funded  meaning  become  independent  of  sensitivity  and  its 
intensity  ?  The  first  aspect  is  evidently  closely  connected  with 
the  second. 

The  process  of  acquirement  of  generic  meaning,  or  of  affective 
generalisation,  may  be  described  as  abstraction  from  the  indi 
vidual  presuppositions  of  the  feeling,  and  the  substitution  of  an 
assumption  of  existence  for  specific  judgment.  Feelings  with 
generic  meanings  are  all  assumption  feelings,  and  their  intensity- 
less  character  may  be  ascribed  to  the  different  attitude  toward 
reality  involved. 

In  understanding  this  proposition  it  is  first  important  to  note 
that  the  acquirement  of  recognitive  meaning  is  the  beginning  of 
acquirement  of  generic  meaning.  The  recognitive  meaning,  func 
tionally  viewed,  consists  in  the  fact  that  the  feeling  of  the  present 


Valuation:   its  Nature  and  Laws 

stands  for  an  actual  feeling  of  the  past  into  which,  it  is  believed, 
it  is  convertible.     It  has  a  representative  capacity.     But  this 
same  feeling,  thus  loosed  from  its  immediacy  by  change  of  pre 
suppositions  from  judgment  to  assumption,  has  also  acquired 
an  anticipatory  character,  a  reference  to  future  actual  judgment 
feelings  into  which  it  can  be  converted.     It  is  here  that  its 
generic  character  is  found.     Such  a  feeling  anticipates  or  repre 
sents—not  one  definite  actual  emotional  reaction  with  its  unique 
and  individual  presupposition— but  a  general  situation  which 
may  be  specialised  in  various  particular  emotions  according  to 
conditions.     The  derived  forms  of  affectivity,  sentiment,  mood, 
and  still  more,  affective  sign,  have  this  recognitive  and  generic 
meaning  in  varying  degrees.     In  general,  then,  the  rise  of  generic 
meaning  is  conditioned  by  abstraction  of  feeling  from  its  indi 
vidual  presuppositions  and  the  substitution  of  assumptions. 

But  this  generic  meaning  we  have  found  to  be  of  two  types, 
instrumental   and   intrinsic,    and   one   of   these,    the   intrinsic, 
is  closely  connected  with  habit.     Every  assumption-feeling  has, 
as  such   a  generic  character,  in  that  it  is  anticipatory  of  specific 
judgment-feelings  which  it  qualifies  when  they  become  actual- 
ised.     But,  as  we  have  already  seen,  there  are  two  types  of 
assumption,    the    explicit   and   the   implicit.     The  first   is   an 
accommodation  in  the  form  of  imaginative  projection  as  deter 
mined   by   subjective    interest.     The    second   is   habit    follow 
ing  upon  repetition  of  judgment.     The  explicit  assumption  is 
always  anticipatory  of  judgment  to  come,  and  is  therefore  in 
strumental.     The   implicit   assumption   and   its   feeling   is   1 
psychical   correlate   of  judgment-disposition,   and   its   meaning 
is   intrinsic.     That   which   these   two   types   have   in   common 
is  the  important  thing,  and  this  is  the  fact  that  both  are  re 
presentatives,    affective    signs,   for   actual  situations    in  which 
existence  is  presumed  or  explicitly  acknowledged  in  judgment. 
Their  representative  function  is  made  possible   through  their 
schematic,  intensity-less  character  already  described.     Both  may 
therefore  properly  be  characterised  as  abstractions,  in  that  their 
generic  meaning  is  acquired  through  abstraction  from  individual 
presuppositions  and  by  the  reduction  of  those  aspects  of  feeling- 
attitude  which  make  them  specific  emotions  or  passions. 

The  difference  between  the  two  types  lies  solely  in  their 
origin  and  function.  There  are  two  ways  in  which  a  feeling 
is  evoked  by  assumption:  (a)  imagination  of  the  existence  of  an 
object  or  class  of  objects,  in  which  process  the  feeling  becomes 


The  Continuity  of  Affective-Volitional  Meaning     133 

generic  and  schematic ;  or  (b)  by  repeated  judgments  passing 
over  into  judgment  habit  and  implicit  assumption,  when, 
embodied  in  general  terms,  in  mere  words,  it  is  an  affective 
sign  for  particular  judgment-feelings. 

Specific   passions    and   emotions   presuppose   explicit   judg 
ment.     Sentiments,  moods,  and  affective  signs  may  be  realised 
on  the  mere  assumption  of  the  object,  the  object  being  often 
the  merest  and  vaguest  universal.     Thus  it  is  that  these  ab 
stracts  inhere  in  mere  words.     The  words  God,  love,  liberty, 
have  a  real  emotional  connotation,  leave  a  trail  of  'affective 
meaning,  because  they  stand  for  an  object  which  is  implicitly 
assumed  to  exist,  an  assumption  generated  by  previous  judg 
ment    reactions.     Let,    however,    this    assumption    pass    into 
judgment,    let     the     explicit    judgment,    I    am    free    or    not 
free,  this  is  or  is  not  my  duty,  God  exists  or  does  not  exist, 
arise  through  some  arrest  of  my  habitual  attitude,  and  again 
passion   and   emotion   appear  with   an  intensity  which   is   de 
termined  by  the  desire  presupposed  and  the  degree  of  arrest. 
We  may  quite  properly  speak  of  the  emotional  connotation  of 
such  words  as  the  funded  meaning  of  previous  emotional  re 
actions  and  the  affective  abstracts  which  constitute  the  psy 
chical  correlates   of  this   meaning   as   the  survivals  of  former 
judgment-feelings.     These  residual  feelings  may,  from  the  point 
of  view  of  function,  be  described  as  affective  abstracts  precisely 
because,  while  every  particular  emotional  attitude  presupposes 
the  apprehension  and  acknowledgment  of   the   existence   of   a 
particular  object,   the  affective  abstract,  the  "  sign  "  of  these 
emotional   attitudes    may   be  experienced  with    all   its   worth 
suggestion  independently  of  the  presupposition   of  the  actual 
feeling. 

IV.  GENERAL  THEORY  OF  THE  ROLE  OF  AFFECTIVE  GENERALS 
IN  PROCESSES  OF  VALUATION— ILLUSTRATIONS 

r.  Criticism  of  the  view  that  they  do  not  determine  particular 
feelings,  are  not  Feelings  of  Value. 

A  view  in  many  respects  similar  to  the  preceding  is  the 
theory  of  Phantasiegefuhle,  feelings  of  the  imagination,  de 
veloped  by  Meinong  and  Saxinger,  and  already  referred  to  in 
an  earlier  paragraph.  This  view  demands  a  special  examina 
tion  here  for  the  reason  that,  while  many  of  the  facts  of  analysis 


1 34  Valuation  :   its  Nature  and  Laws 

are  true    and  while  their  explanation  as  assumption-feelings- 
to  which  I  owe  many  suggestions  for  the  theory  here  developed- 
is  sound  enough,  nevertheless  the  conception  of  their  role 
the  processes  of  worth  determination  is  open  to  serious  cnt 


cism. 


11.  ,    c 

In  so  far  as  introspective   analysis  is  concerned,  Saxmger 
practically  recognises  the  existence  of  such  phases  of  affective- 
experience   as   we   have   described   under   the   terms    affective 
abstract  and  affective  sign,  but  he  gives  them  another  explana 
tion       Elsenhans's  generalisation   and  Ribot's    abstraction 
feeling   have   reference,   he   tells    us,    to    the    same    emotional 
phenomenon  in   different   contexts.      The    generalised    feelings 
are   feelings   of    imagination    following    upon    assumption, 
abstract  feelings  are  feelings  of  imagination  attaching  directly 
to  "substrate-ideas."     For   these  imagined  feelings  he  finds  i 
necessary  to  assume  the  existence  of  dispositions  entirely  distinct 
horn  those  presupposed  by  actual  feelings,  and  accounts  for  1 
direct  attachment  of  imagined  feelings  by  assuming  that 
dispositions   may   be   actualised   by  substrate-ideas   or  general 
concepts,  as  well  as  by  assumptions.     Feelings  of  the  imagina 
tion   differ   from  actual  feelings  in  that  their  intensity  is  not 
modified  by  repetition   and  in  that  the   corresponding  desires 
of  the  imagination  do  not  cease  with  fulfilment.     The  essential 
characteristic   of    this   view  is   to  be   found  in   the   denial    t 
these  feelings  of  the  character  of  real  feelings,  and  therefc 
of  feelings  of  value,  for  the  reason  that  they  are  not  actual, 
but   imagined   feelings    (ScheingefiMe) ;    and   in   the  denial,  i 
the  second  place,  that  these  feelings  influence,  or  are  influence 
by   real  feelings.     This  last  point  is  the  mam  issue. 

Saxinger's    grounds    for    this    view1    are    both    theoretical 
and   experiental.     The   theoretical   may  be  dismissed  in  a  few 
words    for  they  rest  upon  certain  doubtful  assumptions,  and 
in  general  upon  a  conception  of  feeling  which  we  have  foum 
untenable.    In  the  first  place,  it  is  assumed  that  feeling  is  wholly 
subjective,    wholly    different    from    sensitivity,    and 
incapable  of  undergoing  processes  analogous  to  generalisation 
and  abstraction,  incapable  of  acquiring  recognitive  and  selective 
meaning,  a  view  which  our  entire  analysis  has  led  us  t 

i  R.  Saxmger,  Beitrage  zur  Lehre  von  der  emotionalen  Phantasie ;   Zt itschrift  fur 

^S^^  ' 

theorie  und  Psychologic,  Leipzig,  1904. 


The  Continuity  of  Affective- Volitional  Meaning     135 

In  the  second  place,  and  closely  connected  with  this,  is  the  as 
sumption  that  intensity  and  degree  of  meaning  of  feeling  are 
identical,  and  that  therefore  the  criterion  of  the  real  feeling 
is  diminution  of  intensity  with  repetition.  Any  phenomena 
which  do  not  follow  this  law  must  accordingly  be,  not  real  feelings, 
but  some  sort  of  quasi-feelings  which  he  describes  as  feelings  of 
the  imagination.  This  criterion  of  real  feelings  we  must  also 
question,  in  view  of  our  previous  analysis  of  feeling. 

The  facts  of  experience,  on  the  other  hand,  upon  which 
he  bases  his  conclusions,  require  closer  examination.  He 
finds,  as  we  do,  these  phases  of  affectivity  similar  to  other 
feelings  in  all  respects,  except  that  they  do  not  have  that 
sensational  intensity  which  characterises  the  so-called  actual 
feelings,  and  that,  in  their  case,  repetition  does  not  affect  their 
meaning.  These  facts  he  accounts  for  in  terms  of  their 
difference  in  presuppositions.  They  are  all  assumption  feelings, 
having  dispositions  different  from  those  which  give  rise  to 
actual  feelings.  He  believes,  moreover,  that  an  examination 
of  the  facts  discloses  an  independence  of  real  and  imagined 
feelings  which  makes  necessary  this  hypothesis  of  different 
and  independent  dispositions. 

As  to  the  question  of  fact,  we  may  admit  that  much  that 
Saxmger  has  brought  forward  gives  a  certain  plausibility  to 
tms  view,  but  only,  I  think,  when  the  facts  are  mis-interpreted. 
Thus  we  may  admit,  for  instance,  the  truth  of  his  statement  that, 
while  with  time  the  feeling  of  sadness  following  upon  the  death 
of  some  loved  one  may  lose  its  sting,  that  is,  may  decrease 
in  intensity,  it  is  quite  possible  that  the  joy  in  imagining  the 
existence  of  the  departed  loved  one  may,  within  limits,  be  un 
affected  by  this  dulling  of  the  real  feelings.  We  may  also  admit 
the  truth  of  his  illustration  where  the  father  gets  a  certain 
joy  in  imagining  his  son  successful,  while  his  feelings,  determined 
by  the  certain  knowledge  that  his  son  is  a  failure,  remain  un 
affected  by  the  feelings  of  imagination  following  upon  the  as 
sumption.  A  relative  independence  of  these  two  phases  of 
experience,  and  the  possibility  of  their  existing  together,  cannot 
be  denied.  But  it  would  be  a  doubtful  inference  to  conclude 
that  they  are  ultimately  independent.  Such  independence  is 
the  most  only  temporary.  I  am  inclined  to  think  that  in 
both  cases  the  capacity  for  such  feelings  of  imagination  steadily 
decreases  with  the  recognition  of  the  inevitableness  of  the 
actual  fact.  In  so  far  as  my  own  introspection  goes,  with  the 


1 36  Valuation  :   its  Nature  and  Laws 

repetition  of  judgments  of  non-existence  and  their  negative 
worth  feelings,  the  capacity,  as  well  as  the  desire,  for  visual 
isation  and  assumption  of  existence  of  the  objects  which  give 
rise  to  the  feelings  of  imagination,  weaken,  and  the  former 
feelings  remain  only  as  a  mood  or  affective  sign  which  a  word 
or  name  suggests.  This  does  not,  however,  exclude  the  possi 
bility  of  involuntary  recrudescence  of  the  imagined  feeling  under 
certain  favourable  conditions  of  general  mood  or  association. 
The  facts  of  unmediated  revival  of  emotions,  examined  in  the 
study  of  affective  memory,  are  too  unequivocal  to  admit  of  such 
denial. 

Saxinger  recognises,  indeed,  such  cases  of  apparent  in 
fluence,  both  of  assumption  feelings  upon  judgment  feelings 
and  judgment  feelings  upon  assumption  feelings,  but  thinks 
that  such  influence  is  only  apparent,  and  must  be  explained  in 
another  way.  The  question  arises  in  his  discussion  especially 
in  connection  with  the  view  of  Witasek,  that  the  feelings  of 
the  imagination  in  aesthetic  experience  are  influenced  by  the 
actual  sentiment  or  mood  in  which  the  subject  is  at  the  time 
of  the  aesthetic  experience.  It  is  a  well-known  fact  that  some 
moods  and  sentiments  are  completely  antagonistic  to  these 
feelings  and  emotions  of  art.  Thus,  if  we  are  in  a  mood  of  de 
pression,  we  cannot  realise,  even  imaginatively,  the  joyous 
feelings  expressed  by  a  work  of  art.  In  certain  moods  tragic 
emotions  are  unreal  and  even  absurd.  Saxinger  finds  an  ex 
planation  for  these  facts  in  the  theory  that  we  have  here  not 
the  influence  of  one  feeling  upon  another,  but  the  influence  of 
intellectual  processes  of  judgment.  Certain  judgments,  with 
their  accompanying  feelings,  make  impossible  the  activities 
of  assumption,  and  therefore  the  feelings  of  imagination. 
But  precisely  in  this  abstraction  of  the  feelings  from  their  in 
tellectual  presuppositions  we  have  the  weakness  of  the  argu 
ment.  Throughout  our  study  of  feeling  and  desire  we  have  in 
sisted  that  an  essential  characteristic  of  the  feeling  or  desire, 
as  the  case  may  be,  is  its  cognitive  presupposition.  Feelings, 
as  we  have  shown,  do  not  determine  feelings  causally,  nor  do 
feelings  determine  desire.  All  determination  of  affective- 
volitional  meaning  is  through  change  in  cognitive  presuppo 
sitions.  When,  therefore,  the  presuppositions  of  one  feeling 
make  impossible  or  otherwise  influence  the  presuppositions  of 
another,  we  have  influence  of  feeling  upon  feeling  in  the  only 
sense  that  such  an  expression  has  meaning. 


The  Continuity  of  Affective- Volitional  Meaning     137 

Finally,  in  discussing  this  influence  of  one  type  of  feeling  on 
the  other,  Saxinger  has  not  considered  at  all  the  case  of  those 
phenomena,  the  affective  signs  of  words,  which  he  describes 
as  feelings  of  imagination  attaching  directly  to  "  substrate- 
ideas,"  and  which,  on  our  theory,  are  assumption  -  feelings 
of  the  second  type,  where  the  assumption  is  implicit,  following 
upon  presupposed  judgmental  habit.  An  examination  of  these 
phenomena  would  have  shown  him  that  they  constantly  in 
fluence  his  so-called  real  feelings  and  desires.  Their  influence 
is  seen  in  two  significant  situations  :  (a)  where  they  act  as  im 
pelling  sentiments  or  affective  signs,  negating,  inhibiting  par 
ticular  desires  and  emotions  ;  and  (6)  in  those  cases  of  sub- 
sumption  of  the  ethical  type  where,  giving  their  meaning  directly 
to  a  particular  reaction  of  desire  or  feeling,  without  any  mediating 
associations,  they  increase  its  energy  or  affective-volitional 
meaning. 

Throughout  his  entire  discussion  of  these  phenomena  Sax 
inger  has  entirely  failed  to  recognise  the  genetic  relation  of 
assumptions  of  both  types,  explicit  and  implicit,  to  judgment, 
a  relation  which  the  analysis  of  our  first  chapter  made  clear, 
and  which  has  been  persistently  emphasised  in  the  discussions 
that  follow.  Because  of  their  genetic  relations  to  particular 
judgment  feelings,  these  assumption  feelings  are  representative, 
both  in  a  recognitive  and  anticipatory  capacity,  of  actual  feelings, 
and  determine  them  in  ways  already  described,  and  to  be 
shown  in  more  detail  presently.1 

2.   The  Role  of  Affective  Generals  in  Worth 
Determination. 

How,  then,  shall  we  characterise  the  role  of  these  phases 
of  affective  experience  in  the  continuity  of  affective-volitional 
meaning  ?  If  the  results  of  this  criticism  of  Saxinger  are  valid, 
if  these  phases  do  determine  actual  feelings,  that  is,  immediate 

1  In  this  connection  a  paragraph  from  James's  recent  presidential  address,  The 
Energies  of  Men,  is  suggestive.  "As  certain  objects  awaken  love,  anger,  or 
cupidity,  so  certain  ideas  naturally  awaken  the  energies  of  loyalty,  courage,  endurance, 
or  devotion.  When  these  ideas  are  effective  in  an  individual's  life,  their  effect  is  often 
very  great  indeed.  They  may  transfigure  it,  unlocking  innumerable  powers  which,  but 
for  the  idea  would  never  have  come  into  play.  'Fatherland,'  'The  Union,'  'Holy 
Church,'  the  '  Monroe  Doctrine,'  '  Truth,'  '  Science,'  '  Liberty,'  Garibaldi's  phrase 
'  Rome  or  Death,5  etc.,  are  so  many  examples  of  energy-releasing  abstract  ideas.  The 
social  nature  of  all  such  phrases  is  an  essential  factor  of  their  dynamic  power.  They  are 
forces  of  detent  in  situations  in  which  no  other  force  produces  equivalent  effects,  and 
each  is  a  force  of  detent  only  in  a  specific  group  of  men." 


T,8  Valuation:    its  Nature  and  Laws 

O 

reactions    to    concrete    situations,    if    the    distinction    between 
real  feelings  and  feelings  of  the  imagination  is  not  ultimate 
but  the  result  of  a  neglect  of  their  genetic  relations,  and 
finally   the  inference  as  to  the  existence  of  distinct  dispositions 
is  not  necessary,  there  is  nothing  in  the  way  of  our  conception  o 
an     affective  logic."     On  the  other  hand,  our  structural  analysis 
"these  Chases  "(and  our  functional  analysis  of  their  origin)  has 
shown  the  validity   of   their   description   as  generic   forms   < 
affectivity,  and  affords  a  basis  for  the  concept  of  affective  logic 
We  may  now  seek  to  develop  in  more  detail  the  quasi-logical 
relation  of  these  phases  to  the  so-called  actual  feelings. 

Our  main  proposition  is  that  these  affective  generals  and 
affective  signs  are  the  bearers  of  a  funded  meaning  which  is 
tndependen?  of  their  intensity,  and  that  they  represent  par- 
ttTar    feehngs-in    relations    of    substitution,    subsumptjon 
and    mediation-in    the    continuity    of    valuation.     It    b    this 
representative  function  which  should  be  emphasised.    Saxmger 
recognises     indeed,    the    representative    character    of    feeling, 
but  do      not  see  that  this  involves  a  real  function  in  the  pro 
cesses  of  valuation.     "  Assumption-feelings,"  he  tells  us,      take 
the    place    of    feelings    of    value.     The    objects    are    measured 
according  to  their  subjective  worth  when  their  existence  „  non- 
existence  is  assumed."     Meinong  ,s  more  alive  to  this  real  i 
tion  of  the  assumption  feelings  of  the  two  types  in    he  processes 
of  valuation.     He  makes  a  distinction  between   Werther 
Werthalten,  and  recognises  a  subjective  Werthcn  in  *e  case  of 
assumption    feelings  and   a   more   objective  Werthalten 
case  of  judgment  feelings.1     He  even  goes  so  far  as  to  give  a 
certain  advantage  to  these  representative  feelings  m  the  process 
o   valuation,  m  that  they  are  often  present  when  *e  condmon 
of  judgment  are  impossible.     The  importance  of  these  qu. 
tlingf  as  he  calls  them,  for  the  continuity  of  affective-volitional 


e  the   function    of    affective 

generals   and   affective   signs   in   the   following   general 
The  assumption-feeling,  whether  a  feeling  following  upon  an 
actual  explicit   assumption  of  the   existence  or  ™";e^enc 
of  an  object,  or  attached  directly  to  a  word  or  general 
with    its   assumption   of    a    corresponding    objective,     H    con 
nected  with  an  habitual  judgment  wh.ch  has  passed  ove    mt  » 
an  implicit  assumption  of  existence,  as  m  the  case  of 

•  Meinong,  Utter  Annahnun,  Leipzig,  1902,  *apter  v.il,  »  53,  54,  55- 


The  Continuity  of  Affective -Volitional  Meaning     139 

representative  of  our  actual  capacity  for  a  particular  worth 
feeling  in  a  particular  situation.  If  the  assumption  is  explicit 
it  tells  us,  in  imagination,  how  we  should  feel  in  the  actual  situa 
tion.  Attached  directly  to  a  word  or  general  concept,  it  stands  for 
a  series  of  j  udgment- feelings  in  the  past.  Attached  to  an  habitual 
judgment  or  belief  of  the  present  it  is  the  representative  of  a 
j  udgment-disposition . 

3.  Illustrations  of  the  Role  of  Affective  Generals  in  Worth 
Continuities. 

This  general  statement  may  be  made  more  explicit  by 
applying  it  to  certain  concrete  cases  of  affective  substitution, 
subsumption,  and  transition,  with  which  our  studies  began. 
We  may  note  especially  the  cases  of  ethical  and  aesthetic  sub- 
sumptions  where  a  generic  phase  of  emotion,  sentiment,  mood, 
or  affective  sign  gradually  takes  the  place  of  particular  ex 
plicit  emotion,  and  assimilates  new  emotional  attitudes  to  it, 
thus  affording  the  basis  for  transitions  of  feeling. 

In  the  case  of  ethical  sanction  we  have  a  union  of  assumption 
and  judgment.  All  affective  abstracts  are  as  such,  as  "  in 
tensity-less  appreciation,"  assumption-feelings,  while  judgment- 
feelings  are  particular  passions  and  emotions,  and  therefore 
show  the  aspects  of  intensity  and  multiplicity  of  content.  Now 
in  an  ethical  situation  which  is  emotional  an  existential  judg 
ment  is  always  presupposed  and  we  have  always  a  particular 
passion  or  emotion.  But  we  are  familiar  with  changes  of  ethical 
attitude  where  a  new  attitude  becomes  merged  into  an  old,  or 
where  the  old  remains  as  a  qualifying  or  sanctioning  undertone 
of  the  new.  Such  cases  we  found  in  the  phenomena  of  holy 
anger,  or  of  a  sexual  love  becoming  predominantly  maternal  in 
its  nature.  Examination  of  these  subsumptions  (which  we  found 
are  not  mixed  feelings),  shows  the  situation  to  be  of  the  following 
type.  A  new  object,  or  new  aspect  of  an  old  object,  becomes 
the  object  of  an  existential  judgment,  and  therefore  of  a  new 
•emotion,  while  the  old  judgment  does  not  immediately  vanish, 
but  is  represented  by  an  assumption  which  has  an  affective 
abstract  as  its  correlate.  In  the  case  of  holy  anger  the  ex 
istential  judgment  about  God  is  in  abeyance ;  is  for  the  time 
represented  by  an  assumption.  The  feeling  corresponding  to  it 
is  an  abstract,  generic  religious  feeling,  while  the  concrete  object 
is  judged  and  reacted  to  emotionally.  Now  the  importance 


1 4o  Valuation:   its  Nature  and  Laws 

of  the  analysis  lies  in  the  fact  that,  while  the  resultant  attitude 
seems  to  be  a  fusion,  as  a  matter  of  fact  it  is  a  real  subsumption, 
for  it  is  on  the  basis  of  the  capacity  of  retaining  the  assumption 
that  the  union  is  possible.  Should  the  attendant  circumstances 
of  the  anger  be  such  as  to  suppress  the  underlying  religious 
assumption,  the  religious  tone  would  faU  out  In  the  other 
case  the  presence  of  the  feelings  of  the  wife  and  the  mother  in 
the  same  attitude  toward  the  husband,  the  existential  judg 
ments  which  call  out  the  particular  emotions  are  directed  upon 
aspects  of  the  object  which  appeal  to  the  maternal  instinct, 
but  the  disposition  formed  by  the  marital  relation  persists  as 
an  undertone  in  a  feeling  which  has  a  vague  assumption  as 

presupposition. 

One  more  illustration,  one  frequently  used  as  an  extreme 
example  of  mixture  of  apparently  opposing  feelings,  but  whi 
is  really  a  limiting  case  of  subsumption,  will  show  the  situation 
clearly      Rousseau's    Confessions,   which   remains   a   classic 
appreciative    introspection,    describes    how    the    repeated 
sumption  that  Madame  Martens  was  his  mother  passed  over 
into  a  situation  where  mere  imagination  became  almost  betel, 
and  the  correlated  filial  feelings  followed.     But  when  the 
lation  altered  and  became  a  liaison,  at  first  the  new  feelings 
were  coloured  by  the  sanctities  of  the  old  attitude  which  had 
now  passed  back  into  an  assumption.     Such  a  colouring  die 
not  he  tells  us,  and  indeed  could  not,  last.     The  assumption, 
with  its  accompanying  feeling,  was  forced  out,  and  in  the  pr< 
cess  he  suffered  the  pangs  of  remorse.     In  this  extreme  case, 
which    indeed    seems    perhaps    pathological,   we    have    never 
theless  a  type  of  the  emotional  logic  or  fallacy,  as  the  case  may 
be  which  is  characteristic  of  all  ethical  sanction  or  re-adaptation. 
When,  for  instance,  a  man  wakes  to  find  his  beliefs,  which  are 
largely  emotional,  changed,  it  means  simply  that  the  old  belief 
has  for  a  long  time  been  merely  an  assumption  which  now  at 
last  a  new  judgment  has  finally  suppressed. 

Esthetic  subsumption  is  of  a  different  type.  Here  the  im 
mediate  object  of  the  feeling  attitude  is  merely  presented,  not 
judged,  although  there  are  cases  where  judgments  enter  in  as 
partial  presuppositions.  In  the  cases  of  the  Lotos- Eaters  and 
the  Raven,  already  described,  the  actual  objects  of  the  pa 
ticular  feelings  are  all  objects  of  presentation,  not  of  judgment. 
It  would  however,  as  we  have  shown,  be  superficial  analysis  tc 
deny  the  presence  of  conation  and  desire,  at  least  dispositionally. 


The  Continuity  of  Affective-Volitional  Meaning     141 

In  these  cases  the  disposition  is  represented  by  an  assumption 
in  the  one  case  of  the  existence  of  the  desired  rest,  and  in 
the  other  of  irrevocable  fate.  The  mood  in  each  case  is  the 
affective  correlate  of  the  underlying  assumptions.  Now  genetic 
ally  these  assumptions  are  the  products  of  previous  actual 
desires  and  existential  judgments— only  thus  could  the  objects 
of  the  assumptions  have  the  interest  to  hold  the  esthetic  at 
tention,  but  the  possibility  of  holding  all  the  various  particular 
emotions  in  the  unity  of  the  mood  is  conditioned  by  the  fact 
that  the  mood  modifies  the  feelings  connected  with  the  specific 
imagery,  arrests  their  tendency  to  pass  over  into  specific  actual 
feelings,  i.e.,  to  acquire  an  intensity  which  would  disturb  the 
aesthetic  illusion. 


CHAPTER   VI 

THE  LAWS   OF   VALUATION 

I    LAWS  OF  VALUATION-THEIR  NATURE  AND  RANGE 
OF  APPLICATION 


r       - 

discovery  of  laws  ac  -g  ;  i  J  euivalence  or 


by   the  worth 
are  the  laws  governing  the  change  rn  acua  ^ 


if  the  relation  of  the  object  to   ts  wo        "n  die    and 

and  tins  was  certamly  the  cone  us.ono    our  ^-^^  ^ 
if,  further,  some  of  these  ,.      '   .    meanings    of    these 

sensations    or    P-sen"    at  ™.~Sior  all  types 

formulatlons 


142 


The  Laws  of  Valuation  143 

been   made,    although   upon   an   inadequate   analysis    of   worth 
experience,  and  therefore  have  a  definite  place  in  worth  theory 
a  critical  examination  of   their   foundations   and  the  extent  of 
their  application  constitutes  our  first  problem. 

i.  The  Laws  of  Valuation  applicable  to  Extra-Economic 

Values. 

These   psychological   laws   have   been   sought   primarily   by 
the  science  of  economics  for  the  purpose  of  measuring  the  prefer- 
ability  of  one  object  over  another  within  a  restricted' field  of 
goods.     The  method  of  investigation  is  to  study  the  laws  of 
preferabihty  growing  out  of  the  laws  governing  the  consumption 
!  separate  goods  by  an  individual— i.e.,  the  laws  of  subjective 
value— on  the  assumption,  made,  apparently,  by  most  writers  in 
:onomics,  that  these  values  are  unmodified  by  the  individual's 
participation  in  the  economic  activities  of  a  group.     The  laws 
of  objective  value   may  then  be  developed   from   the   laws  of 
subjective  value.     In  general  it  may  be  said  that  this  means  of 
imphfying  the  problem  has  justified  itself  in  the  results  attained 
m  investigating  the  larger  problem  of  valuation  in  general 
shall  first  study  the  laws  governing   the  worth   feeling  of 
the  individual  apart  from   those   complications   introduced  by 
participation  in  the  worth  feelings  of  others,  leaving  this  aspect 
I  the  problem  to  be  treated  separately  in  another  place. 

.he  investigation  of  the  laws  of  valuation  has  been  described 
as  a  larger  problem  than  that  contemplated  by  the  economic 
ctnne    of   consumption.     The    meaning    of    this    will    appear 
when  we  reflect  that,  while  valuation  is  feeling-attitude  toward 
:ts  in  general,  physical  and  psychical,  economic  valuation 
oncerned  only  with  those  feeling  attitudes  towards  objects 
nation  of  which  is  conditioned  by  consumption.     Consump- 
pn  is  but  a  special  case  of  appreciation.    Although  other  modes 
>i  appreciation,  such  as  the  ethical  and  aesthetic,  may  enter  in  to 
ify  the  worths  of  consumption,  and  thus  to  change  the  total 
i  or  funded  meaning  of  the  object,  they  are  nevertheless 
tly  speaking,  no  longer  worths  of  consumption.     The  general 
m  appreciation,  includes,  therefore,  besides  the  worth  feelings 
ch  arise  in  the  process  of  consumption  of  physical  goods 
ler  feelings  which  have  as  their  objects  psychical  qualities  of  a 
higher  order,  growing  out  of  the  physical  objects  by  processes  of 
^eptual  and  ideal  construction.     Such  are  the  esthetic  worths 


144  Valuation:   its  Nature  and  Laws 

imputed  to  a  feast  as  the  result  of  arrangement  of  the  objects  of 
consumption  in  an  harmonious  manner.  Such  also  are  the  values 
which  we  describe  as  cleanliness  and  thrift,  objects  of  appre 
ciation  which  emerge  in  processes  of  consumption,  and  which, 
although  qualities  inhering  in  these  processes,  may  nevertheless 
be  abstracted  from  them  and  become  the  objects  of  appreciation. 
The  laws  of  valuation  must  take  into  account  the  laws  govern 
ing  the  change  in  worth  feelings  corresponding  to  these  different 
types  of  objects. 

But  valuation,  and  therefore  its  laws,  must  include  a  wider 
range  of  feelings  than  these.  Out  of  the  elementary  processes 
of  consumption  and  appreciation  are  developed  other  ideal 
objects  and  constructs  (not  objects  of  sensation  and  presentation 
at  all,  but  merely  of  j  udgment  and  assumption) ,  which  may  both 
modify  the  primary  worths  of  appreciation  and  also  come  to  be 
independently  valued.  The  worth  of  an  object  of  consump 
tion,  as  it  is  being  enjoyed,  may  be  modified  by  instrumental 
judgments  as  to  its  utility  for  other  purposes,  or  by  judgments 
regarding  its  exchange  value  as  determined  by  reference  to 
its  demand  by  other  subjects.  Likewise  the  simple  "condition  " 
worths  of  appreciation,  cleanliness  and  thrift  or  other  qualities 
of  this  nature,  may  become  the  objects  of  a  new  type  of  feelings 
when  they  are  acknowledged  as  attributes  of  the  self  ;  they 
become  psychical  objects  of  personal  worth.  Our  feeling  of  the 
worth  of  these  psychical  objects  may  be  further  modified  in  a  sig 
nificant  manner  by  subsidiary  judgments  regarding  the  existence 
of  these  objects  in  social  groups  or  in  society  at  large,  in  some 
cases  heightened  by  judgment  of  their  absence. 

All  these  facts  lead  us  to  recognise  the  extent  of  the  phe 
nomena  which  the  laws  of  valuation  must  include  in  their 
generalisations,  the  variety  of  objects  of  worth  feelings,  and  the 
variety  of  processes  and  attitudes  which  these  feelings  pre 
suppose.  Our  problem,  from  the  standpoint  of  a  general  theory 
of  value,  is  to  examine  the  psychological  laws  of  feeling  and 
its  modifications,  developed  for  the  purposes  of  economics,  and 
to  determine  the  extent  of  their  application  to  other  types  and 
objects  of  valuation. 


The  Laws  of  Valuation  145 


2.  Laws  of  Valuation  as  Laws  of  Affective-Volitional  Meaning — 
Classification  and  Interpretation. 

These  psychological  laws  may  in  general  be  described  as 
laws  of  relativity  of  worth  feeling.  They  are  formulae  describing 
the  modifications  in  worth  feeling  following  upon  modifications 
of  the  dispositional  presuppositions  of  the  feeling.  The  first, 
the  law  of  the  Threshold,  states  the  fact  that  the  power  of  an 
object,  or  rather  of  a  given  quantity  of  an  object,  to  call  out 
worth  feeling  and  judgment,  is  a  function  of  a  disposition  created 
by  previous  feelings.  The  second,  the  law  of  Diminishing 
Value,  is  a  statement  of  the  fact  that  the  change  in  the  capacity 
of  an  object  for  valuation  is  a  function  of  the  effect  of  previous 
worth  feelings.  Quantity  of  actual  feeling,  whether  in  the  form  of 
repeated  reactions  or  of  intensity  of  a  single  reaction,  diminishes 
the  capacity  of  the  disposition  for  further  actualisation  in  the 
form  of  explicit  feeling  and  judgment.  The  third  law,  that 
of  Complementary  Values,  is  a  formula  for  the  description  of 
the  modification  of  the  capacity  of  an  object  for  calling  cut 
worth  feeling,  as  determined  by  the  combination  of  the  primary 
object  with  other  objects.  Under  certain  conditions  the  working 
of  the  law  of  Diminishing  Value  is  thus  modified  and  the  value 
of  the  object  increased. 

The  interrelation  of  these  laws  is  apparent.  The  threshold 
of  worth  judgment  is  determined  by  both  the  second  and  third 
laws.  The  second  law,  working  alone,  has  the  effect  of  raising 
the  threshold,  but  this  effect  is  modified  by  the  factor  intro 
duced  by  the  third  law.  It  is,  accordingly,  the  effect  of  these 
laws  in  determining  the  value,  or  affective- volitional  meaning, 
of  different  types  of  objects  that  constitutes  the  ultimate 
object  of  their  study.  They  are  further  interrelated  in  that 
they  constitute  the  laws,  formulate  the  conditions,  of  psychical 
progression  or  value  movement.  Movements  to  new  objects,  or 
changes  in  attitude  toward  old  objects,  of  desire  and  feeling  are 
to  be  interpreted  in  terms  of  these  laws. 

Finally,  it  is  to  be  observed  that  these  laws  are  descriptive 
formulas  for  the  effect  of  actualisation  of  a  disposition,  in  actual 
feeling,  upon  the  strength  of  the  disposition  presupposed,  and 
therefore  upon  its  capacity  for  further  actualisation.  When 
this  is  recognised,  the  importance  of  our  preceding  analysis 
of  the  presuppositions  of  worth  feeling,  the  types  of  acts  through 


1 46  Valuation  :   its  Nature  and  Laws 

which  feelings  are  actualised,  becomes  apparent.  The  actual 
effect  of  these  different  presuppositions,  presumptions,  judgments, 
assumptions,  upon  the  feeling,  and  its  corresponding  disposition, 
is  a  problem  for  specific  analysis.  It  is  only  in  the  case  of  dis 
positions  corresponding  to  the  feeling-tone  of  sensation  that 
the  relation  can  be  conceived  as  one  of  direct  causal  stimulation. 
In  the  case  of  feelings  of  value  the  relation  is  of  another  type, 
and  the  underlying  dispositions  must  be  clearly  distinguished. 

II.  THE  LAW  OF  THE  THRESHOLD — ITS  GENERAL 

MEANING 

With  these  general  considerations  in  mind,  we  may  turn 
to  a  study  of  the  first  principle  of  relativity  of  worth  feeling, 
that  which  we  have  described  as  the  principle  of  the  Threshold. 

In  general  this  principle  is  an  expression  of  the  fact  that 
the  power  of  an  object  to  call  out  a  feeling  of  worth,  or  a  feeling 
of  worth  difference,  depends  not  upon  the  object  alone,  but 
upon  the  feeling  or  conative  dispositions  of  the  subject  as  well. 
If  this  principle  is  put  into  quantitative  form,  the  question 
relates  to  the  least  quantity  of  the  object  which  will  produce 
a  modification  in  the  feeling  of  worth.  The  importance  of  this 
principle  is  obvious.  All  worth  theory  is  concerned  with  the 
determination  of  the  principles  of  relative  preferability  of  objects, 
that  is  their  relative  importance  or  affective- volitional  meaning. 
In  order  that  such  degrees  of  preferability  may  be  established, 
it  is  necessary  that  fixed  starting-points  for  such  estimation 
should  be  found  at  the  limits  of  relative  worth,  i.e.  where 
relative  worth  passes  over  into  worthlessness  on  the  one 
hand,  or  into  absolute  unlimited  worth  on  the  other.  These 
upper  and  lower  limits  of  relative  worth  we  may,  on  the 
analogy  of  the  laws  of  intensity,  call  the  upper  and  lower 
thresholds. 

This  concept  of  limits  has  been  defined  with  accuracy  for 
the  limited  sphere  of  consumption  and  utilisation.  The  upper 
and  lower  limits  of  valuation  in  this  sphere  have  been  described 
respectively  as  the  "  existence-minimum "  and  the  "  final 
utility." 

But  that  this  concept  of  limits  is  susceptible  of  extension, 
of  a  much  wider  range  of  application  in  worth  analysis  and  theory, 
a  very  superficial  reflection  makes  evident.  The  objects  of 
worth  judgment,  we  have  seen,  may  be  physical  objects  of 


The  Laws  of  Valuation  147 

consumption,  or  qualities  of  these  objects  for  appreciation ; 
they  may  be  psychical  objects,  such  as  acts  of  persons,  or 
affective-volitional  dispositions  expressed  by  those  acts.  If  we 
take  the  second  group  of  objects,  acts  and  dispositions  pre 
supposed  by  these  acts,  and  consider  the  ethical  judgments 
called  out  by  them,  we  find  that  the  same  act,  or  the  amount 
of  disposition  disclosed  by  that  act,  may  have  worth  or  be 
worthless  according  to  the  disposition  presupposed  in  the  sub 
ject  of  the  judgment.  Thus,  as  an  object  of  personal  worth  it 
may  rise  above  the  threshold  of  value,  when  it  has  not  reached 
the  minimum  required  to  call  out  the  social  moral  judgment. 
So,  also,  a  given  act  may  call  out  a  judgment  of  disapproval, 
personal  or  social,  when  it  is  not  sufficient  to  rise  above  the 
threshold  of  legal  judgment.  There  are  thus,  so  to  speak, 
qualitatively  different  thresholds  marking  off  the  different  spheres 
of  meaning  in  which  the  judgments  are  made.  To  these  cor 
respond  different  modifications  of  feeling,  the  actualisation  of 
which  requires  different  types  of  objects  or  different  amounts 
of  the  same  object.  This  fact  finds  illustration  even  in  the 
case  of  objects  of  strictly  economic  valuation.  A  physical 
object  of  mere  condition  worth,  the  value  of  which  consists 
solely  in  its  capacity  of  satisfying  some  desire  of  the  senses,  may, 
with  increase  or  decrease  in  amount,  call  out  feelings  and  judg 
ments  of  possession  or  of  instrumental  value  which  modify  the 
worth  feeling,  not  only  in  the  direction  of  degree,  but  also  by 
introducing  new  aspects  of  quality. 

From  these  facts  several  important  consequences  follow. 
In  the  first  place,  by  discovering  the  upper  and  lower  limits  of  a 
given  type  of  valuation — the  boundaries  within  which  differ 
ences  of  worth  are  determined  wholly  by  differences  in  quantity 
of  the  object,  we  are  enabled  to  mark  off  the  different  levels  of 
valuation.  The  qualitative  thresholds  mark  the  limits  of  these 
regions.  Thus  we  shall  distinguish  thresholds  of  condition 
worth  of  appreciation,  thresholds  of  personal  worth  of  charac 
terisation,  and  thresholds  of  social  worth  of  participation  and 
utilisation.  In  the  second  place,  in  determining  these  limits  and 
their  mutual  relations,  we  have  a  basis  for  the  analysis,  not  only 
of  the  laws  of  preference  within  a  given  sphere  of  meaning, 
but  also  of  the  laws  of  value-movement  from  one  sphere  of 
meaning  to  another,  and  therefore  the  laws  of  preference  of 
one  type  of  objects  over  another. 

The  different  levels  of  valuation  and  the  different  objects 


148  Valuation :   its  Nature  and  Laws 

of  value  are  the  products  of  perceptual  and  ideal  constructions 
in  the  process  of  acquirement  of  meaning.  While  the  thresholds 
represent  conceptual  limits  within  which  the  valuation  of  a 
given  type  of  objects  moves,  they  nevertheless  correspond  to 
actual  appreciatively  recognisable  differences  in  feeling-atti 
tude,  determined  by  difference  in  presuppositions.  A  genetic 
study  of  these  presuppositions,  and  therefore  of  the  expecta 
tions  or  demands  to  which  they  give  rise,  will  enable  us  to  fix 
the  limits  of  these  demands. 

With  this  understanding  of  the  larger  significance  of  the 
concept  of  the  threshold  in  worth  theory,  we  may  now  return 
to  a  detailed  study  of  the  psychology  of  the  economic  threshold 
of  condition  worths. 


2.  Economic  Thresholds — The  Existence-Minimum 
and  Final  Utility. 

The  limits  within  which  the  relative  valuation  of  an  object 
of  condition  worth  moves  may  be  described  as  the  "  existence- 
minimum,"  and  the  point  of  "  final  utility." 

The  minimum  of  existence  is  that  conceptual  point  at  which 
absolute  value  passes  over  into  relative  value.  Until  the  mini 
mum  of  a  given  good  which  is  necessary  to  existence  is  reached, 
there  can  be  no  estimation  of  relative  value,  for  any  quantum 
of  the  good  has  absolute  value  and  is  capable  of  calling  out 
indefinite  sacrifice.  The  good  has  no  substitute,  and  it  is  only 
among  goods  with  "  capacity  of  substitution  "  that  relative  worth 
or  degrees  of  preferability  can  be  established.  The  upper  threshold 
of  relative  value  in  consumption  is,  then,  the  first  increment  of 
satisfaction  which  rises  above  the  indispensable  minimum.  The 
psychological  meaning  of  this  concept  of  the  upper  threshold 
is  to  be  found  in  the  fact  that  until  this  point  is  reached  the 
entire  personality  is  involved,  that  is,  about  the  fundamental 
want  or  desire,  whether  it  be  for  food  or  warmth,  or  any  other 
object,  the  entire  system  of  conative  tendencies  is  concentrated, 
through  arrest  of  the  fundamental  desire,  and  is  directed  toward 
the  one  object.  When  the  minimum  of  existence  is  reached,  and 
the  tension  or  arrest  of  the  fundamental  tendency  is  relaxed,  the 
estimation  of  worth  becomes  relative,  that  is,  each  successive 
increment  of  the  good  is  estimated  with  reference  to  the  im 
portance  of  the  isolated  particular  wants  and  desires.  It  need 
scarcely  be  said  that  for  many  goods  there  is  no  such  upper 


The  Laws  of  Valuation  149 

threshold  or  minimum  of  existence.  They  find  substitutes  or 
otherwise  cease  to  be  desired  long  before  this  point  is  reached, 
but  their  place  or  relative  importance  in  a  system  of  values  is 
determined  by  their  relations  to  certain  fundamental  goods 
without  capacity  of  substitution,  relations  which  become 
apparent  as  soon  as  our  habitual  valuations  are  disturbed  and 
we  are  forced  to  ultimate  preferences. 

The  lower  threshold,  or  the  point  of  final  utility,  is  again 
that  conceptual  point  at  which  the  minimum  of  worth  tends  to 
pass  over  into  worthlessness.  It  is  the  smallest  quantity  of  the 
good  which  has  the  capacity  of  satisfying  the  least  important 
of  the  subsidiary  wants  or  purposes  of  the  subject.  Psycho 
logically  expressed,  it  corresponds  to  the  most  remote  of  the 
subsidiary  conative  tendencies  connected  with  the  fundamental 
to  which  the  good  ministers.  Now,  whether  any  such  point 
of  final  utility  actually  exists  in  any  concrete  experience  or  not, 
it  is,  like  the  minimum  of  existence,  at  least  an  ideal  limit  useful 
in  describing  actual  processes  of  valuation.1 

In  the  case  before  us — which  is  concerned  with  the  instru 
mental  values  of  economics,  and  therefore  with  worth  feelings 
which  have  as  their  presuppositions,  in  addition  to  judgments 
of  existence  and  non-existence,  subsidiary  utility  judgments, 
it  is  clear  that  these  points  represent  limits  of  relative  instru 
mental  valuation.  Thus  the  lower  threshold,  the  quantum  of 
minimal  worth,  represents  that  point  at  which  the  judgment  of 
its  existence  or  non-existence  has  the  least  importance,  that  is 
where,  with  this  judgment,  goes  an  instrumental  judgment 
referring  it  to  the  least  significant  purpose.  Any  quantity 
below  that  is  worthless,  or  is  valued,  if  at  all,  in  some  other 
attitude,  i.e.,  intrinsically.  Thus  to  the  rich  man  the  small  coin, 
a  penny,  may  be  instrumentally  worthless,  while  as  a  part  of  an 
individualised  whole,  his  wealth,  it  may  have  an  intrinsic  worth. 

1  The  use  of  the  term  Final  Utility  in  this  way  requires  some  explanation  and  perhaps 
apology.  As  introduced  by  Jevons,  and  employed  by  economists  generally,  it  describes 
the  intensity  of  desire  satisfied  by  the  last  increment  of  a  commodity  purchased  or  con 
sumed,  i.e.,  last  in  any  given  process  of  acquisition  or  consumption,  and  may  fall  far 
short  of  the  hypothetical  last  increment  before  worthlessness  is  reached.  It  is  therefore 
only  relatively  final  and  is  identical  with  the  concept  of  Marginal  Increment  or  Marginal 
Utility  which  is  now  gradually  displacing  it.  Since,  however,  Final  Utility  in  this 
sense  is  held  to  be  determined  by  the  law  of  Diminishing  Utility,  according  to  which  the 
value  of  any  stimulus  is  the  degree  of  satisfaction  obtained  by  its  last  repetition,  and 
therefore  varies  with  the  amount  or  rate  with  which  the  stimulus  is  furnished,  final 
utility  as  marginal  tends  to  become  identical  with  final  in  the  sense  of  minimal  value. 
The  very  fact  that  the  somewhat  inept  use  of  the  term  as  identical  with  Marginal  Utility 
is  passing  into  disuse,  may  perhaps  serve  as  an  excuse  for  the  special  application  made 
of  it  in  this  connection. 


150  Valuation :   its  Nature  and  Laws 

The  point  of  minimal  or  final  utility  is,  then,  the  psychological 
point  where  all  relative  instrumental  judgments  fall  away  or  at 
least  are  not  called  out.  The  minimum  of  existence,  on  the  other 
hand,  the  point  where  relative  valuation  passes  over  into  absolute, 
is  psychologically  the  point  of  arrest  or  suppression  of  all  instru 
mental  j  udgments,  where  value  becomes  intrinsic.  We  have  at  this 
point  the  extreme  of  arrest,  and,  as  a  consequence,  concentration 
of  all  subsidiary  tendencies  or  dispositions  about  the  funda 
mental.  The  feeling,  in  this  case  passion,  has  all  the  acquired 
meaning  of  the  other  dispositions,  but  it  is  implicit  and  immediate, 
all  relative  judgments  being  suppressed.1 

There  are  two  points,  therefore,  at  which  relative  and  instru 
mental  value  passes  over  into  intrinsic :  (a)  the  upper  threshold 
where  the  value  becomes  practically  absolute ;  and  (6)  the  lower 
threshold,  where  minimal  worth  passes  into  worthlessness 
unless  the  minimum  in  some  way  acquires  intrinsic  worth. 
The  first  of  these  situations  we  may  easily  understand.  The 
transgredient  reference  of  the  feeling  to  presupposed  conative 
dispositions  has  become  an  impellent  mode  of  quasi-obligation. 
Desire  and  feeling  are  unconditional,  undetermined  by  any  specific 
particular  end.  Here,  we  shall  find  later,  arise  certain  quasi- 
obligatory  feelings  even  in  the  sphere  of  economic  values.  The 
other  situation — where  the  object  which  is  worthless  or  below  the 
lower  threshold  of  instrumental  value,  acquires  intrinsic  worth, 
requires  more  detailed  analysis. 

1  This  concept  of  upper  and  lower  limits  of  relative  valuation  has  been  developed 
for  the  purpose  of  making  possible  descriptive  formulce  for  concrete  worth  attitudes. 
To  what  extent  do  these  conceptual  points  correspond  to  concrete  situations  ? 
What  are  these  situations  in  terms  of  our  analysis?  Have  they  actual  correlates 
or  are  they  purely  conceptual?  It  is  obvious  that,  although  possible,  concrete 
psychological  dispositions  represented  by  the  conceptual  points  are  rarely  actual 
moments  in  experience.  We  approach  them,  but  they  always  tend  to  remain  ideal 
limits.  In  what  sense  then  do  they  enter  as  actual  determinants  into  experi 
ence?  Do  they  at  all?  They  do,  it  would  appear,  and  in  this  way.  Although  the 
actual  feeling  correlated  with  these  thresholds  is  scarcely  realisable,  for  the  limiting 
judgments  which  constitute  their  presuppositions  are  largely  hypothetical,  nevertheless 
such  judgments  are  frequently  represented  by  assumptions,  and  the  feelings  following 
upon  these  assumptions  complete  the  worth  series.  Of  this  concept  of  the  substitution 
of  assumption  feelings,  as  affective  signs  in  the  continuity  of  real  valuation,  we  shall 
make  important  applications  later.  Here  it  is  necessary  merely  to  note  the  phenomenon, 
and  to  observe  that  while  these  substituted  feelings  lack  the  intensities  which  they  would 
have  in  real  situations  where  the  presuppositions  are  existential  judgments,  they  have 
nevertheless  the  worth  suggestion,  in  this  case  the  transgredient  reference,  which  gives 
them  a  functional  place  in  actual  valuation. 


The  Laws  of  Valuation  151 


3.  Modification  of  the  Lower  Threshold  through  Acquired  Meanings 
— Complementary  Values. 

The  acquirement  of  intrinsic  worth  on  the  part  of  an  object 
or  quantity  of  an  object,  in  itself  below  the  threshold  of  value, 
must  be  viewed  as  a  modification  of  feeling  brought  about  by 
inclusion  of  a  new  presupposition.     It  is  already  apparent  that 
an  object,  the  mere  existence  of  which  does  not  call  out  worth 
feeling  may  rise  above  the  threshold  when  it  is  related  through 
instrumental  judgment  to  other  worth  objects  and  their  feelings. 
It  is  also  true — and  this  is  the  phenomenon  with  which  we  are 
here  concerned,  that  a  worthless  object  may  acquire  intrinsic 
value  through  relation  to  an  individuated  whole  which  is  as 
sumed  to  exist,  and  which  has  intrinsic  value.     Such  acquired 
values  may  be  described  at  complementary,  and  to  the  more 
detailed  study  of  such  values  we  shall  devote  a  later  section. 
Here  we  may  content  ourselves  with  a  study  of  the  manner  in 
which  they  modify  the  threshold.     In  the  case  cited,  the  simple 
intrinsic  appreciation  of  the  penny,  when  it  has  no  instrumental 
worth  for  the  subject,  is  a  worth  feeling  with  presuppositions 
different  in  certain  important  points  from  the  threshold  of  in 
strumental  worth,  namely,  in   the   subsidiary   presuppositions. 
In  the  place  of  the  instrumental  judgments  and  their  feelings, 
which  the  object  in  itself  does  not  call  out,  there  enters  as  a 
substitute  an  assumption-feeling,  having  as  its  presupposition 
an  envisaging,  and  momentary  assumption  of  existence,  of  wealth 
as  an  individualised  whole,  of  which  the  minute  object,  though 
instrumentally  worthless,    is    a    part.      That  which  raises  the 
minute  object  above  the  threshold  of  value  in  such  a  case  is  not 
the  intensity  of  the  immediate  feeling,  but  the  subsumption  of 
the  judgment-feeling  under  the  assumption-feeling  described. 

A  like  phenomenon  is  observable  in  other  regions  of  valua 
tion  than  the  economic.  Thus  in  the  ethical  sphere  we  frequently 
find  that  the  theft  of  a  pin,  or  some  other  minute  commission 
or  omission,  although  utterly  insignificant  instrumentally,  may 
call  out  feelings  of  intrinsic  worth  or  ww-worth.  Here,  again, 
it  is  not  the  mere  existence  or  non-existence  of  the  act  which 
calls  out  the  worth  feeling,  but  the  subsumption  of  the  act 
under  the  general  concept  imputes  to  the  act  all  the  worth 
suggestion  of  the  affective  abstract  which  the  assumption  of  the 
existence  of  the  object,  corresponding  to  the  concept,  involves. 


152  Valuation:   its  Nature  and  Laws 

Similarly,  in  the  sphere  of  the  aesthetic,  we  shall  find  that 
an  element,  worthless  in  itself,  may,  as  part  of  an  intuitively 
individuated  whole,  acquire  the  value  of  the  whole.  The 
feeling  may  be  said  to  expand  to  include  the  element  in 
itself  worthless.  With  the  characteristics  of  intuitive  individu- 
ation  and  with  the  nature  of  complementary  aesthetic  values 
we  are  not  here  concerned.  It  will  be  sufficient  to  note  that 
even  in  the  region  of  economic  objects  of  condition  vorths 
this  law  is  at  work.  In  certain  quasi-aesthetic  combir.ations 
of  utilities — as  in  a  festal  meal  or  the  style  of  a  suit  of  raiment 
—a  detail,  valueless  in  itself,  as  a  single  utility,  may  acquire 
an  extraordinary  value. 

From  the  illustrations  before  us  the  general  principle  is 
apparent,  that  an  object  below  the  threshold  of  value  may  become 
the  object  of  intrinsic  valuation  when  there  is  imputed  to  it 
the  worth  which  comes  with  the  assumption  of  the  existence 
or  non-existence  of  an  individuated  whole  of  which  it  is  a  part. 
The  ordinary  explanation  of  this  fact  is  in  terms  of  association, 
the  rise  above  the  threshold  being  conceived  as  due  to  a  sum 
mation  of  minute  dispositional  feelings  below  the  threshold, 
aroused  through  the  stimulation  of  dispositions  associated  with 
the  affective  disposition  corresponding  to  the  primary  object. 
This  concept  of  mere  summation,  which  is  inevitable,  when 
worth  feeling  is  defined  as  pleasure,  is,  however,  untenable.1 
The  modification  of  worth  feeling  which  follows  upon  the 
intrinsic  ethical  or  aesthetic  valuation  of  the  object,  is  not 
a  summation  of  hypothetical  feeling  elements,  but  a  new 
total  feeling  following  upon  an  apprehension  of  a  new-founded 
object  either  conceptual,  or,  as  in  the  case  of  the  aesthetic,  in 
tuitively  given,  which  is  the  object  of  assumption.  We  are 
concerned  here  with  a  phenomenon  of  subsumption  of  feeling, 
not  summation,  with  a  change  in  cognitive  presuppositions, 
not  change  in  stimulation.  The  worth  threshold,  in  contrast 
to  the  simple  hedonic,  must  be  viewed  as  cognitive  in  character. 


4.  The  Independent  Variability  of  Hedonic  and  Worth 
Thresholds. 

The  threshold  of  value,  whether  of  simple  appreciation  with 
its  intrinsic  reference  or  of  utility  with  its  instrumental  refer- 

1  Hoffler,  Psychologic,  Yienna,  1897,  p.  448. 


The  Laws  of  Valuation  153 

ence,  has,  it  will  be  seen,  a  cognitive  character,  which  dis 
tinguishes  it  from  the  merely  hedonic  threshold.  The  con 
fusion  of  these  two  conceptions,  a  procedure  fairly  common 
in  worth  analysis  and  theory,  has  led  to  serious  fallacies  which 
can  be  avoided  only  by  insistence  upon  this  point.  The  inde 
pendent  variability  of  worth  feeling  and  of  pleasantness  and 
unpleasantness,  already  pointed  out  in  a  preceding  chapter,1 
is  further  confirmed  by  our  study  of  the  worth  threshold.  For 
it  is  possible  that,  as  Kriiger  pointed  out,  an  object  may  call 
out  a  fleeting  pleasantness  or  unpleasantness  or  momentary 
impulse  without  crossing  the  threshold,  and  it  is,  on  the  other 
hand,  equally  possible  for  an  object  to  rise  above  the  worth 
threshold  without  an  appreciable  affective  disturbance. 

As  an  illustration  of  the  first  possibility  we  may  consider 
the  case,  an  entirely  possible  one  certainly,  where  a  trivial 
musical  phrase  in  its  aspect  of  mere  auditory  stimulation  (or, 
still  better,  a  momentary  organic  sensation)  may  call  out  a  fleeting 
pleasantness  or  unpleasantness  which  does  not  reach  the  point  of 
worth  judgment.  We  are  conscious  of  the  hedonic  change,  but 
there  has  not  been  that  totalisation  of  affective  attitude  through 
cognitive  reference  to  the  object  which  constitutes  the  pre 
supposition  of  worth  feeling.  The  second  possibility,  the  rise 
of  an  object  above  the  worth  threshold  without  accompanying 
pleasure  causation,  is  illustrated  by  the  many  cases  where  a 
disposition  to  worth  judgment  has  been  formed  on  the  basis  of 
immediate  pleasure-causation,  and  where  the  judgment  of  the 
existence  of  the  object  then  calls  out  worth  feeling  even  when 
the  object  fails,  although  sensed  or  presented,  to  give  pleasure. 
Thus  when  the  point  of  satiety  has  been  reached,  it  frequently 
happens  that  actual  enjoyment  of  the  object  (of  taste,  for  in 
stance),  or  imaginative  presentation  of  its  enjoyment,  is  no 
longer  accompanied  by  pleasure,  but  the  judgment  that  the 
object  still  exists,  or  that  it  does  not  exist,  may  call  out  feelings 
of  value. 

This  principle  of  the  independent  variability  of  the  simple 
hedonic  and  worth  thresholds  is  further  substantiated  when  we 
turn  from  the  stimulus-threshold  to  the  threshold  of  difference. 
There  may  be  appreciable  differences  in  the  hedonic  intensity 
without  changes  in  worth  feeling,  and  changes  in  the  worth 
attitude  without  appreciable  changes  in  hedonic  intensity. 
For  the  first  of  these  possibilities  a  simple  illustration  will  suffice. 

1  Chapter  in, 


Valuation:   its  Nature  and  Laws 

In  the  apprehension  of  an  object,  more  especially  a  work  of  art, 
we  may  distinguish  between  the  emotional  attitude  toward  the 
object  as  a  whole,  and  the  particular  affective  tones  of  the  ele 
ments    as  attention  passes  from  one  phase  to  another, 
there  'may  be  appreciable   changes  in  feeling-tone  which  are 
sufficient   to  lead   to   movement   from   one   aspect   to   another 
in  readaptation  of  attention,  but  not  sufficient  to  lead  to  that 
readaptation   of   judgmental    attitude  which  is  presupposed  in 
value  movement-or  in  change  of  worth  attitude  toward  the 
object  as  a  whole.     Similarly,  repeated  acts  of  apprehension  may 
lead  to  appreciable  changes  in  feeling-tone,  due  to  fatigue  of  the 
nervous  dispositions  connected  with  sense  stimulation,  without 
necessarily  leading  to  a  change  in  worth  attitude,  that  is  t 
change  in  cognitive  presuppositions  presupposed  in  such  change 
of  attitude.     The  worth  feeling  toward  the  object,  founded 
these   elements   of    presentation,   presupposes   another   disposi 
tion      These   phenomena,   we   shall    find,   are    of   considerable 
importance    when   we    come   to   consider   the   second   law 

valuation. 

A  similar  situation  exists  in  the  case  of  economic  feelings  oi 
value   that  is  where  the  feeling  follows  upon  instrumental  judg 
ments.     Let  us  take  the  concrete  case  of  an  individual  in  the 
enjoyment    of    any    object    which     causes     pleasure.     With 
the    total   attitude,   which   may   vary    in    time    length, 
may  be  appreciable   differences  in  pleasure,   as,   for  instance, 
in    the    consumption    of    a  food,   without    cognisable    change 
in  valuation  until  there  has  entered  a  change  of  attitude   that 
a  change  in  the  judgment  presuppositions  of  the  feeling,  through 
the  judgmental  reference  of  the  object  to  other  purposes,  1 
purposes  of  the  future,  etc.     Here  again  cognisable  differences 
in  worth  feeling  do  not  coincide  with  the  least  perceptil 

donic  changes. 

The  second  possibility— variation  in  the  worth  attitude  with 
out  variation  in  intensity  of  pleasure-may  also  be  illustrate 
from  the  sphere  of  feelings  of  economic  value.     From  the  poii 
of  view  of  pleasure  it  is  a  matter  of  indifference  whether,  beside 
the  quantity  of  the  object  being  enjoyed,  other  objects  c 
same  nature  exist  or  do  not  exist,  but  for  the  worth  feeling  it 
is  of  considerable  moment.     The  inclusion  within  the  presuppo 
sitions  of  the  feeling  attitude  of  such  accompanying  judgments 
may  bring  about  a  cognisable  difference  in  the  feeling  of  value, 
while  the  feelings  of  simple  pleasure    remain  the  same. 


The  Laws  of  Valuation  155 

In  these  phenomena  of  difference  in  variability  of  the  threshold 
of  difference  for  pleasure-causation  and  the  threshold  of  differ 
ence  for  feelings  of  value,  we  have  a  situation  analogous  to 
that  which  has  been  pointed  out  in  the  sphere  of  sensation  and 
perception.  It  has  been  argued  that  there  also  a  distinction 
must  be  made  between  the  threshold  of  difference  for  sensation 
and  for  perception.  Within  limits  a  stimulus  may  be  varied 
without  producing  an  appreciable  difference  in  the  object  cog 
nised.  There  is  a  difference  for  sensation,  but  not  for  per 
ception.  This  means  that  the  perception  of  difference  involves 
the  acquirement  upon  the  part  of  the  sensations  of  a  meaning 
which  expresses  itself  in  readaptation,  in  modification  of  attitude 
in  the  form  of  judgment.  Between  these  changes  in  meaning, 
or  attitude,  are  changes  for  sensation  which  are  not  significant 
for  cognition.1  In  a  somewhat  similar  way  it  may  be  said  that 
worth  feeling  and  feeling  of  worth  difference  belong  to  the  cogni 
tive  level,  i.e.,  are  emotion,  sentiment,  and  mood,  and  not  the 
hypothetical  element  of  pleasure.  There  may,  therefore,  con 
ceivably  be  change  in  the  hedonic  redundancies  without  any 
change  in  the  value  reaction.  This  means,  simply,  that  varia 
tions  may  be  superficial  in  the  sense  that  they  do  not  involve 
any  totalisation  of  consciousness  in  a  unitary  affective  attitude. 
They  do  not  penetrate  into  or  involve  the  personality. 

The  formulation  of  this  distinction — between  the  hedonic 
and  worth  thresholds,  involves  important  consequences  for 
our  theory.  For,  the  fact  having  been  established  that  the 
rise  of  an  object  above  the  threshold  of  worth  or  worth  differ 
ence  involves  cognition  of  affective-volitional  meaning,  trans- 
gredient  and  immanental  reference,  and  therefore  judgment  or 
assumption  as  presuppositions,  it  will  now  be  necessary,  in  ex 
amining  the  psychological  laws  operating  to  modify  these  thresh 
olds,  to  study  each  type  of  worth  feeling  for  itself,  and  to 
refuse  to  extend  uncritically  the  laws  modifying  pleasure- 
causation,  and  therefore  the  hedonic  threshold,  to  worth  feeling 
and  the  perception  of  worth  differences  in  general. 

1  Stumpf,  Tonpsychologie,  Vol.  I,  p.  33.     Also  Stout,  Analytical  Psychology,  Vol.  I, 
p.  48,  and  Manual  of  Psychology,  p.  120. 


156  Valuation:    its  Nature  and  Laws 


III.  THE  LAW  OF  DIMINISHING  (OR  LIMITING)  VALUE  :  CRITICAL 
STUDY  OF  ITS   PSYCHOLOGICAL  BASIS — INTERPRETATION 

i.   Its  Historical  Significance. 

The  preceding  study  of  the  concept  of  the  threshold  of  value, 
and  its  relation  to  the  hedonic  threshold,  has  made  it  clear 
that  for  an  object  to  call  out  a  feeling,  or  a  feeling  of  difference, 
of  worth  or  value,  a  change  in  its  cognitive  presuppositions 
is  necessary.  Whether  in  a  given  psychological  situation 
a  given  object  or  quantity  of  an  object  will  rise  above  the  thresh 
old  of  worth  feeling,  or  feeling  of  worth  difference,  depends 
upon  the  attitude  or  disposition  (conative  and  judgmental) 
created  by  previous  valuations.  The  problem  immediately 
arises  as  to  the  existence  of  definite  laws  governing  this  rela 
tivity,  whether  any  factors  can  be  discovered  which  modify 
in  a  uniform  way  the  dispositional  presuppositions  of  worth 
feeling,  and  thus  the  worth  feelings  themselves. 

The  theory  of  value  developed  for  the  purposes  of  economics 
has  formulated  a  law  of  relativity  for  utility  in  consumption, 
basing  its  formulation  upon  the  psychological  laws  governing 
the  modification  of  "  sensation-feelings  "  and  their  corresponding 
dispositions.  The  factors  of  frequency  of  stimulation,  quantity 
of  stimulation,  and  limitation  of  capacity  for  stimulation,  in  pro 
ducing  the  effects  of  habit,  or  dulling  of  sensitivity  and  of  satiety, 
afford  the  psychological  basis  for  the  general  law  of  Diminishing 
Value  and  for  the  more  specific  law  of  Marginal  Utility  connected 
with  it.  On  the  assumption  that  all  feeling  of  value  is  identical 
with  pleasure,  it  would  follow  that  this  law  of  relativity  would 
apply  to  all  types  of  worth  feeling. 

But  on  this  point  there  is  difference  of  opinion.  Thus,  on 
the  one  hand,  we  find  Ehrenfels  contending  for  an  extension  of 
the  principle  of  Grenz-nutzen  to  a  more  general  principle  of 
Grenz-frommen,  while  Kreibig  maintains  that  this  principle  is 
a  special  formulation  of  the  general  law  of  relativity,  applicable 
only  to  the  limited  sphere  of  the  instrumental  values  of  economic 
goods.  It  is  clear  that  the  answer  to  this  question  depends 
upon  the  more  fundamental  problem — whether  these  psycho 
logical  factors  which  affect  the  dispositions  underlying  feel 
ings  of  pleasure  act  in  precisely  the  same  manner  upon  the 
conative  and  judgmental  presuppositions  of  worth  feelings, 


The  Laws  of  Valuation  157 

whether  habit  and  satiety  which  correspond  to  the  factors  of 
repetition  and  quantity  have  the  same  functional  meaning  for 
worth  feelings  as  for  pleasure-causation.  These  questions  can 
be  answered  only  in  the  light  of  a  psychological  analysis  of 
these  factors. 

As  a  preliminary  to  this  analysis  we  must  examine  more 
fully  the  formulation  of  this  principle  of  relativity,  and  bring 
to   light   the   psychological   assumptions   upon   which   the   law 
rests.     The  first  formulations  go  back  to  Bentham  and  Ber 
noulli.1     The   problem   of   Bentham  was  the  relation  of  wealth 
to  happiness,  and  it  arose  in  connection  with  the  question  of 
the  relation  of  the  distribution  of  goods  to  his  "  greatest-happi 
ness  "  principle.     As  a  result  of  his  analysis,  he  develops,  in 
the  Panomial  Fragments,  the  following  principle  which  he  dignifies 
with  the  name  of  an  axiom :  "  The  effect  of  wealth  in  the  produc 
tion  of  happiness  goes  on  diminishing  as  the  quantity  by  which  the 
wealth  of  one  man  exceeds  that  of  another  goes  on  increasing  ; 
in  other  words,  the  quantity  of  happiness  produced  by  a  particle 
of  wealth  (each  particle  being  of  the  same  magnitude)  will  be 
less  and  less  at  each  particle  ;   the  second  will  produce  less  than 
the  first,  the  third  less  than  the  second,  and  so  on."     The  com 
mon  element  in  his  various  formulations  of  the  principle  may 
be  expressed  in  the  following  way.     The  minimum  of  existence 
being  assumed  as  given,  when  this  quantity  of  the  good  increases 
in  any  constant  relation,  happiness,  the  primary  worth,  increases, 
other  things  being  equal,  within  certain  limits,  but  not  in  the 
same  proportion  ;    rather  is  the  rate  of  increase  of  happiness  in 
relation  to  wealth  a  constantly  lessening  one,  although  Bentham 
does  not  formulate  any  law  governing   the  rate   of  decrease. 
Bernoulli's  famous  law  also  assumes  the  minimum  of  existence, 
and  he  finds,  as  does  Bentham,  the  increase  of  happiness,  in 
proportion   to   the   increase   of  wealth,   a  constantly   lessening 
quantity.     Bernoulli  goes  beyond  Bentham,  however,  in  attempt 
ing  to  formulate  the  law  quantitatively.     Happiness  is  conceived 
to  grow  in  arithmetical  proportion  as  the  quantity  of  wealth 
increases   geometrically.     Finally    Gossen,    under   whose   name 
the  law  of  Diminishing  Value  has  largely  entered  into  Political 
Economy,   formulates    the    law   in   the   following   words,    thus 
leading  up  to  the  concept  of  marginal  or  final  utility  :    "  With 

1  For  a  general  sketch  of  the  history  of  these  formulations  see  Kraus,  Zur  Theorie  des 
Wertes  .Halle,  1901,  chap,  iv  ("Grundlagen  der  modernen  Werttheorie").  especially 
pp.  58-69. 


158  Valuation:   its  Nature  and  Laws 

the  increase  of  the  quantity  of  the  object  the  worth  of  each 
additional  increment  must  suffer  deciease  until  finally  zero  is 
reached  "  ;  or,  again,  "  The  first  increment  of  an  object  of  value 
has  the  highest  worth,  the  second  less  worth,  until  finally  worth- 
lessness  is  reacheckfcdt 

These  formulations  of  the  law  of  Diminishing  Value  all 
agree  in  the  genera,  conception  that  the  satisfaction  of  desire 
or  pleasure,  when  it  is  increased  quantitatively,  finally  results 
in  the  loss  of  capacity  for  that  desire  (and  with  it  feeling) — 
and,  therefore,  in  movement  of  desire  to  new  objects,  and  the 
formation  of  new  dispositions.  For  this  reason,  in  view  of  the 
functional  aspect  of  the  law,  as  a  cause  of  new  adaptations,  we 
shall  describe  it  henceforth  as  the  law  of  Limiting  Value. 

2.  Psychological  Basis  of  the  Law — The  more  General  Laws  of 
(a)  Dulling  of  Sensitivity  with  Repetition,  and  (b)  Satiety — 
Critical  Study. 

When  we  seek  for  the  psychological  basis  of  the  law,  we  find 
that  all  these  formulations  of  Bentham,  Bernoulli,  and  Gossen 
alike,  assume  the  identity  of  worth  feeling  with  pleasure, 
and  therefore  the  dependent  variability  of  the  two.  All  that 
was  necessary,  therefore,  to  establish  psychologically  this 
general  principle  of  relativity  was  an  appeal  to  the  laws 
governing  the  physiological  conditions  of  sensitivity  to  pleasant 
ness  and  unpleasantness.  Bentham  definitely  based  his  formu 
lation  of  the  principle  upon  the  more  fundamental  psycho- 
physical  laws  of  dulling  of  sensitivity  through  repetition  and 
habit,  satiety,  and  limitation  of  capacity  of  appropriation ;  and 
Fechner  likewise,  in  subsuming  the  law  of  Bernoulli  under  his 
more  general  law  of  sensitivity,  does  so  on  the  assumption  of 
the  universal  applicability  of  these  laws. 

But  this  assumption  immediately  arouses  suspicion  of  its 
soundness,  and  cannot  be  admitted  without  careful  scrutiny. 
If,  as  our  studies  of  the  threshold  indicated,  pleasure-causation 
and  worth  feeling  may  vary  independently,  i.e.,  the  intension 
and  extension  (depth  and  breadth  in  the  personality)  may  vary 
independently  of  the  intensity  and  multiplicity  of  the  hedonic 
redundancies,  and  if  there  may  be,  to  that  extent,  intensity-less 
acts  of  appreciation,  the  presuppositions  of  which  are  conative 

1  Gossen,  Ent-wickelung  der   Gesetze  des   Menschlichen   Verkehrs   und  der  daraus 
fliessenden  Regclnfiir  Mcnschlichcs  Handeln,  1853,  P-  31- 


The  Laws  of  Valuation  159 

and  judgmental,  and  not  sense  stimuli,  it  would  follow  that 
these  psycho-physical  laws  governing  sensation-feelings,  and 
the  general  principle  of  relativity  built  upon  them,  should  not 
be  uncritically  extended  to  judgment  and  -ssumption  feelings — 
that  is  to  worth  feelings,  without  a  sr  ill  analysis  in  each 
case.  In  fact,  Fechner  himself,  even  or.,  the  assumption  that 
pleasure  and  feeling  of  value  are  identical,  \  arns  us  against  any 
overhasty  generalisation  of  the  principles  governing  the  pleasure 
values  of  simple  sensation  to  cover  more  complex  feelings.  In  the 
Vorschide  der  Aesthetik  (p.  76)  he  says  :  "A  really  mathematical, 
merely  psycho-physical  measure  of  the  intensity  of  pleasantness 
and  unpleasantness  could  be  found  only  in  connection  with  a 
knowledge  of  the  universal  ultimate  cause  of  pleasantness  and 
unpleasantness  ;  until  then  it  can  be  nothing  more  than  an 
estimate  of  more  or  less." 

Much  more,  then,  if  the  relative  independence  of  pleasure- 
causation  and  worth  feeling  be  established,  is  a  separate  ex 
amination  of  the  application  of  these  laws  to  the  latter 
class  of  feelings  necessitated.  As  Brentano  truly  says : 
"  For  all  psychical  phenomena  which  have  their  grounds  in 
psychical  occurrences  within  the  organism  (not  in  external 
stimuli),  or  are  called  forth  by  other  psychical  phenomena, 
a  measure  of  intensity  fails."  It  is  entirely  conceivable,  for 
instance,  that  in  total  attitudes  of  appreciation,  or  valuation, 
two  components  of  the  total  feeling  may  be  distinguished,  the 
worth  feeling  which  has  psychical  presuppositions,  conative 
tendency  and  judgment,  and  the  hedonic  redundancies  with 
their  sensational  conditions  both  organic  and  peripheral.  To 
the  latter  aspect  the  laws  of  dulling  of  sensitivity  and  satiety 
might  apply,  while  for  the  former,  already  found  to  be  inde 
pendently  variable,  the  factors  of  repetition  and  quantity  might 
have  another  significance.  It  is  at  least  conceivable  that  in 
crease  in  depth  and  breadth  of  feeling  tone  might  go  on  side  by 
side  with  the  reduction  of  the  intensity  and  multiplicity  of  the 
hedonic  redundancies. 

(a)  The  Law  of  Dulling  of  Sensitivity.  Critical  Analysis — Its 
Application  Limited  to  Sensation  Feelings  and  the  Re" 
dundancies  of  Feelings  of  Value. 

The  two  principles  of  dulling  of  sensitivity  and  satiety 
are  often  confused,  and  treated  as  identical  because  of  their 


1 60  Valuation  :   its  Nature  and  Laws 

having  the  same  practical  effect  upon  the  estimation  of 
the  utility  of  objects  of  consumption  corresponding  to  isolated 
organic  wants  or  tendencies.  Thus  in  estimating  the  utility 
of  a  given  increment  of  sugar,  it  is  a  matter  of  indifference  whether 
the  lowered  utility  imputed  to  it  is  the  result  of  diminution  of 
desire,  through  lowering  of  nervous  sensitivity  by  repetition 
of  stimulation,  or  whether  it  is  the  consequence  of  the  peculiar 
organic  sensations  of  satiety  produced  by  the  quantity  of  the 
stimulus.  The  practical  consequences  for  the  estimation  of 
utility  are  the  same.  But  when  we  come  to  examine  the  psy 
chology  of  the  two  phenomena,  differences  appear  of  such  a 
nature  as  to  make  separate  studies  of  the  two  cases  necessary. 

It  was,  we  have  seen,  to  the  effect  of  repetition  of  satisfaction 
of  desire  in  producing  habit  and  dulling  of  sensitivity  (and 
to  its  corollary,  limitation  of  capacity  of  appropriation)  that 
Bentham  chiefly  appealed  to  substantiate  his  general  principle 
of  Limiting  Value.  Repetition  of  stimulation  so  modifies  the 
physiological  or  dispositional  conditions  of  the  feeling  as  to  change 
the  relative  intensity  of  the  feeling  in  the  direction  indicated  by 
the  law.  It  is  clear  that  the  degree  of  worth  feeling  which 
follows  upon  the  judgment  or  assumption  of  the  existence  or 
non-existence  of  any  object  will  be  determined  by  the  disposition 
created  by  previous  acts  of  consumption  or  appreciation,  but  it 
does  not  follow  that  repetition  will  have  the  same  effect  upon  all 
these  dispositions.  As  the  presuppositions  of  the  different 
feelings  vary,  so  on  the  side  of  the  object,  the  factor  of 
frequency  or  quantity  may  vary  in  meaning.  We  must  there 
fore  consider  the  influence  of  the  factor  of  repetition  upon  the 
different  types  of  feeling  separately. 

The  effect  of  repetition  upon  one  type  of  worth  feeling, 
the  case  where  the  feeling  of  value  is  conditioned  by  preceding 
sensation-feelings,  is  simple  and  evident.  An  object  which 
has  the  capacity  to  satisfy  a  sense-tendency  becomes  an  object 
of  worth  when  it  is  judged  or  assumed  to  exist  or  not  exist. 
The  sensation  feelings  which  arise  in  the  process  of  consumption 
are  not  worth  feelings,  but  the  worth  feelings  are  determined  solely 
by  the  dispositions  created  by  the  processes  of  consumption. 
The  repetition  of  stimulation,  the  dulling  of  sensitivity,  lowers 
the  capacity  of  judgment  or  assumption  of  existence  to  call  out 
our  worth  feeling,  for  the  reason  that,  in  this  case,  the  modifica 
tions  of  sensation-feelings  are  the  sole  determinants  of  the  worth 
feelings.  Thus  sugar  is  an  object  of  worth  because  it  satisfies 


The  Laws  of  Valuation  161 

a  sense-tendency,  and  the  stimulation  of  the  sense-organ  gives 
pleasure.  But  the  fatigue  of  nervous  substance  through  suc 
cessive  stimulations  reduces  the  sensitivity.  There  is,  to  be 
sure,  a  limited  degree  to  which  fusion  of  sensations,  and  the 
simultaneous  moderate  stimulation  of  different  sense-tendencies 
may  produce  a  greater  degree  of  pleasure  for  a  longer  time' 
But  this  qualitative  sublimation  of  the  gross  quantitative  factor 
is  possible  only  within  very  narrow  limits.  Ultimately  sensitivity 
is  dulled,  and  the  judgment  or  assumption  of  the  existence  of 
the  stimulating  object  fails  to  call  out  worth  feelings,  unless  the 
judgment  is  modified  by  secondary  instrumental  judgments 
Conative  activity  is  arrested,  and  we  have  value  movement 
toward  another  object. 

When,  however,  we  turn  from  immediate  pleasure-causation 
to  the  study  of  worth  feelings,  that  is  feelings  where  the  relation 
e  object  to  the  disposition  is  not  one  of  immediate  stimu 
lation,  but  is  mediated  by  presumption,  judgment,  and  assump 
tion,  it  is  not  so  certain  that  the  principle  of  habit  or  dulling  of 
sensitivity  can  be  applied  directly  as  a  description  of  changes  in 
worth  feeling.  We  must  rather  examine  in  detail  the  effect  of 
repetition  on  each  type  of  feeling. 

Let  us  first  consider  certain  feelings  of  simple  appreciation 
Uowmg    upon    presumption   or    existential   judgment.      Per 
sonal  fame,  as  expressed  in  the  applause  of  others,  is  normally 
the  object  of  worth  feeling.     This  feeling  is  not  the  result  of 
the  stimulation  in  the  applause,  but  of  a  judgment  of  existence 
an  ideal  object  which  the  applause  expresses.     Repetition 
of  this  applause  and  of  the  accompanying  judgment  is  ordinarily 
flowed    by    emotional    indifference.     That    which    was    first 
recognised  with  intensity  of  emotion  is  finally  recognised  with 
out  emotional  disturbance.     Or  again,  we  may  take  the  case 
*  a  man  (e.g.,  a  miner)  whose  occupation  brings  him  face  to  face 
with  constant  danger.     The  judgment  of  the  existence  of  the 
is  at  first  accompanied  by  more  or  less  intense  emotional 
Disturbance,  but  with  repetition  of  the  judgment  his  sense  of 
danger  becomes  dulled.     It  would  appear,  then,  that  repetition 
1  ]udgment  feelings  involves  dulling  of  sensitivity,  and  there 
fore  diminution  of  worth  feeling. 

The  first  part  of  this  inference  is  certainly  true,  and  the  latter 

2ems  to  be  a  necessary  corollary  from  the  first.     But  before 

admitting   its   necessity   we   must   observe   more   closely   what 

actually  takes  place  in  such  cases.     The  phenomena  are  more 


,62  Valuation:    Us  Nature  and  Laws 

complex  than  appears  at  first  sight.    Can  we  logically  infer 
thatP  because  of  this  dulling  of  sensitivity,  the  conative  dis 
positions  corresponding  to  these  two  objects,  fame  and  danger 
have  been  equally  diminished,  and  that  the  objects  have  tost 
their  worth,  positive  or  negative?     I  think  not     For,  m  t 
fir      pice,  the  sensitivity  which  is  dulled   is   the   emotional 
dlturbance,   and   this   owes    its    first    intensity    to    coiUras  • 
to  the  disturbance  of  some   presumption  or  judgment  hab 
which  preceded,  and  to  the  effort  involved  in  the  readaptation 
mciden?  to  realisation  in  a  new  judgment  of  the  existence 
the   formerly  non-ex.stent  object      With   the  growth   o  ^con 
viction  the  emotional  disturbance  disappears  and  the  mte    ity 
o   sensitivity  diminishes.     But  it  does  not  follow  that  the  mean 
ing  of  the  feeling,  its  transgredient  or  immanental  reference  il 
depth  and  breadth  in  the  personality,  has  diminished 
matter  of  fact,  experience  shows  us  an  entirely  f  «^f  at™e 
When  once  the  conviction  of  the  existence  of  the  fame  or 
danger    as  the  case  may  be,  has  been  formed,  the  feeling  as 
leXnt    mood,    or  affective  sign  is  always   present   as  an 
und  ™'  in  the  emotional  experience  of  the  person,  co  ounng 
all  his  worth  judgments.     It  constitutes  a  presupposition  . 

of  this  fact  is  to  be  found 


in  the  substitution  of  assumption,  implicit  or  «£ 
ment.     In  another  connection1  it  was  shown  that 
Tudgment  is  of  the  relational  type,  that  is.  it  appeals  only  after 
'arrest,  is  association  after  disjunction.     In  all  other  cases  reahty 
is  assumed.     In  all  existential  judgment,  adaptat.on,  and 
ore  emotional  disturbance,  is  involved,  with  its  correspondmg 
intensity.     As  with  repetition  the  judgment  passes  over  in  o 
ssumption    the    intensity  decreases,   the    energy   involved    m 
emZnal  disturbance  is  used  up    and  dulling  of  —  ty 
follows     But  this  sensitivity  is  only  one  aspect  of  the  1 
attitude      When   judgment   has   given   place   to   assumption, 
with  its  feeling  of  the  imagination,  its  generic  sentiment  or  afl 
tive  sign  the  sensitivity  has  indeed  diminished  m  intensity-  he 
feeling  becomes  practically  intensity-less-but  the  meamng  of  the 
ee  ing  itslpth  and  breadth  in  the  personality,  may  remain  un- 
mpafred  or  may  even  increase.     In  the  illustrations  cons.dered, 
heTnded  meaning  of  the  idea  of  fame  or  of  danger  remains  m 
the  form  of  such  affective  correlates.    These  correlates  may  1 


1  Chapter  II,  p.  42  ff- 


The  Laws  of  Valuation  163 

reinstated,  either  by  the  explicit  assumption  with  its  feeling  of 
the  imagination,  or  by  the  implicit  assumption  embodied  in  a 
general  concept,  the  feeling-tone  of  words.  When  the  feeling 
is  of  the  first  type,  the  subject  often  continues  to  realise  in 
imagination  the  felt  meaning  of  objects  and  ideas  when  actual 
sense  experience  or  specific  adaptive  judgment  is  without  affec 
tive  reaction.  In  phenomena  of  the  second  type  the  mere 
affective  abstracts  themselves,  which  have  the  funded  meaning 
of  past  experience  (e.g.  of  fame  and  danger),  and  still  more — as 
shown  in  the  illustrations  of  the  preceding  chapter,1  those  which 
have  a  social  origin  and  meaning,  continue  as  felt  presuppositions 
of  further  emotional  experience. 

The  question  arises,  finally,  as  to  the  effect  of  repetition 
upon  the  assumption  feelings  of  the  two  types  described,  the 
feelings  of  imagination  and  the  affective  abstracts  or  signs  which 
take  the  place  of  particular  emotional  reactions.  Repetition 
does  not,  I  think,  reduce  the  worth  suggestion  of  these  affective 
experiences.  They  are  modified  only  when  judgment  makes 
the  assumption  untenable. 

The  fact  that  these  phases  .of  affectivity  are  in  some  way 
different  from  sensation  feelings  and  particular  emotions  or  judg 
ment-feelings,  has  not  escaped  the  observation  of  students  of 
these  phenomena.  This  difference  is  ordinarily  described  as 
consisting  in  the  fact  that  the  feelings  of  the  imagination  (Sax- 
inger)  or  the  affective  abstracts  and  signs  (Paulhan)  are  not 
subject  to  the  law  which  governs  particular,  "  real "  feelings,  i.e., 
the  law  of  diminution  of  intensity  with  repetition,2  and  has  been 
explained  by  the  theory  that  the  actualisation  of  these  feelings, 
whether  by  explicit  imagination  or  by  the  verbal  correlate  of 
habitual  judgment,  does  not  involve,  as  in  the  case  of  other 
feelings,  the  expenditure  of  energy,  and  therefore  the  diminution 
of  the  capacity  of  the  affective  dispositions  for  actualisation. 
Now  the  fact  of  this  difference  we  may  admit,  but  not  its  ex 
planation.  Repetition  affects  the  intensity  of  all  sensitivity, 
even  of  the  hedonic  accompaniments  of  worth  feeling,  but  the 
meaning  of  the  feeling  is  relatively  independent  of  its  intensity. 
Dulling  of  sensitivity  for  emotional  disturbance  cannot  be  cor 
related  with  decrease  in  worth  or  affective- volitional  meaning. 

1  Chap,  v,  p.  137,  note. 

2  Chap,  v,  p.  128. 


1 64  Valuation  :   its  Nature  and  Laws 


(b)   The  Law  of  Satiety  does  not  apply  to  Feelings 
of  Value. 

The  second  psychological  principle  to  which  the  law  of 
Diminishing  Value  is  referred  for  its  foundation  is  the  principle 
of  satiety.  In  determining  whether  these  formulations  are 
applicable  to  all  acts  of  appreciation,  we  found  it  necessary 
to  treat  separately  the  two  principles  of  habit  and  satiety, 
although  from  the  standpoint  of  their  practical  effect  upon 
the  determination  of  the  utility  of  objects  they  could  be  treated 
as  identical. 

What  is  to  be  understood  by  satiety  seems  perfectly  clear 
in  the  case  of  the  consumption  of  a  simple  good  which  corre 
sponds  to  an  isolated  sense  tendency.  As  described  in  the  most 
general  terms,  it  is  a  law  connecting  intensity  of  sensation  with 
quality  of  feeling.  With  increase  in  the  intensity  of  the  sen 
sation  there  is  a  relatively  slower  increase  in  intensity  of  pleasure 
until  a  critical  point  is  reached  where  the  pleasure  begins  to 
fall  off.  Finally,  a  second  critical  point  in  the  curve  is  reached 
where  the  positive  coefficient  of  pleasure  passes  over  into  a 
negative  coefficient  of  unpleasantness  or  pain.  The  second 
point  is  described  as  satiety  or  Ubersattigung.  Now  this 
conception  of  an  entire  change  in  quality  of  the  affective  tone 
of  the  sensation  quality,  while  a  description  quite  sufficient  for 
the  uses  of  the  economist  in  his  account  of  the  effect  of  quantity 
upon  utility,  is,  from  the  standpoint  of  psychological  analysis, 
a  fiction,  an  undue  simplification  of  the  phenomenon.  As  a 
matter  of  fact,  the  situation  is  not  so  simple.  What  we  have 
here  is,  not  pure  pleasure  until  a  certain  point  is  reached  and 
then  a  sudden  change  to  unpleasantness,  but  rather  a  mixed 
phenomenon.  The  pleasure  arising  from  the  stimulation  of 
any  organic  tendency  is  made  up  of  the  affective  tone  of  the 
sensation  itself  plus  the  affective  tone  of  organic  sensations 
associated  with  it.  With  over-stimulation  the  latter  gradually 
become  unpleasant,  but  at  first  not  sufficiently  so  to  modify  the 
dominantly  pleasurable  tone  of  the  total  psychosis.  That 
which  constitutes  the  moment  called  satiety  is  really  the  pre 
dominance  of  the  unpleasantly  toned  organic  content.  Such 
a  thing,  then,  as  transformation  of  pleasure  into  its  negative 
does  not  exist.  What  has  been  so  described  is  really  a  driving 
out  of  the  pleasure  feeling  attached  to  the  content,  as  at  first 


The  Laws  of  Valuation  165 

experienced,  by  a  more  intense  unpleasant  feeling  arising  from 
competing  foreign  contents. 

Now  while,  as  it  has  been  said,  this  interpretation  of  the 
phenomenon  does  not  in  any  way  affect  the  consequences  drawn 
from  the  principle  of  satiety  for  the  utility  of  goods  in  consump 
tion,  it  does  modify  considerably  our  conception  of  the  bearing 
of  the  principle  of  satiety  upon  acts  of  appreciation  in  general. 
In  the  first  place,  it  is  not  even  clear  that  we  can  say  that  satiety 
is  the  normal  consequence  of  satisfaction  of  desire  itself,  that 
is,  that  it  lies  in  the  very  nature  of  desire.  It  is  even  abstractly 
conceivable  that,  if  the  successive  increments  of  a  good  could 
be  appropriated  in  such  a  way  as  not  to  create  organic  dis 
turbances  and  thus  introduce  new  content,  the  principle  of 
dulling  of  sensitivity  would  hold  for  isolated  tendencies,  while 
the  principle  of  satiety  would  not. 

Much  more  important  do  these  considerations  become 
when  we  turn  to  the  appreciation  of  so-called  ideal  goods. 
When  we  recognise  that  the  phenomena  of  satiety  are  not  in 
herent  in  the  appreciative  activity  itself,  but  are  accompanying 
phenomena,  sensational  redundancies  secondary  to  appre 
ciation,  then  new  light  is  thrown  upon  certain  complicated 
facts  of  appreciation.  In  the  first  place,  we  may  notice 
that  the  appreciative  consciousness  itself  makes  a  clear  dis 
tinction  between  the  role  which  these  accompanying  experi 
ences,  described  as  satiety,  play  in  the  two  cases  of  con 
sumption  and  ideal  appreciation.  We  may  express  this  by 
saying  that  when  the  phenomenon  of  satiety  enters  into  the 
process  of  consumption,  the  subject  transfers  its  meaning  to 
the  object  of  desire  ;  he  has  enough  of  the  object  itself.  But 
when  in  the  pursuit  of  ideal  goods  or  in  acts  of  appreciation, 
the  unpleasant  organic  sensations  appear  as  accompaniments  of 
over  -  stimulation  and  fatigue,  he  recognises  their  secondary 
character,  and  does  not  transfer  them  to  the  object;  they 
do  not  affect  the  value  of  the  object.  This  we  express  in 
our  judgments  that  the  thirst  for  knowledge,  beauty,  or  good 
ness  is  insatiable.  Psychologically  this  means  that  the  sense 
of  value  as  affective-volitional  meaning  persists  as  an  undertone 
in  consciousness.  The  transgredient  or  immanental  reference 
of  the  feeling  goes  beyond  the  organic  unpleasantness  of  the 
moment.  Desire  remains  present  dispositionally  in  the  form 
of  sentiment,  mood,  or  affective  sign.  A  mere  imagination, 
assumption,  of  the  existence  of  the  object  suffices  to  call  up 


1 66  Valuation :   its  Nature  and  Laws 

these  representative  worth  suggestions,  when  the  actual  pro 
cesses  of  perception  and  judgment  involved  in  the  realisation  of 
the  objects  would  be  accompanied  by  the  unpleasant  conditions 
which  we  have  described  by  the  term  satiety.  The  worth 
moment  does  not  lie,  then,  in  the  hedonic  accompaniments, 
but  in  the  transgredient  and  immanental  reference  to  dispo 
sitions  presupposed,  which-  may  persist  with  change  in  hedonic 
redundancies.  It  is  true  that  these  accompaniments,  this  satiety, 
may  become  the  object  of  a  negative  worth  judgment,  and  that 
this  negative  worth  may  then  compete  with  the  positive  worth 
of  the  object,  but  in  this  case  we  have  a  new  situation.  The 
state  of  satiety  has  been  presented  as  an  object  of  worth  feeling, 
and  is  itself  no  longer  worth  feeling,  or  an  aspect  of  it. 

It  appears,  then,  that  when  the  feeling  of  satiety,  the 
peculiar  mass  of  unpleasant  organic  sensations  that  goes  by 
that  name,  arises  in  connection  with  judgment  feelings,  this 
does  not  mean  that  the  worth  or  judgment-feeling  has  passed 
over  from  a  positive  to  a  negative  quality,  but  rather  that  the 
physiological  processes  involved  in  the  presentation  and  judg 
ment  of  the  object,  and  in  the  accompanying  emotional  ex 
perience,  have  been  exhausted  and  that  the  pleasurable  accom 
paniments  have  passed  over  into  unpleasant.  It  is  not  the  dis 
position,  the  capacity  for  worth  feeling  itself,  which  is  exhausted, 
but  other  dispositions  involved  in  the  presentation  and  judg 
ment  of  the  object.  In  such  cases  the  valuation  may  persist 
in  the  form  of  an  assumption  feeling,  during  the  experience 
described  as  satiety. 

(c)  Conclusion — The  Limited  Application  of  the  Laws  of  Dulling 
of  Sensitivity  and  Satiety. 

In  considering  the  bearing  of  these  two  factors  of  habit 
and  satiety  on  worth  feeling  in  general,  some  such  analysis 
as  the  preceding  is  all-important  if  we  would  avoid  confusion. 
Thus  Gossen,  who  takes  appreciation  of  a  work  of  art  as  illus 
trating  the  general  law  of  Limiting  Value,  fails  to  distinguish 
the  several  factors  involved  in  such  appreciation.  He  tells  us 
that  "  a  work  of  art  will  bring  to  the  artist  who  beholds  it  the 
highest  enjoyment  in  that  moment  when  he  has  observed  it 
long  enough  to  grasp  all  its  elements.  This  enjoyment  will 
sink  steadily  with  continued  study,  and,  after  a  longer  or  shorter 
time,  according  to  the  nature  of  the  object  and  the  observer, 


The  Laws  of  Valuation  167 

the  latter  will  become  tired  ;  satiety  will  appear,  even  when  he 
seeks  to  enjoy  works  of  the  same  kind.  If,  after  a  longer 
or  shorter  time,  according  to  the  object  and  the  observer,  a 
desire  for  the  repetition  of  the  pleasure  arises,  he  will  then, 
on  account  of  his  previous  knowledge  of  the  work  of  art,  reach 
the  highest  point  of  enjoyment  in  a  shorter  time.  But  this  point 
will  be  reached  the  less  easily  and  the  less  frequently  the  more 
often  and  the  more  frequently  the  repetition  has  taken  place." 
But  in  this  description  Gossen  fails  to  distinguish  two  different 
types  of  feeling  involved  in  this  appreciation.  Had  he  observed 
more  closely,  he  would  have  found  that,  while  this  is  a  true 
account  of  what  takes  place  in  the  purely  sensuous  pleasure  in 
the  formal  factors,  nevertheless  the  sentiment  felt  toward  the 
content,  the  object  expressed  by  the  work  of  art,  might  quite 
easily  have  remained  unaffected  by  repetition.  The  feeling 
in  this  latter  case  is  an  assumption  -  feeling  and  presupposes 
dispositions  quite  different  from  those  the  stimulation  of  which 
affords  the  pleasure  of  the  formal  element.  Actualisation  of 
these  dispositions  has  by  no  means  the  same  consequences  as 
stimulation  of  dispositions  involved  in  pleasure  causation. 

IV.  EXTENT  OF  THE  APPLICATION  OF  THE  LAW  OF  LIMITING 
VALUE  TO  IDEAL  OBJECTS — THE  GENERAL  PROBLEM 

The  preceding  study  of  the  two  principles  of  habit  and  satiety, 
upon  which  the  general  law  of  Limiting  Value  is  based,  has 
shown  them  to  hold  only  for  sensation-feelings  and  for  intensities 
of  emotional  disturbance  involved  in  judgment-feelings.  The 
latter,  the  feeling  intensities  which  diminish  with  repetition  of 
judgment,  may,  indeed,  be  sensation  -  feelings  also,  resonances 
of  the  organism  following  upon  the  psychical  activity  of  judg 
ment.  To  the  worth  feelings  themselves,  the  affective-volitional 
meaning  acquired  in  processes  of  judgment  and  assumption,  the 
laws  of  dulling  of  sensitivity  and  satiety  do  not  necessarily 
apply,  for  they  are  not  sensation-feelings.  We  should  avoid, 
however,  at  least  at  this  stage  of  the  discussion,  the  inference 
that  the  law  of  Limiting  Value  is  not  universally  applicable,  as 
claimed  by  its  founders.  Because  the  psychological  analysis 
upon  which,  as  a  theory,  it  was  based,  is  faulty,  it  does  not 
follow  that  it  fails  as  a  description  of  fact.  It  is  entirely  con 
ceivable  that  it  represents  the  fate  of  all  processes  of  valuation 

1  Gossen,  Entwickelung  der  Gesetzet  etc.,  p.  5. 


1 68  Valuation:   its  Nature  and  Laws 

whatsoever,  but  that  the  principle  rests  upon  the  limitation  of 
other  capacities  than  those  of  sense  stimulation. 

"  There  are  two  types  of  valuation  in  which  the  object  as  sense 
or  presented  is  not  the  immediate  cause  of  feeling,  but  in  which 
the    feeling -disposition   is  mediated  through  intellectual  pro 
cesses  of  judgment  or  assumption.      They  are  worth  feehngs 
which  rise  upon  processes  of  ideal  reconstruction  of  the  objec 
through  judgment.    These  reconstructions  may  be  either  the  work 
of  instrumental  judgments,  in  which  case  the  corresponding  values 
are  instrumental  values,  or  processes  of  aesthetic  reconstruction 
which   individualise   the   object   as  an  harmonious  whole,  am 
make  it  of  intrinsic  value.     We  must  now  examine  these  wortl 
feelings  and  their  modifications  to  determine  to  what 
this  law  of  valuation  holds. 

The  objects  of  such  types  of  valuation,  and  their  fundt 
meanings,  are   "founded"   objects  in   the   sense   of  our  intro 
ductory  analysis.     The  various  types  of  such  objects  already 
enumerated,    the    psychical    objects    of    appreciation     such    as 
beauty  and  grace  of  form  in  perceptual  objects-founded  qualities 
which  emerge  in  the  activities  of  consumption  and  in  instinctive 
activity  generally,  such  as  cleanliness,  manners,  etc     the  qu; 
ties  and  dispositions  of  persons,  viewed  either  in  their  intrinsic 
worth  as  worths  of  the  person,  or,  instrumentally,  as  the  basis 
of  social  participation,  and  finally  objects  of  utilisation    of  ex 
change-more   especially   the   medium   of   exchange-all         se, 
although  they  inhere,  so  to  speak,  in  objects  of  immediate  sensa 
tion  and  presentation,  are  themselves  not  the  objects  of  such 
processes,  but  rather  of  judgment  and  assumption      As  such 
their  funded  meaning  is  not  determined  by  simple  feeling,  1 
by  feelings  which  have  as  their  presuppositions  the  processe 
of  judgment  and  assumption  of  which  they  are  the  objects 
It  is  of   course  possible,  and  it  frequently  occurs,  that  the* 
worths  or  meanings  are  attributed  to  the  sensations  and  pr. 
sentations   themselves,   and   then  we  speak   of   them   as 
plementary  values  of  the  latter,  but  for  psychological  analysi 
they  must  be  kept  quite  distinct. 

Let  us  now,  with  these  considerations  in  mind,  turn  first  tc 
the  problem  of  the  application  of  the  law  of  Limiting  Valu 
to  the  objects  of  utilisation  and  exchange,  the  type  of 
tension  of  simple  appreciation  which  takes  place  through  t 
interpolation  of  instrumental  judgments. 


The  Laws  of  Valuation  169 

i.    Application   of   the  Law  to  all  Instrumental   Values, 
(a)  The  Law  of  Marginal  Utility  and  its  Explanation. 

Economic  worth  theory  took  its  rise  from  the  obvious  applica 
tion  of  the  principle  of  Limiting  Value  to  objects  corresponding 
to  isolated  sense  tendencies.  Its  extension  to  valuation  in 
general  was  made  possible,  as  we  have  seen,  historically,  by  the 
unanalysed  substitution  of  wealth  in  general,  or  money,  for  the 
particular  homogeneous  good  corresponding  to  an  isolated  sense- 
tendency.  In  the  concept  wealth,  and  still  more  in  that  of 
money,  we  have  the  most  abstract  possible  symbol  for  objective 
value  and  thus  indirectly  for  subjective  worth.  If  therefore  the 
principle  of  relativity  developed  for  isolated  processes  of  con 
sumption  holds  for  the  valuation  of  successive  increments  of 
wealth  in  general,  it  would  seem  that  the  inference  as  to  the 
universality  of  its  application  was  justified,  and  with  it  the 
hedonistic  assumptions  of  its  founders. 

Now,  as  a  matter  of  fact,  we  shall  see  that  the  merely  em 
pirical  law  of  modification  of  worth  feeling,  developed  by  Bentham 
and  his  successors,  does  hold  within  limits  for  the  valuation  of 
successive  increments  of  wealth ;  but  it  does  not  follow  that  the 
psychological  explanation  of  this  empirical  law  is  the  same  in  both 
cases.  Since  the  value  feelings  accompanying  successive  incre 
ments  of  wealth  are  judgment- feelings  while  in  consumption 
the  feelings  are  sensation  -  feelings,  the  presuppositions  being 
different,  the  law  of  their  modification  may  be  different.  As 
illustrative  of  our  position,  we  may  refer  to  another  region  of 
psychology.  If  the  principle  of  relativity  called  Weber's  Law, 
developed  first  for  the  perception  of  differences  of  intensity, 
is  found,  as  Wundt  maintains,  to  hold  also  for  the  perception  of 
relative  differences  in  pitch,  it  does  not  follow  that  the  explana 
tion  is  the  same.  Intensity  and  pitch  are  two  different  things, 
in  fact  independently  variable.  The  subsumption  of  this  law 
of  relative  pitch  under  the  law  of  relative  intensity,  because  the 
empirical  law  is  the  same,  would  be  unjustified.  Likewise,  with 
out  further  analysis,  we  cannot  infer  that,  because  the  empirical 
law  of  the  modification  of  judgment-feelings  in  instrumental 
valuation  is  the  same  as  that  in  immediate  pleasure-causation, 
the  two  phenomena  are  identical. 

The  feelings  of  value  corresponding  to  the  concept  wealth  are, 
with  certain  significant  exceptions  which  we  shall  consider  later, 


i  70  Valuation  :   its  Nature  and  Laws 

judgment  feelings.    Whatever  be  the  concrete  object  symbolising 
wealth,  the  essence  of  the  conception  lies  in  the  fact  that,  through 
the  interpolation  of  certain  judgments  as  to  its  reapplicability  for 
different  purposes,  whether  through  extension  of  simple  use  or 
through  exchange,  the  object  is  removed  from  the  immediacy  of 
simple  pleasure-causation,  of  stimulation  or  consumption.     The 
presupposition  of  such  a  feeling  of  value  is  then  the  fundamental 
conative  tendency  as  modified  by  subsidiary  judgments  as  to 
reapplicability  of  the  object.     An  object's  value  in  use,  notwith 
standing  the  extension  of  its  applicability  through  exchange, 
always  retains  at  least  'a  faint  reference  to  the  fundamental 
desires  to  which  it  first  afforded  satisfaction,  but  the  more,  as  in  a 
"  money-economy,"    it  becomes   abstracted  from  this  primary 
reference  and  fills  the  role  of  mere  symbol  of  exchange,  the  more 
its  purely  instrumental  character  becomes  emphasised.     Now  the 
consequence  of  this  is  that  successive  increments  of  any  good, 
which,  after  a  certain  point  in  the  immediate  or  direct  satisfac 
tion  of  desire  is  reached,  can  then  produce  a  modification  of  feeling 
only  through  exchange  for  other  objects,  are  not  direct  stimuli, 
but  can  modify  feeling  only  through  the  indirect  process  of  calling 
out  judgments. 

What  is  the  effect,  then,  of  successive  increments  of  wealth, 
making  necessary  repeated  judgments  of  reapplicability  of  the 
good,  upon  the  economic  threshold  or  threshold  of  value  feeling  ? 
Obviously  such  increase  raises  the  threshold  of  value, 
new  judgment  of  application  diminishes  the  capacity  of  the  good 
for  application,  i.e.  with  each   added  increment  of  the  good, 
especially  of  the  abstract  symbol  money,  the  more  significant 
and  important  fundamental  condition  worths  of  the  subject  be 
come   satisfied,    the   possibilities    of   reapplication,    always    for 
any   individual   a   limited   universe,   become   progressively   ex 
hausted  and  the  relation  between  the  object  and  the  immediate 
appreciation   becomes   more   and  more   mediated   and  remote. 
In  other  words,  in  order  that  a  significant  instrumental  relation 
may  be  established  between  the  new  increment  and  a  condi 
tion    or    personal  worth,  the    increment    must   be  appreciably 

greater. 

It  is  clear  then,  that  the  economist  is  justified  in  extend 
the  conception  of  Diminishing  Utility  or  Limiting  Value  to  include 
all  those  values  mediated  by  successive  judgments  of  reapplic 
tion,  as  well  as  to  those  conditioned  by  repetition  of  immediat< 
stimulation.     Moreover,  the  law  of  Marginal  or  Final      tihty, 


The  Laws  of  Valuation  1jl 

deduced  from  this  more  general  principle,  according  to  which 

J  value  of  a  sum  of  goods  depends,  not  upon  the  "total" 

ity,  but  upon  the  marginal  or  "  final "  degree  of  utility  of  the 

last  addition,  is  universally  applicable  to  all  objects,  the  value 

1  which  consists  in  their  capacity  for  reapplication.     It  is  not 

so  clear  however,  that  this  fact  is  to  be  explained  by  reference 

the  laws  limiting  our  capacity  for  mere  stimulation.     It  is 

.ther  probable  that  more  complex  processes  of  imaginative  and 

ideal  construction  of  ends  are  involved. 

(b)  Certain  Limitations  to  the  Law  of  Marginal  Utility. 

Certain  significant  limitations  of  this  law  will  bring  the  situation 
into  clearer  relief.  It  has  been  recently  pointed  out  by  Simmel ' 
that  what  we  may  call  the  curve  of  value,  or  funded  meaning 

f  money,  for  any  individual  cannot  be  viewed  as  a  plotting  of 
progressive,  gradual  changes,  but  must  rather  be  looked  upon  as 
representing  a  series  of  discrete  stages.  Thus,  if  the  individual 
has  a  given  income,  he  lives  normally  in  a  universe  of  certain 
relatively  fixed  possibilities  from  which  many  of  the  most  signif 
icant  condition  and  personal  worths,  such,  for  instance,  as  esthetic 
satisfactions  and  social  position,  in  so  far  as  they  are  dependent 
upon  money,  are  excluded.  Gradual  additions  to  his  income 

re  valued  by  him  according  to  the  law  developed  above  for 
within  that  relatively  fixed  system,  the  reapplications  become 
progressively  less  and  less  important.  But  if  the  increment 

i  of  sufficient  amount  to  change  the  universe  of  possibilities  the 
level  of  valuation  is  changed.  New  worths  of  condition  'and 
person  come  within  the  individual's  horizon,  new  possibilities  of 
application  appear,  and  the  curve  begins  again  on  a  new  level 

/\S   cMmmfM    rprmrl^e:     nno-n-t-ifir     ,,.\^         •*.    • 

uaiKb,  c          itv.  when  it  is  grasped  as 


whoi  K.        '  '  e     as  a  ™e 

whole  may  be  transmuted,  sublimated,  into  quality.     Of  course 

ZshoidsrT,level  the  T agam  begms  to  w°rk'  modify-e  ^ 

esnold  in  the  manner  described  above 

A  second  modification  of  this  law  appears  closely  connected 
fee  one  previously  described.     In  the  former  case,   the 
charac  enst.c  feature  is  to  be  found  in  the  fact  that  if  a  certain 
f  money,  say  a  million  dollars,  is  received,  its   subjec 
tive  worth  ,s  not  to  be  calculated  according  to  the  law  of  margmal 
.  is  at  first  a  direct  appeal,  without  mediating  in- 
rumental  judgments,  to  the  worths  of  the  personality  (social 

1  Simmel,  Philosophic  des  Geldes,  pp.  250-76. 


172  Valuation  :   its  Nature  and  Laws 

esteem,  independence,  etc.).  The  threshold  is  qualitative,  not 
quantitative.  In  a  similar  manner,  such  a  sum  of  money  acquires 
this  qualitative  character  when  it  is  viewed  as  an  ideal.  In  the 
case  of  the  miser,  for  instance,  who  sets  before  himself  a  certain 
sum  viewed  as  a  unity,  as  an  ideal,  the  value  of  the  successive 
increments  of  money  does  not  follow  the  law  of  marginal  utility. 
As  he  nears  his  goal  the  smallest  increment  may  have  almost  an 
absolute  value,  and  even  in  the  earliest  stages  of  the  process  of 
acquisition,  the  value  of  any  increment  is  not  to  be  calculated  ac 
cording  to  this  law.  The  significant  feature  of  this  situation  is  to 
be  found  precisely  in  the  suppression  of  instrumental  judgments 
and  the  substitution  for  them  of  an  assumption  of  the  possible 
application  of  the  wealth  as  a  whole.  To  the  minute  increment 
is  imputed  the  assumption  feeling  that  goes  with  the  individuated 
whole,  or  we  may  say  that  the  most  insignificant  moment  in  the 
series  borrows  worth  from  the  end  feeling  of  the  series.  As  we  have 
seen  in  our  study  of  the  threshold,  intrinsic  valuation  of  a  minute 
object  as  part  of  an  individuated  whole  lowers  the  threshold  of 
worth  for  that  object.  Whereas,  then,  in  the  instrumental 
valuation  of  wealth  each  successive  amount  calls  forth  worth 
feeling  only  through  the  mediation  of  a  series  of  instrumental 
judgments — and  value  decreases  with  the  remoteness  of  the 
object  from  the  fundamental  desire,  in  the  intrinsic  valuation 
of  wealth,  on  the  other  hand,  as  an  individuated  whole,  the 
valuation  which  follows  upon  this  individuation,  is  subject 
neither  to  the  dulling  of  sensitivity  which  follows  upon  repetition 
of  sensation  feelings  nor  to  the  limitation  of  judgment  capacity 
which  inheres  in  instrumental  valuation. 

A  consideration  of  these  phenomena  enables  us  then  to  under 
stand  the  limited  application  of  the  law  of  marginal  utility 
to  the  founded  ideal  objects  of  economic  utility  and  exchange, 
without  subsuming  it  under  the  laws  of  dulling  of  sensitivity 
and  satiety,  which  apply  only  to  pleasure.  And — what  is  still 
more  important— it  enables  us  to  dispense  with  the  psychological 
fiction  of  the  continuous  variability  of  intensity  of  pleasure  with 
changes  in  increments  of  wealth,  an  assumption  which  underlies 
much  of  economic  worth  theory.  As  a  matter  of  fact,  however, 
it  must  be  recognised  that  some  economists  have  employed  this 
law  as  a  purely  empirical  principle  without  accepting  the  theo 
retical  fiction  of  infinitesimal  changes  in  worth  feeling.1 

1  Oskar  Kraus,  Zur  Thtvie  des  Wertes,  p.  64. 


The  Laws  of  Valuation 


v 


tmuity  and  extension  of  ,    i  lle  a  certam  con- 

reconstructions  °n 


of  sensitivity) 


It  is        H  °f  habU  <dulllng 

of 

of  desiiwhi  inn>Onf  ty,P<!  °f  reconstr-ti<»>  of  objects 

which  a 

intuitive  instead   of 

valuation  is  developed  I 

individuates  a  CTO™  n 

monious  group  rfetonen°s 

emerge  new  psychicT  o,,  H  H     K 

valued.     We  have  no     /«  ^      JuCtS  Which  are  intrinsically 

of  indivlduation   and!  stuZte  1      "'""  °f  'heSe  PrOCeSS- 

idy  the  laws  governing  the  intrinsic 


s  rec°nstruction  is 

fOm  °f  intrinsic 
construct'°n  ^  intuitive,  in  that  it 

Whok  of  ™anlng,  an  har- 
'  Pr°CeSS  °f  i 


1 74  Valuation  :   its  Nature  and  Laws 

valuation  of  these  psychical  objects,  more  especially  the  extent 
of  the  applicability  of  the  law  of  Limiting  Value. 

i.  Description  of  the  Laws  of  Complementary  Values- 

(a)   In  Economic  Valuation. 

The   general   characteristic   of   this   type   of   reconstruction 
is  to  be  found  in  the  fact  that  the  elements  of  a  total  group  of 
objects,  or  the  part-processes  of  a  total  conative  process  of  con 
sumption  or  acquisition,  are  so  related  to  each  other  as  to  be 
complementary.     In  other  words,  when  related  to  each   other 
as  elements  in  an  individuated  whole,  the  value  of  the  whole,  the 
degree   of  satisfaction   of  the  conative  tendency  presupposed, 
exceeds  the  value  of  the  sum  of  the  elements  taken  separately. 
The  values  acquired  in  such  processes  are  described  as  ( 
plementary  Values,  and  the  objects  toward  which  these  feelings 
are   directed   are    ideal    objects.      The    complementary   values 
thus  acquired  may,  indeed,  be  imputed  to  the  elements  taken 
separately,   but   they  are  not  the  values  of  the  elements    but 
of  the  object  founded  upon  the  harmonious  arrangement 
elements.     We  must  first  consider  this  doctrine  as  formula t, 
for  economics,  and  then  determine  the  extent  of  its  applicati- 
to  other  types  of  values. 

As  a  consequence  of  the  law  of  Marginal  Utility  the  psycho- 
logical  basis  of  which  we  have  been  considering,  the  econoi 
without  difficulty  draws  the  inference  that  the  value  of  a  supp  y 
of  goods  is  determined,  not  by  addition  of  the  utilities  actually 
got  from  the  different  parts,  that  is  by  the  sum  of  the  very  differ 
ent  utilities  of  the  different  portions,  but  by  the  utility  of  its 
least   important  part  multiplied  by  the  total  number  of   ea 
equally  large   part.     To  this  general  law  an  exception  is  how 
ever,   universally  recognised,   namely    where   a  sum   of   g od 
constitutes  a  unity  and,  as  such,  displays  a  certain  utility  effect 
whTch  is  not  equal  to  the  sum  of  the  utilities  of  the  separate 
"arts      As  a  consequence,   it   further  follows   that  any  given 
part  of  the  whole  has  imputed  to  it  a  utility  over  and  above 
Siat  which  it  would  have  as  an  *^**ti^^£ 
the  law  of  Marginal  Utility.     A  stock  illustration  is  that  of 
hun  er  and  his  powder,  ball,  and  flint.     Each  of  these  alone  is 
useless      All  of  them,  taken  together  with  the  labour  involved 
Tn  employing  them,  equal  the  value,  or,  in  the  more  accurate 
terms   of   the   economist,   the   discounted  value,   of   the   game. 


The  Laws  of  Valuation  175 

Because  each  alone  is  useless,  and  is  of  utility  only  in  combina 
tion,  the  value  of  each  is  equal  to  the  value  of  the  whole. 


(b)  Extra-Economic  Valuation — ^Esthetic  and  Ethical. 

In  this  illustration  we  have  a  unity,  an  individuated  whole, 
the  instrumental  value  of  which  arises  from  the  grouping  of  the 
elements,  the  parts  of  the  whole  being  interrelated  by  instrumental 
judgments.  But  the  same  general  principle  is  evident  in  cases 
where  the  instrumental  judgments  recede  into  the  background  and 
intrinsic  valuation  takes  their  place.  We  have  already  considered 
cases  in  which  wealth  viewed  as  a  whole  acquires  a  complementary 
value  of  an  intrinsic  character,  a  value  which  not  only  exceeds 
quantitatively  the  sum  of  the  instrumental  values  of  the  parts,  but 
changes  qualitatively  as  well,  where  quantity  is  sublimated  into 
quality.  In  the  two  cases  referred  to — where  the  reception  of  a 
round  sum  of  money  leads  to  its  intrinsic  valuation  as  a  whole 
(rather  than  to  an  instrumental  valuation  of  the  parts,  according 
to  the  law  of  Marginal  Utility,  as  when  the  amount  is  gradually 
acquired),  or  where  the  miser  values  his  wealth  as  a  whole  without 
reference  to  the  instrumental  value  of  the  parts — in  these  cases 
the  instrumental  judgments  have  sunk  into  the  background  and 
are  represented  merely  by  a  vague  assumption  of  instrumental 
value,  the  total  amount  being  valued  intrinsically  without  explicit 
reference  to  the  instrumental  value  of  the  parts. 

But  we  find  still  other  cases  where  the  general  principle  of 
complementary  values  applies  to  individuated  wholes,  where  the 
element  of  utility,  and  of  instrumental  judgments,  is  entirely  lack 
ing,  where  both  the  whole  and  the  parts  are  objects  of  intrinsic 
appreciation.  These  are  perceptual  and  ideal  unities  which  have 
an  aesthetic  or  ethical  character.  Two  illustrations  will  suffice 
to  indicate  the  character  of  these  constructions.  As  examples  of 
such  perceptual  wholes  we  may  take  a  face  in  which  all  the  features 
taken  together  are  charming,  or  a  landscape  in  which  the  river, 
the  hills,  the  valleys,  are  all  said  to  be  beautiful.  None  of  them 
would  be  beautiful  without  the  others  ;  each  has  all  the  beauty  of 
the  whole.  Of  such  ideal  constructs  the  individuality  of  a  person  is 
typical.  The  person  is  an  ideal  construct,  a  complex  of  traits 
or  dispositions  all  of  which,  when  thus  combined  to  make  the 
unique  whole,  create  that  founded  object,  the  person.  The 
intrinsic  worth  of  the  person  is  something  over  and  above  the 


176  Valuation:   its  Natiire  and  Laws 

separate  values  of  the  parts,  and  to  the  separate  trait  or  dis 
position  is  imputed  all  the  complementary  value  of  the  whole. 

(c)  General  Characterisation  of  the  Law. 

The  phenomena  described  by  the  principle  of  Complementary 
Values  are  now  before  us.  From  the  illustrations  given  it  is 
seen  to  be  a  process  by  which  the  range  of  valuation  is  extended, 
both  in  the  sphere  of  physical  objects  of  sensation  and  perception 
and  in  the  region  of  ideal  objects  of  imagination  and  judgment, 
a  process  by  which  the  working  of  the  law  of  Limiting  Value, 
which  holds  both  for  sensation  and  judgment  feelings,  is  modified. 
It  is  this  aspect  of  the  situation  which  has  received  special 
attention  from  Professor  Patten a  in  his  theory  of  consumption. 
The  older  doctrine  of  consumption,  he  tells  us,  does  not  take  into 
account  all  the  elements  of  pleasure  and  utility.  Besides  the 
gross  quantity  of  the  goods,  and  the  relation  of  this  quantity  to 
the  capacity  of  the  elementary  wants,  there  are  in  all  groups 
of  goods  capacities  for  rearrangement  which  are  outside  the  category 
of  quantity,  that  is  are  qualitative,  aesthetic.  A  group  of  goods, 
harmoniously  arranged,  is  capable  of  giving  indefinitely  greater 
pleasure  than  the  mere  sum  of  the  separate  pleasures  of  each  of  the 
components  of  the  group.  The  complementary  values  are  then 
imputed  to  the  components.  This  principle  is  conceived  as 
coming  in  to  modify  the  working  of  the  laws  of  habit  and  satiety. 
These  laws,  as  we  have  seen,  apply  to  all  sensation  feelings,  and 
consequently  the  law  of  Limiting  Value  applies  to  all  worth 
judgments  determined  by  such  feeling-dispositions.  But  since 
there  is  in  these  objects  of  sensation  feeling  a  certain  capacity  for 
harmonious  combination  which  extends  their  capacity  for  con 
tinued  stimulation,  these  laws  are  to  a  degree  modified.  "  Since 
these  aesthetic  goods  may  be  said  to  be  goods  without  the  point  of 
satiety  which  is  found  in  simple  economic  goods,"  and  "  since 
simple  aesthetic  goods  may  be  said  to  be  the  result  of  the  blending 
of  distinct  pleasures  into  one  group  and  the  aesthetic  pleasures 
seem  to  be  the  largest  harmonious  grouping  of  pleasures  that 
society  can  produce,"  it  would  follow,  Professor  Patten  thinks, 
that  "  progress,  in  the  sense  of  increase  of  value  for  the  in 
dividual  and  society,  must  lie  in  the  direction  of  harmonious 
consumption."  Despite  the  unfortunate  terminology,  which 
identifies  worth  with  pleasure  causation  and  speaks  of  summation 

1  S.  N.  Patten,  The  Consumption  of  Wealth,  Philadelphia,  1889. 


The  Laws  of  Valuation  177 

of  pleasures,  we  may  accept  this  as  a  true  account  of  the  ex 
tension  of  the  activities  of  simple  appreciation  through  the 
rearrangement,  reconstruction  of  the  objects  of  desire  and  feeling. 
But  Professor  Patten  has  added  to  this  conception  another 
of  equal  importance  for  our  study.  In  his  pamphlet  Economic 
Causes  of  Moral  Progress,1  he  attempts  to  show  that  at  least  many 
of  our  aesthetic  and  ethical  ideals,  relatively  permanent  and  pro 
gressive  sources  of  satisfaction,  are  but  qualitative  expressions, 
names  for  these  complementary  goods,  the  value  of  which  is 
imputed  to  the  economic  goods.  Thus  comfort,  cleanliness, 
thrift — he  might  have  added  taste  and  manners — are  qualitative 
terms  for  the  process  of  harmonising  goods  or  objects  of  sensuous 
feelings,  and  of  ejection  of  inharmonious  elements.  Even  the 
home  and  its  attendant  virtues,  the  state  with  its  justice,  are 
groups  of  such  values,  partly  economical,  partly  ethical.  If  we 
extend  the  conception  of  consumption  to  include  all  the  activities 
involved  in  acquisition  and  utilisation  of  goods,  we  may  accept 
this  account  of  the  origin  of  the  primitive  ethical  and  aesthetic 
objects.  The  importance  of  such  a  conception  lies  in  the  fact 
that,  if  these  qualitatively  different  worths  may  thus  be  related 
to  the  elementary  economic  values  as  complementary,  we  have 
a  means  of  co-ordinating  these  two  groups  of  values. 

But  it  must  be  noted,  in  the  second  place,  that  these  qualita 
tive  expressions  for  harmonious  consumption  and  activity, 
these  aesthetic  and  ethical  ideals,  are  new  objects  of  worth  feeling, 
ideal  objects,  founded  it  is  true  on  perceptual  activities,  but 
nevertheless  now  no  longer  the  objects  of  sensation  and  perception, 
but  of  judgment.  As  such  ideal  objects  of  judgment,  they  them 
selves  now  become  objects  of  new  worth  feelings  and  are  capable 
of  ideal  reconstruction  into  new  individual  unities  or  wholes. 
Thus  the  worths  of  the  personality,  the  ideals  of  goodness,  nobility, 
obligation,  inner  peace,  freedom,  perfection,  etc.,  are  qualitative 
terms  for  complementary  values  arising  from  the  harmonious 
co-ordination  of  these  fundamental  dispositions.  They  are 
objects  of  desire  and  worth  feeling  founded,  not  in  perceptual, 
but  in  ideational  activity  and  construction.  To  what  extent 
these  ideal  objects  may  become  relatively  permanent  and  pro 
gressive  sources  of  satisfaction  is  a  problem  to  which  we  must 
later  turn  our  attention. 

1  S.    N.    Patten,    Economic    Causes  of   Moral  Progress,   Annals  of  the  American 
Academy  of  Political  and  Social  Science,  Vol.  Ill,  Sep.  1892. 


I78  Valuation:   its  Nature  and  Laws 

3.    Psychological   Interpretation   of   the   Law   of    Complementary 

Values. 

The  extension  which  has  thus  been  given  to  the  economic 
concept    of    Complementary    Values-to    include    all    acquired 
affectfve-volitional   meanings   which    arise    m   the   P^ess   o 
cerceptual   and   ideal  reconstruction    of   individualised  whole; 
out    of    the    simple    elements    which   constitute    the   primary 
obiects  of  desire  and  feeling— opens  up  a  vista  of  mnumeral 
ob  ects  of  secondary  acquired  value,  and  raises  the  quest,, 
the  limits  of  such  ideal  construction  and  acquirement  of  meaning^ 
When  this  process  of  perceptual  and  ideal  construction  is  viewed 
in  this  larger  way,  as  a  form  of  conative  continuity,  it  is  seen 
that  these'constructions  are  all  in  the  service  of  continuity -rf 
valuation   and  constitute  readaptations  of  conative   tendency 
after  arrest      The  facts  of  this  conative  continuity  or  acquire 
m  nt  Tworth  are  adequately  enough  described  by  this  formula 
™l  Complementary  Values,  but  the  attempt  to  answer  the  question 
of  the  limits  of  the  process  requires  that  the  principle  1  e  properly 
interpreted. 

(a)  The  Objects  of  Complementary  Value:  Forms  of  Combination, 

Perceptual  and  Ideal— Founded  Objects. 
The  interpretation  ordinarily  given-that  the  complementary 
acquired  value  is  in  some  sense  a  sum  of  the  quantities  of  pleasure 
Tthe  elements,  although  that  product  is  recognised  as  being 
something  more  than  the  mere  sum  of  the  elements-fails  to 
take  account  of  the  really  significant  feature  of  the  process 
That  feature  is  to  be  found  in  the  fact  that  ttj.  acqu,red  value  has 
as  its  object,  not  the  elements,  but  a  new  object  founded  on  the  «- 
arrangement  of  the  elements.  Even  in  the  simple  case  of  th 

mbfnation  of  food-stuffs  and  the  order  of  their  consumption, 
th  object  of  the  acquired  worth  is  a  qualification,  a  meaning 
which  hough  it  may  be  conceptually  abstracted  from  the  pro- 
STcan  beSintuitively  realised  only  in  the  consumption  ofjhe 
elements  themselves.  When  we  consider  the  further 
pi  cTtton  o  these  food  elements  with  the  objects  of  other  senses- 
Delights  flowers,  and  music  of  the  banquet,  the  refinements 
"process  of  serving  and  eating,  the  psychical  character  of 
the  object  of  the  acquired  worth  becomes  more  evident,  f 


The  Laws  of  Valuation  179 

have  in  the  terms  "  taste "  and  "  manners  "  the  means  of 
separating  the  form  of  the  process  from  the  material  elements. 
The  psychical  objects,  cleanliness  and  thrift,  though  equally 
forms  of  combination  developed  in  the  processes  of  consumption, 
can  be  still  more  easily  and  completely  abstracted  from  the 
process,  as  objects  of  judgment  and  assumption,  and  may  be 
referred  in  judgments  of  possession  to  the  self,  the  subject  of  the 
process. 

This  conception  of  the  psychical  object  as  a  form  quality 
emerging  in  conative  process  may,  moreover,  be  extended  to  other 
fundamental  instinctive  activities,  connected  with  acquisition, 
the  hunt,  war,  labour,  etc.,  to  sex  and  parenthood,  and  finally  to 
perceptual  and  ideational  activities  involved  in  knowledge  and 
art.  The  range  of  possible  psychical  objects,  the  construction 
of  which  adds  complementary  value  to  the  objects  of  immediate 
sensation  and  perception  upon  which  they  are  founded,  is  con 
sequently  very  large.  We  may,  moreover,  again  recall  the  fact 
that  personality,  when  intrinsically  valued,  is  a  "  character," 
a  form  of  combination  of  certain  elements.  In  this  case  the 
elements  are  certain  qualities  or  dispositions  which  emerge  in  the 
process  of  ideal  construction  involved  in  sympathetic  Einjiihlung. 
Of  this  special  application  of  the  principle  we  shall  make  con 
siderable  use  later. 

(b)   Psychological  Laws  of  Complementary  Values. 

These  psychical  objects  may,  therefore,  be  founded  on  per 
ceptual  and  ideational  activities  alike ;  they  are  the  products  of 
individuating  activities,  both  perceptual  and  ideational.  Is  it 
possible  to  refer  this  general  law  of  Complementary  Value  to  more 
ultimate  psychological  principles,  as  in  the  case  of  the  other  two 
laws,  of  the  Threshold  and  of  Limiting  Value  ? 

In  approaching  this  question,  it  is  well  to  observe  the  fact 
that  we  are  here  concerned  with  the  formulation  of  general 
psychological  laws  of  acquirement  of  meaning  through  individua- 
tion  of  psychical  objects — with  general  laws,  therefore,  which  have 
specific  application  in  different  spheres  of  concrete  valuation. 
In  the  second  place,  we  must  distinguish  between  the  general 
laws  of  acquirement  of  meaning,  of  which  the  law  of  Comple 
mentary  Values  is  an  expression  for  worth  theory,  and  the 
particular  laws  of  combination  of  elements  by  which  the 
meaning  is  acquired. 


180  Valuation:   its  Nature  and  Laws 

The  Law  of  Complementary  Values  has  accordingly  two 
aspects.  From  the  point  of  view  of  conative  activity  there  is 
progress,  to  a  new  (psychical)  object.  From  the  point  of  view  of 
content  there  is  a  combination  of  elements  in  which  an  individu 
ated  whole  emerges  with  a  meaning  not  found  in  the  elements 
taken  separately.  Wundt  has  described  these  two  aspects  as  the 
law  of  Heterogeneity  of  Ends,  and  the  law  of  Creative  Resultants, 
or  more  simply,  Resultants.1 

The  principle  of  Resultants  is,  then,  a  law  of  combination  of 
elements  and  acquirement  of  meaning  through  that  combination 
and  reconstruction.  Every  resultant  is  an  individuated  whole, 
either  perceptual  or  ideal.  The  complementary  value  is  the 
affective-volitional  meaning  acquired  in  that  construction. 
What  are  the  psychological  factors  or  laws  involved  ? 

Wundt  formulates  but  one  such  law,  that  of  increase  through 
contrast.  Here  he  has  in  mind  the  increase  of  effect  for  feeling 
and  will  which  the  contrasting  elements  exercise  upon  each  other. 
But  while  the  principle  of  contrast,  when  viewed  in  all  its 
aspects,  may  be  said  to  comprehend  all  the  principles  of  intuitive 
construction,  nevertheless  for  our  purpose  it  is  necessary  to 
formulate  more  specific  laws  for  such  ordering  of  content. 

We  may  distinguish,  then,  three  specific  laws  of  intuitive 
combination  of  elements  into  total  resultants  :  (i)  the  law  of 
contrast  in  its  narrower  sense ;  (2)  the  law  of  the  total  series  (the 
principle  formulated  by  Fechner  for  aesthetics  as  the  Folgegesetz) ; 
and  (3)  the  law  of  end  feelings  (which  Fechner  formulated  in  its 
aesthetic  application  as  the  Versdhnungsgesetz).  The  meaning  of 
these  laws  may  be  stated  very  briefly,  for  their  fuller  develop 
ment  will  be  required  at  various  points  where  they  are  applied. 

The  principle  of  simple  contrast  merely  states  that  an  object 
of  desire  and  feeling  when  contrasted  with  its  opposite,  or  when  its 
existence  is  contrasted  with  its  non-existence,  gets  an  imputed 
value  which,  by  itself,  it  has  not  either  intrinsically  or  instru- 
mentally.  The  law  of  the  "  total  series  "  is  a  formulation  of  the 

1  Wundt,  Physiologische  Psychologic,  Fifth  Edition,  Leipzig,  1903,  Vol.  Ill, 
chap,  xxn  ;  also  Kreibig,  Psychologische  Grundlcgung  eines  Systems  der  Wert- 
theorie,  Wien,  1902,  pp.  59-63,  for  a  similar  study.  Wundt  describes  them  as 
laws  of  psychical  causality  in  distinction  from  physical  causality,  pointing  out  that, 
whereas  in  physical  causality  there  is  a  quantitative  equivalence  between  antecedent 
and  consequent,  in  psychical  process  the  resultant  always  shows  an  increase  or  acquire 
ment  of  meaning.  Calling  this  acquirement  of  affective-volitional  meaning  "increase 
of  energy,"  he  finds  in  this  contrast — between  equivalence  of  energy  in  the  physical 
sphere  and  increase  of  psychical  energy — the  fundamental  difference  between  physical 
and  the  psychical  causality.  While  making  use  of  his  descriptive  formulae  we  need 
not,  and  shall  not,  for  reasons  developed  in  the  introductory  chapter  and  elsewhere 
in  our  discussion,  conceive  them  in  terms  of  causality. 


The  Lazvs  of  Valuation  181 

fact  that  the  ordering  of  objects  of  desire  and  feelings  in  a  graded 
series,  or  with  certain  relations  of  contrast  and  repetition,  as  for 
instance  in  rhythm,  gives  rise  to  an  imputed  value  of  the  whole 
which  is  not  a  sum  of  the  value  of  the  separate  elements.  The 
law  of  "  end  feelings  "  recognises  the  fact  that  the  worth  of  a 
series  of  elements  is  determined  by  the  final  moment  of  the 
series  and  its  relation  to  the  preceding  moments.  In  all  these 
specific  forms  of  arrangement  of  elements  form-qualities  are 
created  which  constitute  the  real  objects  of  the  imputed  value. 

VI.  EXTENT  OF  APPLICATION  OF  THE  LAW  OF  LIMITING 
VALUE  TO  IDEAL  OBJECTS  OF  INTRINSIC  VALUE 

We  are  now  in  a  position  to  return  to  the  problem  with  which 
this  discussion  of  the  founded  ideal  objects  of  valuation  took  its 
rise — namely  the  extent  of  the  application  of  the  law  of  Limiting 
Value  to  the  intrinsic  valuation  of  such  objects.  The  im 
portance  of  the  question  is  far-reaching.  In  the  succeeding 
chapters  we  shall  have  occasion  to  consider  the  more  specific 
problem  of  the  laws  governing  the  actual  concrete  judgments  of 
the  worth  of  ideal  objects  of  aesthetics  and  ethics.  Here  we  are 
concerned  merely  with  the  general  theoretical  question,  whether 
the  nature  of  these  objects,  and  of  their  corresponding  worth 
feelings,  is  such  as  to  make  this  general  law  applicable.  Upon  this 
question,  as  we  have  seen,  there  is  a  significant  difference  of 
opinion.  Kreibig  and  Meinong  deny  its  application,  while 
Ehrenfels  affirms  it.  The  problem  is  in  the  first  place  obviously 
a  question  of  fact.  Are  there  any  objects  of  desire  and  feeling 
which  have  the  power  of  calling  forth  continuous  intrinsic  valua 
tion  ?  It  is,  in  the  second  place,  however,  a  question  of  psycho 
logical  analysis  of  what  is  involved  in  the  act  of  valuation.  Thus 
an  analysis  which  recognises,  with  Brentano,  intensity-less  acts  of 
valuation  would  be  in  a  position  to  infer  the  existence  of  such 
objects,  for  the  laws  of  dulling  of  sensitivity  and  satiety  apply 
only  to  hedonic  intensities.  In  general,  any  theory  which  makes 
the  fundamental  distinction  between  pleasure-causation  and 
worth  feelings,  between  sensation-feelings  and  feeling  actualised 
by  judgments  and  assumptions,  would  leave  the  question  open. 

If  then  we  start  with  the  immediately  given  facts  of  experi 
ence,  with  our  appreciation  of  those  perceptual  and  ideal  objects 
which  emerge  in  our  instinctive  conative  activities  and  in  our 
ideal  constructions,  we  are  at  first  sight  disposed  to  admit  the 


1 82  Valuation:    its  Nature  and  Laws 

working  of  the  law  of  Limiting  Value  in  this  sphere  as  well. 
The  complementary  values  which  emerge  in  harmonious  con 
sumption  modify,  it  is  true,  the  effects  of  habit  and  satiety, 
and  extend  the  limits  of  valuation,  but  ultimately  these  limits  are 
reached.  The  acquirement  upon  the  part  of  a  group  of  com 
plementary  objects  of  sense  stimulation,  of  meanings  such  as 
sensuous  beauty,  taste  in  eating,  etc.,  removes  the  object  from 
the  immediate  gross  satisfaction  of  desire,  and  postpones  the 
dulling  of  sensitivity  and  satiety  ;  but  since  these  meanings  are 
founded  on  some  stimulations  and  are  conditioned  by  them,  they 
are  ultimately  subject  to  the  law  inherent  in  all  feelings  of  sense. 
Sensuous  beauty  palls,  refinement  of  living  becomes  wearisome, 
and  we  seek  refreshment  in  crude  contrasts  of  ugliness  and 
coarseness.  There  is  a  return  value-movement  from  the  acquired 
to  the  fundamental. 

The  same  nemesis  seems  to  follow  the  so-called  ideal  feelings, 
the  valuation  of  those  ideal  objects  the  worth  of  which  is  deter 
mined  by  emotion  and  sentiment,  the  ideal  objects  of  know 
ledge,  art,  and  morals.  With  the  loss  of  novelty,  with  each 
successive  act  of  appreciation,  there  is  diminution  of  enthusiasm, 
loss  of  the  emotional  uplift  which  constitutes  the  peculiar  glory 
of  the  first  realisations.  James  has  described  this  process  in  his 
usual  picturesque  and  impressive  way  in  his  chapter  on  emotions, 
and  it  is  a  phenomenon  everywhere  present  in  our  appreciation  of 
persons  and  objects.  Friendship  begins  with  full  and  resonant 
emotions,  love  with  an  ecstatic  devotion,  to  become  with  use 
habits  with  scarcely  perceptible  affective  intensity.  Perhaps 
the  most  striking  illustration  of  all  is  the  effect  of  repetition 
upon  sympathetic  participation  (Einfuhlung).  With  each  act  of 
participation,  the  judgment  of  the  existence  of  the  ideal  object, 
the  disposition  in  the  alter,  acquires  more  certainty,  the  character 
is  more  fully  realised,  but  the  realisation  has  become  habit,  the 
emotional  resonances  are  deadened.  It  would  seem,  then,  that 
those  ideal  objects  the  value  of  which  lies,  not  in  their  being 
sensed,  but  in  their  being  judged  or  assumed  to  exist  or  not  exist, 
also  lose  their  capacity  for  calling  out  intense  emotions.  If, 
then,  we  should  equate  the  worth  or  funded  meaning  of  these 
objects  with  this  capacity,  there  would  be  no  question  that  the 
intrinsic  valuation  of  such  objects  is  subject  to  the  law  of  Limit 
ing  Value.1 

1  And,  doubtless,  in  some  moods  of  retrospection  the  mind  turns  from  the    unded 
meaning  of  the  object  to  the  emotional  accompaniments  of  the  earlier  appreciations. 


The  Laws  of  Valuation  183 

Apparently,  then,  we  may  conclude  that  the  repeated  ex 
perience  of  these  ideal  emotions  and  sentiments  is  accompanied 
by  the  same  dulling  of  sensitivity  for  intensity  of  affective  dis 
turbance  which  comes  with  repetition  and  habit  in  the  case  of 
repeated  sensation-feelings,  and,  in  those  experiences  where  the 
intensity  of  the  emotion  is  unusual,  by  the  phenomena  of  satiety. 
We  might  be  led,  therefore,  to  conclude  further  that  the  intrinsic 
valuation  of  ideal  objects  is  subject  to  the  law  of  Limiting  Value. 

On  the  other  hand,  with  equal  truth  it  may  be  said  that  these 
experiences  do  not  modify  our  feeling  of  the  worth  of  the  objects, 
do  not  determine  our  worth  judgments,  as  habit  and  satiety 
have  been  seen  to  modify  our  judgments  in  the  case  of  objects  of 
sensation-feeling.  The  worth  of  the  object  seems  to  be  deter 
mined  by  other  modifications  of  consciousness  than  this  feeling 
intensity,  and  to  persist  throughout  these  changes  in  hedonic 
accompaniments,  as  relatively  permanent  and  progressive  sources 
of  satisfaction.  This  analysis  of  the  facts  of  our  experience  would 
therefore  indicate  that  in  such  processes  we  have  a  "  mixture  of 
phenomena  "  in  which  the  worth  of  the  object  and  its  capacity 
for  calling  out  feeling  intensities  are  independently  variable. 
In  other  words,  as  we  have  previously  insisted,  habit  has  its  own 
worth  feelings,  its  own  affective-volitional  meaning. 

This  apparent  antinomy  has  been  clearly  expressed  in  a 
passage  in  Mimsterberg's  discussion  of  the  relation  of  worth 
experience  to  psychology.  To  a  remark  of  Windelband,  that 
"  all  interest  and  appreciation,  all  valuation  on  the  part  of 
men,  have  reference  to  individual  and  particular  appearances," 
Miinsterberg  answers  :  "  If  by  value  we  mean  the  psychological 
feeling-process  which  is  called  forth  in  objective  causal  happen 
ing,  such  sense  of  value  appertains  to  repeated  no  less  than  to 
single  reactions.  Of  course  there  are  acts  the  attraction  of  which 
lies  in  their  novelty  and  singularity,  but  all  that  is  good  and  noble 
gains  depth  of  feeling- tone  through  repetition  just  as  displeasure 
with  the  low  and  coarse  grows  through  repetition."  x  With  the 
truth  of  Miinsterberg's  observation  we  may,  on  the  basis  of  our 

The  lover  looks  back  with  regret  upon  his  ecstasy,  the  religious  devote  cries  out  in  the 
words  of  the  hymn  :  "  Where  is  the  blessedness  I  knew  when  first  I  saw  the  Lord  !" 
And  not  rarely  the  artist  and  enjoyer  of  works  of  art  alike  long  for  the  emotional 
accompaniment  of  their  earlier  appreciations,  and,  at  times,  put  more  value  upon  them 
than  upon  their  realised  ideals  and  their  attained  insights.  Nevertheless  we  rightly 
look  upon  these  backward  movements  as  temporary  phenomena,  in  what  is  still  a  con 
tinuous  valuation  of  the  ideal  object.  The  lover  still  loves  and  believes  and  appreciates. 
The  ideal  objects  remain  relatively  permanent  and  progressive  sources  of  satisfaction. 
1  Miinsterberg,  Grundziige  der  Psychologic,  pp.  39  and  40. 


1 84  Valuation  :   its  Nature  and  Laws 

own  analysis,  readily  agree.  That  the  increase  of  depth  of  feeling- 
tone,  with  repetition,  is  an  empirical  law  of  some  types  of  worth 
feeling  no  one  could  deny,  unless  under  the  influence  of  the 
prejudice  that  the  laws  of  habit  and  satiety,  which  apply  to  sensa 
tion  feelings,  apply  to  all  feelings.  But  the  recognition  of  this 
empirical  law  necessitates  logically,  if  we  are  to  explain  it,  a 
distinction  between  the  attributes,  depth  of  feeling-tone  and 
intensity.  Clearly  depth  of  feeling-tone,  which  is  here  equated 
with  degree  of  worth,  cannot  be  equated  with  degree  of  intensity 
of  affective  disturbance.  Depth  of  feeling  must  be  taken  as 
a  description  of  a  modification  of  a  special  aspect  of  feeling, 
other  than  the  affective  tone  of  sensations  and  presentations, 
for  the  intensity  of  such  hedonic  tone  is  also  subject  to  dulling 
of  sensitivity  with  repetition. 

The  question,  in  so  far  as  it  is  one  of  psychology  merely, 
seems  to  depend  for  its  solution  upon  the  answer  to  two  pre 
liminary  questions — namely  :  what  is  the  nature  of  these  ideal 
objects,  and  what  is  the  nature  of  the  psychical  acts  of  which 
they  are  the  objects  ?  For  the  modifications  of  feeling  are,  as 
we  have  seen,  determined  by  the  character  of  the  presuppositions 
of  the  feeling. 

As  to  the  nature  of  the  objects  of  this  intrinsic  valuation, 
it  has  already  been  shown  that  they  are  ideal  and  "  founded," 
and  that,  while  the  ideal  or  conceptual  objects  of  utility-judg 
ments  are  subject  to  the  limitations  inherent  in  this  type  of 
construction,  there  is,  on  the  other  hand,  in  the  case  of  the  in 
dividuating  type  of  construction  which  creates  these  objects  of  in 
trinsic  worth,  no  limit — at  least  none  that  can  be  determined 
a  priori — to  the  acquirement  of  meaning.  The  laws  of  this 
individuating  construction,  already  analysed,  indicate  this  fact 
clearly. 

When  we  turn  to  the  feeling  side  of  this  process  of  intrinsic 
valuation  of  ideal  objects,  and  examine  the  presuppositions  of 
these  feelings,  we  find  that  the  cognitive  acts  presupposed  are 
judgments  and  assumptions  (explicit  and  implicit) ;  such  ideal 
objects  are  not  objects  of  sensation  and  presentation  at  all. 
Now  the  detailed  analysis  of  feelings  conditioned  by  such  acts, 
has  shown  us  that  the  laws  of  dulling  of  sensitivity  and  satiety, 
upon  which  the  principle  of  Limiting  Value  is  based,  do  not  apply 
to  these  feelings,  but  only  to  the  hedonic  redundancies  of  the 
acts.  Otherwise  expressed,  the  affective  abstract,  or  "  affective 
sign  "  (an  assumption- feeling  representing  judgmental  habit) 


The  Laws  of  Valuation  185 

is  not  subject  to  these  laws,  but  rather  increases  in  depth  of 
feeling-tone  with  repetition.  The  assumption-feeling  may,  as 
we  have  seen,  attach  itself  directly  to  judgment  or  to  a  word 
which  represents  a  general  concept,  and,  as  thus  representing  the 
acquired  meaning  of  earlier  worth  judgments,  it  constitutes  the 
assured  presuppositions  of  the  new  judgment-feelings.  Each 
successive  judgment  of  value  embodies  the  assumption-feeling  or 
"  habit-meaning "  of  preceding  judgments,  and  thus  makes 
possible  the  increase  of  "  depth  of  feeling-tone,"  of  sentiment. 

This  increase  of  depth  of  feeling  does  not,  however, 
exclude  the  phenomena  of  dulling  of  sensitivity  and  satiety  as 
applying  to  the  "  hedonic  redundancies  "  of  these  successive 
appreciative  acts.  The  causes  of  these  redundancies,  accompany 
ing  acts  of  intrinsic  appreciation  of  ideal  objects  are,  as  Brentano 
has  suggested,  too  complex  to  admit  of  the  formulation  of  specific 
laws  of  their  modification.  In  principle,  however,  the  laws  of 
dulling  of  sensitivity  and  satiety  are  applicable  to  these,  as  to 
all  sensation- feelings.  When  we  say,  therefore,  that  the  acquire 
ment  of  intrinsic  value  on  the  part  of  these  ideal  objects,  the 
permanence  of  these  ideal  sentiments,  is  unaffected  by  these 
laws,  we  mean  to  say  merely  that  the  modification  of  the  intensity 
of  the  accompanying  sensation-feelings  becomes  irrelevant 
for  worth  judgment.  We  have  already,  in  a  number  of  connec 
tions,  seen  the  independent  variability  of  these  two  aspects  of  a 
total  feeling  attitude,  more  particularly  in  the  case  of  the  effect 
of  repeated  appreciations  of  a  work  of  art,  already  analysed, 
and  in  the  distinction  here  made  we  have  the  solution  of  the 
apparent  antinomy  in  intrinsic  valuation. 

VII.  GENERAL  CONCLUSIONS — INFERENCES  FROM  THIS  STUDY 
OF  THE  LAWS  OF  VALUATION  FOR  DIFFERENT  TYPES  OF 
OBJECTS — THE  PROBLEM  OF  THE  LIMITS  OF  ACQUIREMENT 
OF  VALUE 

The  preceding  study  of  the  law  of  Limiting  Value,  and  of  its 
application  to  feelings  directed  toward  different  types  of  objects, 
has  been  purely  psychological  in  character,  in  that  it  was 
concerned  with  the  application  of  the  laws  of  dulling  of  sen 
sitivity  and  of  satiety  to  feelings  with  different  presuppositions. 
It  can  easily  be  seen  that  in  this  principle  we  have  the  means 
of  defining  different  worth  objects  in  terms  of  their  capacity 
for  continuous  valuation,  and  thus  for  determining  the  relative 


1 86  Valuation:   its  Nature  and  Laws 

oreferability,  not  only  of  objects  of  the  same  kind  but  of 
different  groups  or  kinds.  Such  use  of  these  stud.es  we 
TaU  make  ir?  the  succeeding  chapters,  which  treat  of  the 
different  worth  objects  distinguished  in  the  first  chapter  (objects 
of  condition  worth,  personal  worth,  and  impersonal  or  over, 
individual  worth) ;  but  before  turning  to  these  subjects,  and  by 
way  of  concluding  this  discussion,  we  have  still  to  consider  a 
croblem  of  very  general  theoretical  bearings,  the  problem, 
namely,  whether  the  principle  of  Limitation  of  Capactty  (capacity 
of  appropriation,  as  developed  for  consumption  originally)  can 
be  applied  to  valuation  in  all  its  aspects. 

It  was  to  this  general  principle,  it  will  be  remembered,  that 
the  economic  theory  of  value  looked  for  its  foundation :    tc  it 
all  other  laws  were  referred.    That  theory,  we  saw,  is  base 
partly  upon  analysis  of  a  very  limited  sphere  of  experience, 
partly  upon  unwarranted  assumptions  of  an  a  pnon  character 
Because  the  dispositions  underlying  "  pleasure-causation 
linSed  in  capacity,  it  is  assumed  that  all  worth  dispositions, 
Sings    of    value    being    identified   with    pleasure  -  causation 
are   subject   to  the  same  limitation.     Now  that   an   emp.nca 
^dy  of  a  larger  range  of  worth  feelings,  and  thei, •  psychica 
conditions,  has  shown  the  capacity  of  some  ideal  object 
continuous  valuation  to  extend  far,  perhaps '  ""^'te^  beyond 
the   limits   conceived  by  the   theory  we  have   been   , 
examining,  we  may  turn  our  attention  to  this  a  pnon  ^unrpt^ 

The  assumption  is,  of  course,  that  for  every  ideal 
object  which  appreciation  may  distinguish,  and  for  the  w. 
feeling  directed  toward  it,  there  is  a  corresponding  physiologica 
disposition;  that  this  disposition  is  part  of  a  system  the  tota 
energy  of  which  is  a  constant  and  limited  amount  .that  wit 
this  system  there  may  be  redistribution,  but  never  actual  increase 
of  energy '  and  that  the  increase  of  the  energy  of  any  deposition 
?s  therefore  strictly  limited  by  the  mutual  relations  of  the e emen  » 
of  the  system.     This  conception  is  then  simply  earned 
into  the  psychical  sphere  and  from  it  are  deduced  limits  f 
nsvrhical  energy  and  the  acquirement  of  meaning. 
PSyCSuch   ^Sumption   clearly  underlies   *~""*^ 
when  he  says 1  that  "  the  ethical  dispositions,  just  as  litt 
Others    however  insignificant,  possess  the  power  of  extending 
^^Tbeyond  tleir  given  limits,  that  in  then ^tendency 
so  to  do  they  are  held  fast  in  their  former  relations  by  t 

i  System  dcr  Werththeorit,  Vol.  II,  p.  217  note. 


The  Laws  of  Valuation.  jg, 

tendency  inherent  in  all  other  dispositions  to  maintain  their 
former  force.'  And  the  same  assumption  is  present  when  he 
maintains  (m  criticism  of  Kruger's  formulation  of  the  suprem 

cTm6;  ,  f^  V3jUe'  ^  the  desire  for  the  ^crease  of  the 
capacity  of  valuation  itself)  that,  »  the  claim  that  every  subject 

£t±Sf,mUSt  PK?SeSS,the  diSp°Sition  to  value  the^  valuing 
activity  tself  is  capable  neither  of  a  priori  nor  of  empirical  proof  "* 

Obviously  it  ,s  assumed  that  every  ideal  object  constructed  has 
as  ,ts  correlate,  a  definite  physiological  disposition,  and  that  S 
Place  m  a  system  of  values  is  but  a  reflex  of  the  place  of  the 
disposition  in  a  system  of  energies 

More  specifically,  the  question  is  raised  whether  it  is  possible 
for  us  rationally  to  wish  or  strive  for  increase  of  our  wo?" 
positions  m  any  direction.  Can  a  man,  for  instance,  on  the  bas  s 
of  an  experience  of  the  sentiment  of  benevolence,  rat  onally  deste 


vole  ' 

Ehrenfels  s  first  answer  is  unqualifiedly  negative  but 

m  reply  to  Hoffler's  criticism  of  the  same  '  (and  this  fea    thatl 
necessary  for  our  purpose),  he  concludes  that  it  would  be  rational 
or  a  man,  let  us  say  in  a  state  of  exaltation  of  benevolent  emotfe 
leading  to  great  sacrifice,  to  desire  intrinsically,  as      per™ 
possess™,  the  «*,  capacity  for  benevolent    eeling^sTe  ex 


that  IS'  more  than 

for  the  reason  that  his  desire  is  not  for 
something  which  transcends  the  known  measure  of  the  capacity 
the  disposition  mvolved.     But  to  desire  an  increase  in  the 
disposition  itself  would  be  irrational,  for  it  wouW  TmTolve 

^  S6t  WhlCh  °ther  ^Positions  would  receve 


Obviously  the   problem,  as   here   formulated    is  concerned 
trintal  ValUe  °f  the  d-Po-t10n  for  s™ai 


r  f  qU6Stl0n'  but  merejy  with  the  possibility 

A     \     r    y          hlS  lncrease  from  the  point  of  view  of  I 


out  of  account  the  secondary  question 

— ^,IL^S,  ^jntem  aer  Werttith eori e 'v ol  'll   nn    TTC    i*f.      t       u-rr 
p.  599.  "i  PP-  I/5>  »7«J  a'so  Hoffler,  /•'jyv 


tem  d'r  Mrthtteont,  Vol.  II,  p.  171 


1 88  Valuation:   its  Nature  and  Laws 

of  the   nature  of  that  measure  (which  would  have  to  be  ex 
pressed  not  only  in  terms  of  intensity,  but  also  of  extension, 
that  is  of  its  capacity  for  expression  in  all  the  acts  of  life,  whereas 
Ehrenfels   takes  into  account   only  the  strength  or  intensity), 
we  may  go  directly  to  the  root  of  the  problem.     The  rationality 
of  the  desire  for  increase  of  disposition  must  be  determine 
wholly  by  the  question  of  experience,  whether  such  an  ideal  is  an 
object   with   capacity   of   continuous   valuation,   whether  with 
repetition  or  increase  of  amount,  the  worth  feeling  increase 
accordingly.     The  criterion  being  one  of  intnnsic  and  internal 
meaning,  the  only  way  in  which  the  possibility  and  desirability 
of    the  increase  of    a  disposition  can  be  known  is  by  willing 
it    by  judging  or  assuming  the  object  to   exist    and    mfe 
the  truth  of   such  judgment  or  assumption   from  the  wortl 
experiences    which    result.      The    criterion    is    wholly     mtr 
experiential.     But  for  Ehrenfels  the  real  test  of  the  rationality 
of  the  desire  is  not  an  intra-experiential  test,  not  a  criterion 
formed  in  the  inductive  analysis  of  the  different  types  of  worth 
feelings  themselves.     It  is  inferred  deductively  from  a  purely 
hypothetical  and  abstract  conception  which  is  extra-psychological. 
It  rests  upon  the  assumption  of  specific  nervous  dispositions 
and  of  the  limitation  of  the  energies  of  these  dispositions. 

This  it  seems  to  me,  is  an  essentially  false  way  of  stating  the 
problem      The  "  rationality  "  of  any  valuation  is  essentially  an 
axiological  problem,  and  as  I  hope  to  show  in  the  concluding 
chapter,  the  only  possible  axiological  criterion  is  an  internal 
one    immanent  in  the  process  of  valuation  itself, 
point  of  view,  the  only  sense  in  which  the  question  of     rational 
ity  "    has    any    meaning    is    that    expressed    by    the    purely 
practical  question-can    I   continue   to    desire    and    will    any 
object    or   its   increase,   without   internal   contradictions,   con 
tradiction  in  this  sphere  meaning,  of  course,  not  contradiction 
of  judgment   with   judgment,  but    of   feeling   with    feeling   • 
of  feeling  with  will  ?     In  the  concrete  case  before  us,  i 
actually  desire  increase  of  a  given  sentiment,  and  in  that  ex 
perience  there  is  nothing  paradoxical,  as  there  is  in  the  case  of  the 
desire  for  indefinite  increase  of  pleasure,  but  with  each  judgment 
of  increase  of  the  sentiment  there  is  an  increase  of  worth  feeling 
then  such  a  desire  is  rational  in  the  only  sense  in  which  t 
word  has  meaning  in  this  connection. 

The  true  way  of  stating  Ehrenfels's  problem  is  rather  this- 
does  such  a  desire  for  indefinite  increase  of  an  ideal  object, 


The  Laws  of  Valuation 


£' 

5  *' 


=»= 
b«™  ,„  i.,«P,s,,,  ts-  tht  j"*"™'  "•  *- 


logical    analysis    of    the    dffierent    t  ,  that    a   PSych°- 

and  the  formulation  of  the  lawtof  ^T  r  T"'11  fedin«s' 
analysis,  give  us  the  sr'  ntifi  valuation  based  upon  this 

pret  the  concrete  facts  of  I  T^3  ™th  which  to  inter- 
disclose.  The  iefulness  o  Ih  °  WWch  °Ur  analysis  «» 

ing  studies  shou  d  not  be  ^LT*?*0"*  *"  °Ur  SUCCeed- 
introduction  of  concep  ions  f  Z  I  7  preJudiced  ^  the 
irrelevant.  °m  a  sPhere  whi<*  is  essentially 


CHAPTER   VII 

I.    VALUES    OF    SIMPLE    APPRECIATION — THEIR    ORIGIN    AND 

NATURE 

i.  Objects  of  "  Condition  Worth,"  Primary  and  Derived 

A   PRELIMINARY   classification   of    worth   objects   distinguished 
three  general  groups — condition   worths,  or  worths   of  simple 
appreciation,   personal  worths,   and   over-individual   or   imper 
sonal  worths.     To   these  corresponded  the  three  fundamental 
types  of  activity  of  valuation,  simple  appreciation,  characterisa 
tion,  and  participation,  the  latter  including  utilisation  of  ob 
jects.     On  each  of  these  levels  of  valuation  the  worth  of  the 
object  is  the  funded   affective-volitional   meaning  acquired   in 
antecedent  psychical  processes,  and  the  different  qualifications 
in  worth  feeling  on  these  levels,  distinguishable  for  appreciation, 
go  back  to  differences  in  the  processes  by  which  these  worths 
are    funded.     Condition    worths    of    simple    appreciation    are 
determined    by    feelings    of    the    individual   which    presuppose 
merely  presumptions,  judgments,   or  assumptions  of  existence 
or  non-existence  of  objects  immediately  or  remotely  desirable, 
that    is,    objects    which    correspond    to    conative    dispositions. 
They  are  called  condition  worths  because  the  feelings  aroused 
are,  when  abstracted  from  the  object  and  viewed  retrospectively, 
referred  not  to  the  idea  of  the  self,  but  to  the  affective  condition 
of  the  organism.     Personal  worths  are  determined  by  feelings 
which  presuppose  processes  of  sympathetic  Einfuhlung  and  ideal 
construction   of  dispositions   on   the  basis  of  that  experience, 
while  over-individual  worths  are  determined  by  feelings  which 
have    as   their   presuppositions   still   wider  processes  of   social 
participation,  and  consequent  ideal  construction. 

The  three  levels  thus  distinguished  represent  relatively 
distinct  stages  in  valuation,  and  are  related  to  each  other  in 
such  a  manner  that  the  values  of  the  lower  level  are  implicitly 
presupposed  on  the  next  higher  level.  Personal  and  social 

190 


Values  of  Simple  Appreciation  191 

values,  with  their  qualitatively  distinct  feelings  and  differ 
entiated  objects,  emerge  first  in  the  processes  of  simple  appre 
ciation,  being  built  upon  and  developed  out  of  objects  of  "  con 
dition"  worth.  This  general  process,  in  which  the  object  of 
explicit  judgment  and  feeling  becomes  the  object  of  implicit 
assumption,  and  gives  place  to  a  new  object  on  a  higher 
level,  we  may  describe  as  a  progression  or  Value  Movement. 
The  general  problem  that  confronts  us  in  our  succeeding  studies 
is,  accordingly,  the  description  of  the  processes  involved  in  these 
progressions,  the  discovery  of  their  conditions  and  laws,  and 
the  reduction  of  their  empirical  uniformities,  in  so  far  as  possible, 
to  the  general  laws  of  conation  and  feeling  described  as  the  laws 
of  Valuation. 

While  this  is  the  general  problem — the  genetic  background 
of  our  picture,  the  immediate  foreground  is  taken  up  by 
the  values  and  objects  of  condition  worth.  These  objects 
are  the  primary  worth  objects,  and  their  values  the  primary 
values.  From  these,  as  has  been  said,  the  personal  and  social  or 
over-individual  values  are  derived.  But  within  the  region  of 
condition  worths  themselves,  there  are  certain  phenomena  and 
laws  of  valuation  which  require  study  for  their  own  sake.  To 
this  problem  the  present  chapter  is  devoted,  and  it  may  be 
described  as  the  application  of  the  general  laws  of  Valuation  to 
the  specific  class  of  condition  worths. 

Objects  of  condition  worth  are  of  varied  character,  and 
may  for  convenience  be  divided  into  physical  and  psychical. 
To  the  physical  objects  belong  primarily  the  so-called  economic 
goods,  desire  for  which  is  conditioned  by  certain  primary  sense- 
tendencies,  the  stimulation  or  gratification  of  the  same  being 
described  as  consumption.  To  these  must  be  added  other 
objects  of  sense  tendency  which  for  various  reasons  do  not  enter 
directly  into  economic  calculations,  although  they  must  of 
necessity  indirectly,  such  as  the  satisfaction  of  fundamental 
sex  and  gregarious  instincts.  Psychical  objects  of  condition 
worth  are  the  qualities  of  physical  objects,  which,  arising  in 
the  processes  of  consumption,  pursuit,  acquisition,  become  the 
objects  of  new  appreciations,  and  either  add  complementary 
worth  to  the  primary  objects,  or,  when  made  the  objects  of 
judgment,  call  out  new  worth  feelings. 

This  distinction  between  primary  and  derived  objects  and 
values  of  simple  appreciation  constitutes  the  starting-point 
for  the  detailed  studv  of  values  of  condition  and  their  laws. 


192  Valuation:   its  Nature  and  Laws 

It  is  desirable,  therefore,  to  develop  the  distinction  more  fully. 
By  a  primary  value  (or  Stammwert,  as  Ehrenfels  calls  it)  is 
understood  any  object  which  serves  to  satisfy  immediately 
any  fundamental  instinctive  sense-tendency.  The  mere  pre 
sumptions  or  judgment  of  existence  or  non-existence  of  such 
objects  is  followed  by  a  feeling  of  intrinsic  value.  By  funda 
mental,  in  this  sense,  we  understand  a  relatively  constant  in 
stinctive  tendency,  with  its  corresponding  passional  or  emotional 
psychosis  (or  instinct  feeling),  which  may,  according  to  the 
cognitive  acts  through  which  it  is  actualised,  be  predominantly 
desire  or  feeling.1 

It  is  not  necessary  for  our  purposes  that  we  should  determine 
just  what  objects  and  values  are  primary  and  what  secondary. 
For  some  purposes  of  ethical  and  social  science  such  a  classi 
fication  is,  it  is  true,  necessary — and  many  attempts  have 
been  made  in  this  direction,  notably  the  recent  efforts  of 
Ehrenfels  and  Schwartz,  but  for  the  special  ends  of  the  present 
study,  it  is  quite  sufficient  to  recognise  that  there  are  certain 
fundamental  conative  tendencies,  such  as  hunger,  sex,  ex 
pression  of  bodily  energy,  etc.,  the  satisfaction  of  which  gives 
immediate  and  unconditional  "  condition  "  worth,  that  for  any 
given  individual  they  are  primary  and  original,  and  that  the 
simplest  object  corresponding  to  these  tendencies  are  therefore 
primary  objects  of  simple  appreciation. 

Much  more  important  is  the  derivation  of  the  secondary 
from  the  primary.  These  secondary  or  derived  values  emerge, 
it  was  said,  in  the  processes  of  pursuit,  acquisition,  and  con 
sumption  of  the  primary  objects.  As  complementary  values, 
they  are  first  imputed,  as  additional  values,  to  the  primary 
objects ;  but  they  may  ultimately  be  abstracted  from  the  primary 
objects,  ideally  reconstructed  and  independently  valued,  when 
they  become  the  objects  of  personal  and  social  values. 

In  the  preceding  chapter  we  have  seen  in  a  general  way 
how  these  complementary  values  emerge  in  the  processes  of 
valuation.  On  the  side  of  the  object  they  are  the  resultants  of 

1  The  term  fundamental  carries  with  it  no  implications  as  to  higher  objective  worth 
or  underived  character.  As  Schwartz  in  his  criticism  of  Nietzsche's  worth  theory 
rightly  points  out,  there  are  no  conative  tendencies  or  dispositions  which  by  reason 
of  their  originality  carry  with  them  implications  as  to  their  primary,  least  of  all  ex 
clusive,  worth.  At  the  same  time,  when  we  view  the  history  of  the  individual  as 
well  as  that  of  the  race,  we  find  that  certain  instincts  and  conative  tendencies  to 
which  the  fundamental  passions  correspond  do,  in  a  certain  sense,  act  as  poles  about 
which  others  concentrate.  The  individual  at  birth  may  from  a  psychological  point 
of  view  be  looked  upon  as  a  group  of  more  or  less  loosely  co-ordinated  conative 
tendencies,  impulses,  and  instincts. 


Values  of  Simple  Appreciation  193 

perceptual  and  ideal  reconstruction ;  on  the  side  of  conative 
tendency  they  involve  readaptations  of  attitude  after  arrest. 
Continuous  valuation  of  any  isolated  physical  good,  satisfaction 
of  any  single  tendency,  is  followed  by  dulling  of  sensitivity 
and  satiety,  and  the  reconstruction  of  the  object,  together 
with  modifications  of  the  attitude,  constitutes  a  value  move 
ment  in  which  the  continuity  of  valuation  is  maintained. 
Our  present  task  is  to  study  these  value  movements  in  more 
detail,  to  discover  whether  these  transformations  of  attitude  and 
reconstruction  of  objects  cannot  be  reduced  to  general  types, 
and  their  conditions  and  laws  determined.  To  this  end  we  shall 
seek,  first,  to  classify  and  analyse  the  general  phenomena  of  value 
movement,  and  then  to  distinguish  those  which  are  characteristic 
of  sirnple  appreciation  —  those  by  which  primary  condition 
worths  acquire  complementary  value — from  the  more  complete 
developments  from  one  level  of  valuation  to  another.  From 
this  understanding  of  value  movements  and  their  relation  to 
the  general  laws  of  valuation,  we  shall  turn  to  the  study  of  two 
special  forms  of  psychical  development,  in  which  objects  of  con 
dition  worth  acquire  ethical  and  aesthetic  values,  and  shall  seek 
to  show  how  these  acquired  complementary  values  modify  our 
simple  economic  judgments. 

Our  general  problem  may  be  made  more  definite  by  con 
necting  it  with  the  results  of  the  preceding  chapter.  The  law 
of  the  Threshold,  the  law  of  Limiting  Value,  and  the  law  of 
Complementary  Values  are,  we  found,  the  fundamental  laws 
of  worth-process  in  general.  We  have  seen  how  both  of  these 
last  two  laws  modify  the  threshold.  The  law  of  Limiting  Value 
modifies  the  threshold  of  value  judgment  in  the  sense  that  old 
objects  and  habitual  quantities  of  these  objects  lose  their  value, 
and  discontinuous  value  movement  to  new  objects  results.  The 
law  of  Complementary  Values,  of  which  the  continuous  value 
movements  to  be  described  are  but  special  applications,  extends 
both  the  lower  and  upper  threshold  of  valuation.  How  these 
thresholds  are  modified  by  acquired  ethical  and  aesthetic  values 
constitutes  the  chief  problem  of  our  study. 


o 


1 94  Valuation :   its  Nature  and  Laws 


II.  VALUE  MOVEMENTS  IN  GENERAL — DEFINITION  AND  CLASSI 
FICATION — THEIR  RELATION  TO  THE  LAWS  OF  VALUATION 

i.  Definition  of  Value  Movements. 

By  the  term  value  movement  we  understand,  then,  any 
reconstruction  of  the  object,  or  readaptation  of  attitude,  from 
which  there  results  continuity  of  valuation.  The  term,  thus 
technically  used,  has  acquired  a  permanent  place  in  the  recent 
contributions  to  worth  theory.  The  term  is  primarily  due  to 
Ehrenfels,1  who  uses  it  to  describe  any  transference  of  worth 
feeling  from  one  object  to  another.  Schwartz  2  makes  use  of 
the  same  conception,  although  he  gives  it  another  name,  Motiv- 
Wandel,  thereby  describing  the  fact  that,  in  the  pursuit  of  one 
object  new  objects  gradually,  and  often  unconsciously,  take  the 
place  of  the  old.  Both  writers  then  seek  to  classify  and  formu 
late  the  laws  of  these  value  movements,  and  ultimately  to  in 
terpret  them  in  terms  of  the  fundamental  laws  of  value.  To 
these  types  of  value  movement  they  give  the  term  Directions 
of  movement,  indicating  thereby  that  the  changes  that  take 
place  are  not  lawless,  but  disclose  certain  trends,  the  causes  for, 
and  the  meaning  of  which,  it  is  the  function  of  worth  analysis 
and  theory  to  discover.  This  conception  of  value  movement 
we  may  accept — with  the  one  modification,  however,  that  it 
shall  be  understood  to  include  also  changes  in  attitude  toward 
the  same  object,  as  well  as  changes  in  the  object,  by  which  mean 
ing  is  acquired. 

With  this  understanding  of  the  use  of  the  term  value  move 
ment,  we  may  distinguish  two  general  types,  already  referred 
to,  the  discontinuous  and  the  continuous.  We  have  the  former 
when,  through  the  working  of  the  laws  of  habit  and  satiety, 
there  is  simply  transference  of  conative  tendency  from  one 
object  to  another.  The  one  acquires  meaning,  the  other  loses 

1  System  der  Werththeorie,  Vol.  I,  p.  135. 

*  Psychologic  des  Willcnst  etc.,  p.  203.  This  passage  may  well  be  quoted  in  its 
entirety :  "  This  relation  [Value  Movement]  is  found  regularly  under  certain  conditions, 
namely,  when,  originally  only  one  tendency  having  moved  us,  unexpectedly  a  new  note 
of  our  will,  hitherto  unsounded,  makes  itself  heard.  As  yet  we  know  not  what  it  is 
that  sounds  softly  within  us.  We  still  think  we  are  acting  only  in  the  direction  of  the 
first  motive.  But  the  second  gains  in  force,  at  first  merely  as  a  subsidiary  tone  of  the 
primary,  perhaps  never  to  be  heard  independently.  And  yet  it  may  suddenly  take 
the  place  of  the  first  and  be  heard  for  itself.  When  that  happens,  when  we  gradually 
begin  to  perform  the  acts  done  from  the  old  motive  under  the  influence  of  the  new,  and 
thereby  either  forget  the  old  or  supplement  the  old  with  the  new,  then  we  have  what  is 
called  Alotiv-wandel  or  value  movement." 


Values  of  Simple  Appreciation  195 

it.  On  the  other  hand,  we  have  continuous  value  movement 
where  there  is  gradual  transference  of  conative  tendency  to 
a  new  object  or  a  new  aspect  of  the  same  object,  where  the  new 
worth  feeling  becomes  complementary  to  the  old  or  the  old 
feeling  remains  as  an  undertone  in  the  new. 


2.  Classification  of  Value  Movements — Their  Directions. 

By  the  directions  of  value  movement,  then,  we  understand 
the  relatively  uniform  types  of  transference  of  worth  feeling  to 
new  objects,  or  to  new  aspects  or  funded  qualities,  of  the 
same  objects,  whereby  new  meaning  is  acquired.  As  modes 
of  acquirement  of  worth  or  affective-volitional  meaning,  they 
are,  in  the  first  place,  distinguishable  only  appreciatively ;  but, 
since  such  movement  presupposes  some  psychical  process  whereby 
the  object  is  modified,  and  also  certain  changes  in  the  functional 
presuppositions  of  the  feeling-attitude  toward  the  object,  these 
appreciative  descriptions  may  be  translated  into  psychological 
terms.  Their  appreciative  character  appears  immediately  in 
the  terminology  employed  in  their  description.  They  have 
been  classified  as  upward  and  downward,  as  forward  and  back 
ward  movement,  as  inward  movement  and  as  movement  toward 
activity.  In  each  case  the  immediate  feeling  or  condition  worth 
of  the  object  has  acquired  some  new  reference,  transgredient 
or  immanental,  not  previously  distinguishable. 

We  may  best  study  these  phenomena  by  a  critical  ex 
amination  of  the  classifications  of  these  directions,  and  of  the 
principles  which  underlie  them.  Ehrenfels  1  distinguishes  four 
principal  types  of  direction,  the  upward,  the  downward,  the 
inward  value  movements,  and  what  he  describes  as  the  move 
ment  toward  activity.  To  such  a  movement  he  gives  the 
general  name  Ziel-folge,  because,  while  for  the  subject  of  the 
value  movement  there  is  no  consciousness  of  end  or  purpose, 
nevertheless,  in  reflection  the  transitions  may  be  seen  to  have 
an  internal  meaning.  We  have  a  Ziel-folge  nach  aufwarts,  up 
ward  value  movement,  when  an  object  valued  immediately  and 
intrinsically  becomes  valued  mediately  because  it  is  instrumental 
to  the  attainment  of  some  new  object  of  value.  The  reverse 
of  this — where  an  object  valued  as  a  means  to  an  end  becomes 
valued  for  itself,  the  former  object  having  been  lost  sight  of,  he 
describes  as  Ziel-folge  nach  abwdrts.  We  have  the  inward  value 

1  System  der  Werththeorie,  pp.  132-141. 


Valuation: 


movement,    «***   ~ 

ierred  from  some  object  of  6        the 

which  the  worth  f  itutrwhen  the  emphJs  of  valuation  is 

toward  activity  takes  place  when  t  i         y       ^  activitles  m. 

shifted  from  the  °Y^  of  Se  object  and'  m  reacting  upon  it 
volved  in  the  presentation  of  the  oD]  e          here  present 

Illustrations  of   these  four  types  movements  are 

in  our  experience.     The  upward  g*^  jnstincts  and  their 
observable  in  connection  with  aU  tun  Q(  ^^  o 

objects.     The  satisfaction  of  hunger   kr    ,  ^  instrument  , 

intrinsic  worth  at  first,  may  c  nie  to  ^.^          be 

to  ideal  and  conceptual  ends,  as  wne,  n  ^^  health 

conscious^  valued  »£££%£;%  Lught  consciously  as 
and  power,  or  when  experience  s  ot  The  downwatd 

means  for  the  development  «*•"££  cases  where  money, 
movement  is,  on  the  ^^trst  as  instruments  to  other 
knowledge,  position,  etc.  sought  «  IUustrations  of  this 

ends,  finally  become  ends  *£•«££»*  and  social  groups, 
tendency  may  also  ^  s^eS^valued  first  merely  as  necessary 
where  frugality  and  self  ^lal  J  own 

means  to  ends,  acquire  an  mtr         va  valuation  of  dis- 

The  inward  value  movement,  toward  t  ^  ^^ 

positions  intrinsically,  often  d«dy  —<      ^          ^  rf 
Lrd  movement,  is  constantly  m  e"  wWch  have 

consideration  the  fund  ament  al  e  hical       p        courage> 
been  historically  developed  in  ^  this  way  fixed  {or  social 

veracity,  etc.,  the  *«ndament"sUee  the  genesis  of  new  dis- 
imitation  and  valuation,  we  may  ^ee  [  th    g  our 

positions  as  objects  of  r'^Tn  ft    conditions  of  life  requiring 
ees.    Every  marked  change  ^  of  disposltion  through 


- 
iost 


sight  of.  activity  is  also  observable  in  con- 

The  movement  toward             V  As  we  shall  see 
nection  with  aU  forms  of  cona  U«  ^tende    y^   ^^ 

later,  it  is  at  the  root  ,     aU  ^  m  activity,  at  first 

esthetic  in  the  broadesl     ense.     VV  to  {unctlon 

directed  toward  some  concrete  ol  |ect  ^e  ^  transference 

with  relative  independence  of  the  end  of  ^ 
of  the  emphasis  of  valuation. 


Values  of  Simple  Appreciation  197 

hunt,  of  love-making,  of  social  communication  of  all  sorts,  all 
instinctive  at  the  start,  and  with  definite  concrete  ends,  may 
become  intrinsically  valued  as  mere  forms  of  activity.  And 
with  this  transference  of  emphasis,  comes  rearrangement  and 
ordering  of  the  activities  as  individualised  wholes  with  a  mean 
ing  of  their  own. 

Schwartz  makes  a  slightly  different  classification  of  these 
phenomena  from  a  somewhat  different  point  of  view.  It 
can  be  shown,  however,  I  think,  that  it  is  really  with  the 
same  uniformities  of  direction  that  his  analysis  is  concerned. 
He  distinguishes  three  fundamental  types  of  value  movement, 
or  Motiv-Wandel.  These  are  the  forward  and  backward  move 
ments  and  the  movement  toward  activity.  By  the  forward  move 
ment  is  understood  the  development  from  mere  condition  worth 
to  ideal  objects  and  values,  either  personal  or  over-individual. 
Typical  illustrations  of  this  movement  are  the  development  of 
passion  into  ideal  love,  of  merely  organic  sympathy  into  conscious 
benevolence,  of  curiosity  into  love  of  knowledge  and  truth.  In 
each  of  these  cases  physical  objects,  the  worth  or  funded  mean 
ing  of  which  is  determined  by  the  modifications  of  feeling  as  a 
condition  of  the  organism,  gradually  pass  over  into  founded  ideal 
objects  whose  worth  is  determined  by  feelings  presupposing  the 
processes  of  judgment  in  which  the  ideal  objects  were  con 
structed.  This  is  described  as  the  forward  movement  because 
it  is  the  normal  direction  which  worth  processes  take  in  con 
tinuous  valuation. 

We  have  the  backward  movement,  on  the  other  hand,  when 
the  emphasis  of  valuation  is  transferred  from  the  object  of  im 
mediate  appreciation  or  condition  worth,  or,  indeed,  from  an 
ideal  object,  personal  or  over-individual,  which  has  developed 
out  of  simple  appreciation,  to  the  mere  hedonic  redundancies 
of  the  feeling  of  value.  In  this  phenomenon,  the  hedonic  reson 
ance  abstracted  from  the  total  attitude,  conceived  as  passive, 
and  presented  in  conceptual  terms  as  quantity  of  pleasure, 
becomes  the  object  of  desire  and  worth  feeling.  This  backward 
movement  is  conspicuously  present  in  all  forms  of  conscious 
hedonism  and  epicureanism,  but  is  also  to  be  observed  in  a 
more  subtle  form  in  sentimental  enjoyment  of  emotions  of  all 
kinds — sympathetic,  religious,  moral,  etc. 

It  will  be  observed  that  Schwartz  does  not  distinguish  a 
special  type  of  inward  value  movement  toward  the  valuation 
of  dispositions.  The  reason  for  this  is  clear  when  we  recognise 


198  Valuation :   its  Nature  and  Laws 

that  it  is  but  one  aspect  of  forward  value  movement.  All 
forward  movement  is  in  the  direction  of  the  construction  of 
ideal  objects  and  their  valuation.  Now  the  disposition  is  an 
ideal  construct  based  upon  the  experience  of  condition  worths, 
and  referred  to  the  self  in  judgment.  It  is,  therefore,  clearly 
a  type  of  forward  movement  of  ideal  reconstruction  of  experi 
ence  in  the  interest  of  continuity  of  valuation.  The  movement 
toward  activity  is  closely  connected  with  the  backward  move 
ment,  although  not  to  be  identified  with  it. 

A  comparison  of  these  attempts  to  classify  value  move 
ments  reveals  two  facts  which,  when  properly  interpreted, 
will  enable  us  to  explain  them  psychologically,  and  to  show 
their  origin  and  function  in  the  processes  of  valuation.  In  the 
first  place,  as  has  already  been  suggested,  these  directions  of 
value  movement  are  appreciative  descriptions  of  certain  uni 
formities  in  modification  of  psychical  content  and  function. 
When  they  are  described  as  movements,  forward  and  backward, 
as  inward  movement  and  movement  toward  activity,  we  have 
before  us  a  complete  system  of  possible  developments  of  desire 
or  conation  from  some  simple  fundamental  desire  as  a  point  of 
departure,  a  system  of  possible  developments  through  which 
worth  may  be  acquired. 

In  the  second  place,  we  find  that  they  fall  into  two  general 
groups :  (i)  the  value  movement  as  transference  of  worth 
feeling  along  the  series  of  means  to  ends,  the  upward  or  down 
ward  movement ;  (2)  value  movement  to  some  new  object,  or 
new  aspect  of  the  old  object,  which  gradually  emerges  in  the 
processes  of  desire  and  feeling  directed  toward  the  primary 
object.  Such  are  the  inward  movement  toward  the  disposition, 
the  movement  toward  activity,  and  the  backward  movement 
toward  the  hedonic  accompaniments.  It  is  with  the  second 
general  class  that  we  are  here  concerned.  They  are,  as  we  have 
described  them,  "  continuous  "  value  movements  in  that  they 
represent  continuous  acquirement  of  meaning. 

In  those  value  movements  where  the  continuity  is  reflective, 
the  change  in  functional  presuppositions  consists  in  the  inter 
polation  of  a  relational  judgment  between  desire  or  feeling 
and  its  object.  The  object  is  of  value  because  it  is  instrumental 
to  the  attainment  of  another  object  of  desire.  The  disposition 
when  viewed  as  object  is  of  value  because  it  is  instrumental 
to  some  desirable  act,  or  because  it  is  identified  with  the  subject. 
Those  value  movements,  on  the  other  hand,  which  are  character- 


l/alues  of  Simple  Appreciation  199 

ised  by  a  gradual  change  in  appreciative  attitude,  by  a  change 
which  takes  place  within  an  emotional  unity  and  continuity,  have 
changes  in  presuppositions  of  another  type.  They  consist  for 
the  most  part  in  the  gradual  substitution  of  assumptions  for 
the  primary  judgments,  the  addition  of  new  judgments  and 
the  combination  of  existential  judgments  and  assumptions  in 
various  ways.  Both  types  of  value  movement  arise,  as  we  shall 
see,  as  adaptations  of  conative  tendency  after  arrest  and  '  dis 
tancing'  of  the  object  of  immediate  desire  and  feeling;  they 
differ  in  the  extent  to  which  the  readaptation  involves  conceptual 
reconstruction. 


3.  Analysis  of  Value  Movements  in  Simple  Appreciation. 

When  we  examine  more  closely  this  second  group  of  value 
movements,  i.e.,  where  an  object  acquires  new  meanings,  it 
is  apparent  that  they  are  all  characterised  by  the  fact  that 
they  begin  in  a  change  of  attitude  toward  the  primary  object, 
in  which  change  new  feelings,  or  modifications  of  feelings, 
emerge,  leading  to  the  imputation  of  complementary  value  to 
the  primary  object,  and  ultimately  to  the  construction  of  new 
ideal  objects.  In  the  first  case  we  have  merely  an  extension  of 
simple  appreciation,  in  the  second  case  an  advance  to  a  new 
level  of  valuation,  as,  for  instance,  of  personal  worths.  The 
inward  value  movement  and  the  movement  toward  activity 
are  first  of  all  value  movements  of  simple  appreciation,  but  they 
are  the  germ  of  later  ideal  constructions  and  values. 

This  gradual  change  in  attitude — with  its  perceptual  and  ideal 
reconstruction  of  the  primary  object — involves  certain  changes 
in  cognitive  presuppositions.  In  general  these  changes  are  of  the 
nature  of  substitution  of  assumption  for  judgment,  of  gradual 
change  of  the  primary  judgment  into  assumption,  explicit  or 
implicit,  and  emergence  of  new  judgments.  The  character  of 
these  substitutions  differs  in  detail  with  the  specific  type  of 
value  movement,  but  the  general  nature  of  the  process  may  be 
described  by  saying  that  the  primary  object  of  judgment,  desire, 
and  feeling,  is  gradually  distanced,  falls  into  the  background, 
the  feeling  of  its  existence  being  retained  merely  as  an  assumption- 
feeling,  while  a  new  object  or  new  aspect  of  the  old  object  comes 
into  the  foreground  as  object  of  presentation  and  judgment. 

To  consider  in  more  detail  the  changes  in  functional  pre 
suppositions  involved  in  the  value  movements  of  simple  appre- 


2OO  Valuation :   its  Nature  and  Laws 

elation,  we  may  start  with  the  type  of  movements  described  as 
substitutions.  Here  it  is  obvious — for  instance,  in  the  case 
where  the  marital  attitude  becomes  gradually  coloured  by  the 
maternal,  or  reversely,  as  in  the  case  of  Rousseau,  described  in 
a  preceding  chapter,1 — that  we  have  to  do  with  phenomena  of 
affective  subsumption,  the  nature  of  which  has  already  been 
described.  As  the  new  aspect  of  the  old  object,  or  the  new 
object,  gradually  becomes  the  object  of  judgment,  and,  there 
fore,  of  new  worth  feelings,  the  old  judgment- feeling  does  not 
disappear,  but  gradually  changes  into  an  assumption-feeling. 
The  old  relation,  the  old  attitude,  the  marital,  for  instance, 
remains  as  a  vague  presupposition  or  assumption,  and  the  affec 
tive  sign  which  goes  with  it  colours  the  new  feeling.  We  have 
in  this  phenomenon  an  emotional  continuity,  whereas  in  the  dis 
continuous  value  movement,  previously  described,  the  transition 
from  one  judgmental  attitude  to  another  is  abrupt. 

The  inward  movement  and  the  movement  toward  activity 
on  the  level  of  simple  appreciation  are  phenomena  of  the  same 
general  type,  of  gradual  readaptation.  In  the  first  case,  the 
inward  movement,  the  judgment- feeling  is  gradually  modified 
until  it  becomes  an  implicit  assumption,  and  gives  place  to  a 
new  modification  of  feeling  and  a  new  object.  The  object  of 
primary  feeling  is  one,  the  existence  or  non-existence  of  which 
is  intrinsically  desired.  Out  of  this  desire  springs  an  act  which 
is  instrumentally  necessary  and  instrumentally  valued.  With 
the  repetition  of  that  act,  and  the  formation  of  habit  or  con 
stancy  of  disposition,  the  object  of  immediate  desire  is  distanced, 
the  control  factor  becomes  subjective,  and  new  assumptions 
and  judgments  emerge.  The  act  itself  is  assumed  to  have  an 
intrinsic  value,  even  when  abstracted  from  the  object.  It 
acquires  an  impulsion,  a  momentum,  so  to  speak,  which  per 
sists  even  when  the  primary  object  sinks  into  the  background. 
Such  an  inward  value  movement  may  be  either  purely  indi 
vidual  and  sub-social,  or  partially  social  in  its  origin  and 
conditions.  It  is  with  its  individual  aspect  that  we  are  first  of 
all  concerned. 

Now,  as  has  been  pointed  out,  the  completion  of  this  value 
movement  leads  to  the  presentation  of  the  disposition  itself  as 
object  of  judgment,  and  to  its  intrinsic  valuation.  But  between 
this  and  the  immediate  desire  for  the  object  or  act  lies  an  inter 
mediate  stage  of  inward  movement,  which  is  the  significant 

1  Chap,  v,  p.  140. 


Values  of  Simple  Appreciation  201 

phenomenon  in  this  connection.  This  intermediate  attitude  is 
characterised  by  an  intensification  of  the  transgredient  reference 
or  tension,  through  the  formation  of  the  dispositional  constant. 
This  modification  of  feeling  has,  however,  as  its  functional 
presupposition,  merely  an  assumption  of  some  ultimate  object 
for  the  transgredient  reference  of  the  feeling.  Thus  the  gradual 
acquirement  of  the  feeling  of  obligation  with  reference  to  any 
act,  a  phenomenon  which  we  shall  later  consider  as  the  most 
important  phase  of  the  inward  movement,  consists  in  adding  to 
the  worth  of  the  object  or  act  the  transgredient  reference  ac 
quired  in  the  formation  of  the  disposition,  this  added  feeling 
having  as  its  presupposition  the  assumption  of  the  existence  or 
non-existence  of  a  more  ultimate  desirable  object  to  which 
the  reference  points.  On  the  level  of  simple  appreciation,  of 
merely  felt  obligation,  this  object  is  always  indefinite.  It  is 
only  through  ideal  construction  and  explicit  judgment,  when 
this  transgredient  reference  is  directed  explicitly  to  such  objects 
as  the  self  or  the  other,  the  law,  the  State,  or  God,  that  assump 
tion  passes  over  into  judgment.  The  process  by  which  obligatory 
character  is  imputed  to  an  object  or  act  may  then  be  viewed 
as  a  continuous  value-movement  in  which  an  old  worth  feeling 
is  gradually  subsumed  under  a  new,  the  new  in  this  case  being 
the  assumption-feeling  corresponding  to  the  dispositional  con 
stant  created.  The  acquired  worth  in  this  case  is  an  increase 
of  the  transgredient  reference. 

The  movement  toward  activity  also  consists  in  the  gradual 
distancing  of  the  primary  object  of  desire  and  worth  feeling 
and  the  interpolation  of  a  new  attitude  or  object.  In  its  complete 
form  the  activity  of  conation  or  feeling  becomes  the  presented 
object  of  new  judgment-feelings,  as  when  for  instance  the  activities 
of  play,  the  hunt,  or  love,  as  in  coquetry,  become  the  object  of 
worth  themselves,  without  any  conscious  reference  to  their 
primary  objects.  But  here  again,  between  immediate  desire 
for  the  object  and  the  stage  of  explicit  valuation  of  the  activity 
directed  toward  the  object,  there  is  an  intermediate  phase  of 
importance  for  simple  appreciation.  It  is  the  stage  where  the 
gradual  substitution  of  the  activity  leads  merely  to  the  imputa 
tion  of  the  new  worth  thus  acquired,  to  the  original  object  of 
desire.  The  process  here  involved  is  relatively  simple.  The 
original  object  of  desire,  the  judgment  of  the  existence  or  non- 
existence  of  which  is  followed  by  worth  feelings,  is  gradually 
distanced,  and  for  the  judgment  a  mere  assumption  is  substituted. 


2O2 


Valuation :   its  Nature  and  Laws 


As  the  desired  object  sinks  into  the  background,  the  activity 
comes  into  prominence. 

Now,  the  characteristic  of  this  acquired  worth,  appreciatively 
described,  is  its  immanental  reference.     The  distancing  of  the 
object  from  immediate  desire,  which  takes  place  when  judgment 
passes  over  into  assumption,  and  the  coming  into  the  foreground 
of  consciousness  of  the  mere  activities  of  conation  and  feeling, 
make  possible  repose  in  the  object,  together  with  the  acquirement 
of  new  worth  through  the  independent  functioning  of  the  activi 
ties  originally  directed  upon  the  object.     When  we  come  to  study 
in  more  detail  the  acquirement  of  aesthetic  worth,  as  the  most  im 
portant  aspect  of  the  movement  toward  activity,  we  shall  find 
that  the  acquired  immanental  reference  of  the  aesthetic  attitude  is 
a  worth  moment  which  has  as  its  corresponding  object  those  ideal 
objects  which  we  mean  by  the  words  beauty,  grace,  sublimity,  etc. 
Finally  we  may  see  that  the  movement  towards  activity,  as  a 
felt  continuity,  must  also  be  viewed  as  an  emotional  subsumption. 
The  new  feelings  which  arise  in  the  perceptual  and  ideal  con 
struction  of  the  object  in  the  interests  of  mere  activity,  are  sub 
sumed  under  the  fundamental  emotional  attitude,  some  senti 
ment  or  mood,  directed  toward  the  primary  object. 

4.  Interpretation  and  Explanation  of  Value  Movements  :    Their 
Relation  to  the  Laws  of  Valuation. 

A  general  review  of  the  changes  in  functional  presuppositions 
which  characterise  all  these  value  movements  in  simple  apprecia 
tion,  discloses  the  fact  that  the  common  element  in  them  all 
is  the   distancing  of  the  primary  object  of  desire  or  feeling, 
and  the  interpolation  of  new  feeling  attitudes  with  modified 
presuppositions.     The    change    of    judgment    into    assumption, 
the  assimilation  of  new  judgments  to  these  assumptions,  and 
through  it  all  a  felt  continuity,  are  the  significant  factors.     It  i 
important  to  emphasise  both  the  fact  and  the  nature  of  this 
continuity.     These  processes  are  continuous  in  that,  in  contrast 
with  the  reflective  developments  or  the  movements  determined  by 
instrumental  judgments,  all  these  movements  of  simple  apprecia 
tion,  the  affective  substitutions,  the  inward  movement  and  the 
movement  toward  activity,  constitute  a  gradual  assimilation  of  a 
new  meaning  to  an  old  object  or  a  new  object  to  an  old  meaning. 
In  the  second  place,  the  character  of  the  continuity  is  best 
scribed  as  emotional,  as  an  emotional  subsumption.     The  mean 


Values  of  Simple  Appreciation  203 

term  of  the  transition  is  emotional ;  the  assumption  feeling, 
representing  the  judgment  habit  of  former  valuations  of  the 
primary  object,  is  retained  as  the  background  or  "  affective 
sign  "  which  gives  colour  to  the  new  feeling.  Such  subsumptions 
we  may  describe  as  forms  of  emotional  logic. 

The  justification  of  this  conception  of  an  emotional  as  dis 
tinguished  from  an  intellectual  logic  has  already  been  attempted 
in  an  earlier  chapter.1  In  that  chapter  also  the  nature  and  con 
ditions  of  these  affective  subsumptions  have  been  treated  in  their 
more  psychological  aspects,  and  illustrations  of  the  fundamental 
types  of  subsumption  developed  in  detail.  Here  it  is  important 
merely  to  emphasise  the  fact  that  the  value  movements  of  simple 
appreciation  are  of  this  general  type. 

But  the  significance  of  these  emotional  continuities  really 
appears  only  when  they  are  viewed  genetically  as  readaptations 
(progressions  or  regressions,  as  the  case  may  be),  following  upon 
the  working  of  those  laws  of  interest  or  of  affective-volitional 
meaning,  described  as  the  laws  of  valuation.  The  principal 
law,  the  law  of  Limiting  Value,  which,  we  have  seen,  applies  to 
all  objects  of  condition  worth,  and  to  all  instrumental  construc 
tions  growing  out  of  valuation  of  objects  of  condition  worth, 
may,  in  its  functional  aspect,  be  interpreted  as  a  law  of  arrest, 
and  the  value  movements  as  readaptations  after  arrest.  The 
general  law  of  acquirement  of  Complementary  Value  describes  in 
general  terms  certain  forms  of  readaptation  after  arrest.2 

Functionally  viewed,  the  law  of  Limiting  Value  formulates 
the  conditions  under  which  arrest  and  discontinuity  of  conative 
process  appear.  When  reduced  'to  their  psychological  causes, 
these  arrests  were  seen  to  be  of  two  general  types.  In  the  first 
place  there  is  the  arrest  of  any  isolated  conative  tendency  due 
to  habit  and  satiety  following  upon  repetition  or  over-stimulation 
of  the  sense  tendencies  involved.  In  the  second  place,  however, 
and  equally  important,  is  the  modification  of  feeling,  and  arrest 

1  Chap,  v,  pp.  121  ff. 

2  The  importance  of  the  moment  of  arrest  in  value  movement  must  be  emphasised. 
The  criticism  made  by  Hegel  upon  those  theories  of  the  development  of  mind  which 
proceed  in  a  purely  "affirmative  manner"  may  still  hold  to-day  if  we  translate  their 
intellectualistic  terminology  into  affective-volitional.     In  speaking  of  these  methods,  he 
says  :  "  Their  ruling  principle  is  that  the  sensible  is  taken  (and  with  justice)  as  the  prius 
and  the  initial  basis,  but  that  the  latter  phases  that  follow  from  this  starting-point  present 
themselves   as   emerging  in  a  solely  affirmative  manner,   and  the  negative  aspect  of 
mental  activity  by  which  the  material  is  transmuted  into  mind  and  destroyed  as  a 
sensible  is  misconceived  and  overlooked"  (Philosophy  of  Mind,  William  Wallace,  The 
Clarendon  Press,  sec.  442,  p.  61).     So  also  we  may  say  that,  while  the  condition  worths 
are  the  prius  of  higher  levels  of  valuation,  the  later  phases  arise  only  through  arrest  of 
the  primary. 


2O4  Valuation:   its  Nature  and  Laws 

of  conation  when  the  limit  of  instrumental  judgment  is  reached, 
when  the  reapplicability  of  the  object  in  question  for  the  satisfac 
tion  of  primary  wants  ceases,  or  when  acts,  instrumentally 
valuable  in  the  attainment  of  objects,  become  inapplicable,  and 
when  attainment  is  no  longer  probable,  or  even  possible. 

These  then  are  the  causes  of  arrest  of  conative  tendency  which 
lead  to  value  movement.  But  it  is  all  important  to  emphasise  the 
fact  that  these  arrests  of  conative  tendency  may  be  complete 
or  only  partial,  for  in  the  difference  of  the  degree  of  the  arrest 
we  have  the  source  of  the  distinction  between  different  types  of 
value  movements.  Where  the  arrest  is  complete,  the  value 
movement  is  discontinuous,  and  we  have  transference  of  cona 
tion  and  feeling  to  a  new  object.  Where,  on  the  other  hand, 
the  arrest  is  only  partial,  we  have  value  movement  of  the  con 
tinuous  type,  the  gradual  assimilation  of  a  new  object  to  an  old, 
or  gradual  change  in  attitude  toward  the  primary  object,  by 
which  complementary  value  is  acquired.  The  value  of  the 
primary  object  remains  as  an  assumption,  a  presupposition  of 
the  new  feelings. 

Finally  we  may  note  the  teleological  character  of  these 
value  movements.  Functionally  viewed,  they  are  readapta- 
tions  after  arrest,  whereby  continuity  of  valuation  is  secured 
and  meaning  is  acquired.  The  description  of  the  directions 
of  these  movements  or  types  of  adaptations,  is,  we  have 
already  seen,  appreciative  in  character,  and  therefore  pre 
supposes  the  postulate  of  all  appreciative  descriptions, 
enhancement  of  worth  or  acquirement  of  meaning,  and  is 
teleological  in  its  nature.  But  when  these  value  movements 
are  described  as  teleological,  it  is  obviously  not  meant  that  they 
are  determined  by  an  explicit  consciousness  of  end.1  Such 
explicit  reference  to  ends  appears  only  upon  the  higher  levels  of 
valuation.  Looking  back  from  these  higher  levels,  we  may  see 
that  the  complementary  values  of  objects  of  condition  worth 
contain  the  germ  of  developments  to  higher  levels,  but  the  objects 
of  valuation  on  these  higher  levels  are  not  foreseen  in  simple 
appreciation.  In  the  value  movements  of  simple  appreciation 
the  teleology  is  wholly  immanental. 

1  The  use  made  of  Wundt's  descriptive  formula  (Heterogeneity  of  Ends)  in  describing 
these  progressions  constitutes  an  explicit  denial  of  such  character  to  the  processes, 
complementary  values  which  arise  in  these  movements  are  unforeseen,  are  not  prese 
in  idea,  but  gradually  emerge  in  the  acquisition  or  enjoyment  of  objects  of  primary  con 
dition  worth      The  teleological  character  appears  rather  as  immanental  and  is  described 
by  the  application  of  Wundt's  second  principle  the  Law  of  Resultants.     The  results 
these  value  movements,  of  these  perceptual  and  ideal  constructions  of  the  object,  have  i 
worth  or  meaning  not  in  the  elements. 


Values  of  Simple  Appreciation  205 

III.  ETHICAL  AND  ESTHETIC  VALUES  ACQUIRED  IN  VALUE 
MOVEMENTS  OF  SIMPLE  APPRECIATION  —  THEIR  ROLE  AS 
DETERMINANTS  OF  THE  ECONOMIC  VALUES  OF  ACQUISITION 
AND  CONSUMPTION 

i.   Ethical  and  ^Esthetic  Values  as  Modifications  of  Feelings 
of  Condition  Worth. 

There  are  two  fundamental  ways  in  which  the  simple  appre 
ciation  of  the  condition  worth  of  the  object  may  be  modified. 
The  worth  attitude  may  acquire  the  attribute  of  obligation  or 
of  aesthetic  repose  in  the  object.  These  acquired  feelings  then 
become  the  basis  of  imputation  of  new  worth  to  the  object. 

It  is  of  the  utmost  importance  to  recognise  the  relativity  of 
these  appreciative  differences.  The  a  priori  distinctions  drawn 
by  Kant  do  not  maintain  themselves  upon  closer  analysis. 
Genetically  viewed,  the  more  primitive  ethical  and  aesthetic 
values  arise  in  the  very  processes  of  consumption  and  acquisition. 
The  obligation  to  cleanliness  and  thrift,  the  aesthetic  values  of 
taste  and  refinement  in  living,  are  acquired  almost  imperceptibly. 
And  from  the  analytical  point  of  view  also,  it  is  impossible  to  say 
that  obligation  is  something  fundamentally  different  from  desire 
or  that  the  aesthetic  is  desireless  appreciation.  They  are 
merely  appreciative  distinctions  within  the  total  worth  process, 
modifications  of  attitude  by  which  meaning  is  acquired.  And, 
when  the  problem  is  approached  from  the  side  of  economic 
analysis,  it  is  seen  that  even  the  economist  cannot  keep  his 
province  distinct,  but  is  forced  to  see  his  general  laws  modified 
by  the  intrusion  of  ethical  and  aesthetic  motives.  These  facts, 
which  constitute  the  bane  of  the  various  worth  sciences  when  they 
seek  to  work  alone,  are  precisely  those  which  contain  the  greatest 
promise  for  worth  theory  as  a  whole.  An  act  or  an  attitude 
acquires  ethical  worth  or  meaning  when  it  becomes  obligatory, 
and  for  an  act  to  be  felt  as  obligatory,  means  that  it  has  become 
the  object  of  a  new  kind  of  worth  feeling.  An  object  acquires 
aesthetic  worth  or  meaning  when  it  becomes,  as  we  say,  beautiful, 
the  term  being  used  in  its  larger  sense  to  include  all  the  modifica 
tions  of  the  aesthetic.  And  for  an  object  to  be  felt  as  beautiful, 
means  that  it  has  become  the  object  of  a  new  kind  of  worth 
feeling.  In  each  case  the  oughtness  or  the  beauty  of  act  or  object 
is  a  funded  meaning  or  worth  acquired  in  some  process  and 
imputed  to  the  object  or  act. 


;06  Valuation  :   its  Nature  and  Laws 

When  we  seek  to  describe  appreciatively  these 

precisely  described  the  sense  ;  of  beauty  ,s  a  d^«    g 


sw 


saart 

r^renftoward^ctivity  aheady  descnhed. 


1  Chap,  in,  pp.  68-71. 


Values  of  Simple  Appreciation  207 

2.    The  Ethical  Mode  of  Simple  Appreciation — Its  Modification 
of  the  Condition  Worth  of  Objects. 

(a)  Analysis  of  the  Impellent  Mode  of  Obligation. 

An  attempt  to  analyse  more  completely  the  sense  of  obligation 
shows  that  it  is,  in  one  sense,  an  ultimate  mode  of  appreciation 
which  is  not  further  reducible.  Our  description  of  it  as  a  deepen 
ing  of  the  transgredient  reference  as  impellent  mode,  is  in  reality 
only  a  description  and  not  a  definition.  Strictly  speaking, 
definition  is  impossible.  If  we  wish  to  go  further  in  our  fixation 
of  this  mode  of  experience,  it  must  be  by  a  method  of  interpolation 
of  the  mode  in  a  series  of  equally  ultimate  meanings,  and  by 
determination  of  their  psychological  equivalents.  But  in  follow 
ing  this  procedure  we  are  immediately  met  by  a  considerable 
difference  of  opinion.  Thus  Simmel  finds  obligation  to  be  a 
fundamental  modality  of  thought  (Denk  Modus)  ;  others  de 
scribe  it  as  a  mode  of  will  and  still  others  as  a  mode  of  feeling. 
On  closer  inspection  it  will,  nevertheless,  appear,  I  think,  that 
there  is  no  necessary  irreconcilability  in  these  different  views. 
Worth  experience,  of  which  obligation  is  a  mode,  is  seen  always 
to  be  feeling  with  certain  cognitive  presuppositions,  while  the 
distinction  between  feeling  and  will  has,  on  closer  analysis, 
shown  itself  to  be  a  relative  difference  of  intent  and  not  of 
content. 

Simmel  finds  obligation,  das  Sollen,  in  one  aspect,  a  mode  of 
thought  which  lies  midway  between  the  judgments  of  non- 
existence  and  existence.  Speaking  of  the  various  possible 
attitudes  toward  objects,  he  says  :  "  One  could  arrange  them 
all  in  a  phenomenological  series  which  extends  from  the  mere 
presentation  of  an  object  for  our  thought,  without  existence,  to 
complete  reality.  Das  Wollen,  das  Hoffen,  das  Konnen,  das 
Sollen,  all  these  are,  so  to  speak,  mediate  stages  between  non- 
existence  and  existence,  which,  for  one  who  has  never  experienced 
them,  we  could  as  little  define  as  we  are  able  to  say  what  being  or 
thinking  really  is:  there  is  no  definition  of  obligation."1  Here 
then  the  attitude  of  obligation  is  indirectly  defined  by  giving  it  its 
place  in  a  continuous  vital  series,  in  much  the  same  fashion 
that  the  relations  of  attitudes  of  feeling  and  will  were  defined 
.  in  Brentano's  series.  What  is  it  that  distinguishes  the  attitude 
of  obligation  from  the  adjacent  terms  or  modes  in  the  series  ? 

1  Simmel,  Einleitung  in  die  Moralwissenschaft,  Berlin,  1892,  Vol.  I,  p.  8. 


2o8  Valuation :   its  Nature  and  Laws 

Evidently  it  is  its  different  cognitive  presuppositions.     It  will  be 
recalled  that  we  found  the  difference  between  feeling  and  will 
in  difference  of  intent  or  meaning  of  the  same  content  according 
as  the  object  is  recognised  as  independent  of,  or  dependent  upon, 
the  subject.     With  the  recognition  of  independence  of  the  object 
we  have  feeling,  with  the  recognition  of  dependence  we  have 
desire  or  volition,  according  to  the  degree  of  the  dependence. 
Now  worth  feeling  is  feeling  with   judgments  or   assumptions 
as  its  presuppositions,  and  the  sense  of  obligation  being  a  mode 
of  worth  feeling,  we  must  look  for  its  differentia  in  these  terms^ 
The  feelings,  "  I  wish,"  "  I  hope,"   "  I  can,"   "  I  should, 
must  "   "  I  will,"  represent  an  ascending  scale  of  changes  i 
these  presuppositions  in  which  the  dependence  of  the  act  or  object 
upon  the  self  comes  more  clearly  to  consciousness, 
presupposes  merely  the  judgment  of  non-existence;   '       hope 
this  judgment  as  qualified  by  an  assumption  of  possibility ; 
must  "  represents  this  possibility  as  having  passed  over  into  ex 
plicit  judgment  of  existence.  The  feeling  "  I  ought"  lies  midway 
between  power  and  constraint. 

Obligation,  when  thus  inserted  into  a  series  of  cognate  attitudes 
is  seen  to  be,  in  one  aspect,  a  mode  of  thought.     But  this  serial 
method  of  interpretation  also  enables  us  to  answer  the  other  ques 
tion  as  to  whether  it  is  a  mode  of  feeling  or  will.    Common  speech 
allows  either  description.     For  just  as  we  fed  need  or  hope  or  our 
power  to  do  a  thing,  so  we  may  equally  well  be  said  to  feel 
obligation  or  the  necessity  of  an  act.     The  sense  of  obligation 
may,  therefore,  very  properly  be  described  as  a  mode  of  worth 
feeling      On  the  other  hand,  when  seen  prospectively,  in  1 
light  of  the  succeeding  attitude  of  will,  the  sense  of  oughtness 
seems  to  be  nearer  to,  in  fact  a  preliminary  stage  of  volition. 
The  validity  of  both  descriptions  appears  when  we  apply  the 
results  of  our  preceding  study  and  recognise  that  the  distinction 
between  feeling  and  will  is  itself  not  ultimate,  but  that  will  is 
also   feeling  with   certain   characteristic   presuppositions, 
sense  of  obligation  appears  to  be  a  mode  of  will,  and  not  feeling, 
only  when  the  term  feeling  is  limited  to  those  hedomc  con 
comitants  which  go  with  emotional  disturbance.     The  feeling  ot 
obligation  is  by  no  means  to  be  identified  with  emotional  di< 
turbance,  for   the   deepest   obligation   may  be   represented 
affective  abstracts  and  even  by  affective  signs,  often  t 
emotional  connotation  of  a  word. 

If  this  analysis  of  the  sense  of  obligation  is  adequate,  it  is  not 


Values  of  Simple  Appreciation  209 

difficult  to  show  its  place  in  the  system  of  value  movements  by 
which  condition  worths  acquire  complementary  value.  Its 
subsumption  under  the  inward  movement  has  already  been 
suggested  in  a  preceding  paragraph,  where  it  was  seen  to  be  a 
movement  from  the  desire  for  the  object  itself  to  an  attitude  where 
the  desire  is  qualified  by  a  consciousness  of  the  disposition  pre 
supposed.  It  is  an  inward  movement  in  which  the  object,  or  act 
directed  toward  its  realisation,  acquires  all  the  meaning  of  this 
inward  reference.  The  "  distancing  "  of  the  object,  its  detach 
ment  from  immediate  satisfaction  through  arrest,  is  followed  by 
the  deepening  of  the  transgredient  reference,  which  in  turn  is  the 
sign  of  the  added  element  of  a  subjective  control  having  its 
locus  in  the  pre-formed  disposition.  The  mere  feeling  of  the  non- 
existence  of  the  object  is  supplemented  by  the  assumption  of 
possibility,  the  feeling  "  I  can,"  which  then  passes  over  into 
judgments  of  existence  and  necessity.  The  impellent  mode  which 
later  develops  into  explicit  sense  of  obligation  is,  therefore,  a 
transition  stage  between  desire  springing  out  of  non-existence 
and  the  sense  of  the  "  must  be  "  which  comes  with  the  judgment 
of  existence.1 

(b)    Pre-Ethical  and  Quasi-Ethical  Impulsions  and  Obligations: 
Their  Sub-Personal  and  Sub-Social  Character. 

The  feeling  of  obligation  in  its  simplest,  pre-ethical  form, 
as  impellent  mode,  is  accordingly  merely  a  new  modification 
of  worth  feeling.  This  modification  is,  as  has  been  already 
pointed  out,  at  first,  in  its  germinal  form,  objectless.  Its  apparent 
object  is  still  the  primary  object  of  simple  desire,  the  simple 
desire  and  feeling  having  been  insensibly  qualified  by  the  deep 
ening  of  the  transgredient  reference.  Its  real  object r  however, 
is  the  disposition  presupposed,  and  ultimately  the  ideal  construct 
of  the  self  or  the  social  group,  to  which  the  feeling  is  referred 
when  the  progression  or  value  movement  has  reached  the  stage 
of  explicit  acknowledgment  and  characterisation  of  the  pre 
suppositions  of  the  feeling.  For  this  explicit  acknowledgment 

1  This  origin  of  the  movement  toward  obligation  as  dependent  upon  arrest  is  well 
described  by  Simmel  in  Vol.  II,  p.  387  of  the  Einleitung,  "  Das  Sollen  nimmt  eine 
mittlere  Stellung  zwischen  dem  Miissen  und  dem  Wollen  ein  ;  beim  Miissen  stellen  sich 
der  Handlung,  die  schlieslich  aus  einem  iiberwiegenden  Grunde  doch  gewollt  wird,  sehr 
starke  Wollungen  entgegen  ;  bei  dem  freien  Wollen  gar  keine  ;  bei  dem  Sollen  eine 
gewisse  Anzahl,  deren  Uberwindung  das  Maass  des  sittlichen  Verdienstes  angiebt."  He 
is  here  describing  obligation  on  the  higher  levels  of  personal  and  social  worths. 
P 


2io  Valuation:   its  Nature  and  Laws 

• 

and  characterisation  in  judgment,  certain  processes  of  sym 
pathetic  imitation  and  projection,  or  pinfuhlung,  are  necessary. 
But  the  object,  later  to  be  characterised  and  acknowledged,  is 
vaguely  anticipated  in  the  transgredient  reference  of  the  feeling. 
The  objects  of  the  feeling  of  obligation  are  therefore  always  ideal 
objects,  and  it  is  their  worth  which  is  reflected  in  the  feeling  of 
obligation. 

The  feeling  of  obligation  has,  accordingly,  different  degrees  of 
explicitness.  It  may  be  merely  felt,  as  in  simple  appreciation, 
or  it  may  be  explicitly  referred  to  some  ideal  object.  In  the 
first  case  some  object  of  the  new  qualification  of  feeling  is  vaguely 
assumed,  while  in  the  second  case  some  ideal  object,  such  as  the 
ideal  of  the  self  and  its  dispositions,  or  the  idea  of  social  good  or 
law,  is  constructed,  and  to  this  the  feeling  is  referred. 

We  may  distinguish  three  levels  of  development  at  which  this 
deepened  transgredient  reference  may  appear,   and  they  may 
be    described,    in   Guyau's    terms,    as    the    three    psychological 
equivalents  of  obligation.1     The  first  equivalent  is  that  which 
we  have  described  as  belonging  to  the  level  of  simple  appreciation, 
the  dynamic  suggestion,  the  transgredient  reference  which  an 
attitude  attains  in  an  individual  as  the  result  merely  of  the 
formation  of   conative   dispositions  or  constants  through  rep 
etition   and  habit.      This   equivalent  we  may  describe  as  the 
instinctive,  pre-ethical,  obligation  on  the  level  of  simple  apprecia 
tion  of  objects.     The  second  equivalent  is  the  tension,  deepening 
of  the  transgredient  reference,  which  arises  through  sympathetic 
projection    (Einf uhlung) .     The    presumption    of    the    existence 
of  the   feeling   and  feeling-disposition  in  another,  or  in  social 
groups,    deepens    the    transgredient    reference    of    the    feeling 
as  condition  of  the  subject.     This  Guyau  describes  as  tension 
arising  from  "  fusion  "  of  desires  and  feelings.     The  third  equiva 
lent  is  the  further  increase  of  tension  or  transgredient  reference 
which  arises  at  the  stage  of  reflection  when  the  object  of  the 
feeling  is  explicitly  characterised  and  acknowledged  as  a  demand, 
personal  or  over-individual,  social  or  ideal,  as  the  case  may  be. 
This   third   equivalent   corresponds    to    the    distinctly    ethical 
obligation.     The  others   may   be   described   as   the   pre-ethical 
and  quasi-ethical,  respectively. 

The  situations  in  which  this  deepening  of  the  worth  con 
sciousness,  described  as  the  sense  of  obligation,  appears  most 

1  Guyau,  Esquisse  d'une  Morale,  sans  obligation  et  sanction,   Book   I,  chap,   in, 
especially  p.  127. 


Values  of  Sim-pie  Appreciation  211 

marked  are,  as  the  Kantian  analysis  rightly  discovered,  those 
in  which  oppositions  between  personal  and  over-individual  worths, 
and  between  personal  or  over-individual  worths  and  condition 
worths  occur,  where  in  fact  preference  and  sacrifice  appear. 
Such  situations  presuppose  processes  of  sympathetic  projection 
and  the  ideal  construction  of  personal  and  over-individual 
objects  toward  which  the  feeling  of  obligation  is  directed.  The 
feeling  of  obligation  in  such  cases  presupposes,  therefore,  explicit 
judgments  of  the  existence  or  non-existence  of  the  objects 
and  explicit  references  of  them  to  the  self  or  the  alter.  With 
these  we  sfyall  be  concerned  in  later  chapters.  But  while  the 
higher  levels  of  obligation  presuppose  these  processes  of  social  pro 
jection  and  imitation,  it  must  not  be  thought  that  the  feeling  of 
obligation  is,  as  simple  impellent  mode,  exclusively  social  in  its 
origin,  or  that  the  value  movement  toward  the  ethical  is  con 
ditioned  by  the  presentation  of  the  attitude  as  a  personal  or  social 
worth.  This  is  as  far  from  the  truth  as  it  would  be  to  say  that  the 
aesthetic  mode  of  worth  experience  is  socially  conditioned  because 
most  of  the  specific  aesthetic  activities  are  social  in  their  origin. 
The  primary  sense  of  obligation  is,  as  Guyau  in  his  splendid 
analysis  shows,  both  sub-personal  and  sub-social.  Mere  instinctive 
feeling,  as  a  simple  condition  worth,  prior  to  its  presentation 
either  as  an  attitude  of  the  self  or  the  alter,  has  in  it  the  potenti 
ality  of  that  situation  of  contrast  and  opposition  out  of  which  the 
sense  of  obligation  arises.  Guyau  has  described  this  primary 
sense  of  obligation  as  strictly  correlative  to  the  sense  of  capacity. 
Wherever  in  the  face  of  arrest  emerges  the  sense  of  "  I  can," 
there  tends  to  follow  from  it  imperceptibly  the  sense  of  "  I 
ought." 

(c)  Illustrations  of  Various  Instinctive,  Quasi-Ethical  Obligations. 

Let  us,  then,  first  examine  some  expressions  of  this  sub- 
personal,  sub-social  sense  of  obligation  prior  to  ideal  construction 
and  conscious  reference  of  feeling  to  ideals.  There  is  scarcely  any 
conative  tendency  which  cannot  under  certain  conditions  acquire 
the  sense  of  obligation.  We  are  hardly  aware  of  the  constant 
undertone  of  obligation  which  accompanies  our  simplest,  most 
instinctive  acts.  One  of  the  most  noticeable,  and  also  most 
instructive,  is  the  sense  of  obligation  which  the  mere  possession 
of  brute  physical  strength  may  acquire.  We  begin  to  exert  our 
strength  for  a  given  end.  If  opposition  appears,  we  frequently 


212  Valuation  :   its  Nature  and  Laws 

find  ourselves  impelled  to  persist  long  after  the  object  of  the 
original  effort  loses  its  interest.  The  place  of  worth  has  been 
usurped  by  the  attitude.  Inward  value  movement  has  taken 
place.  All  this  would  be  simple  enough  in  those  cases  where 
persistence  of  effort  is,  under  the  stimulus  of  social  imitation  and 
contrast,  ideally  presented  as  a  personal  worth,  or  as  an  object 
with  over-individual  reference,  but  the  point  here  is  that  ex 
perience  shows  us  this  obligation  emerging  on  the  level  of  simple 
appreciation,  sub-personally  and  sub-socially,  prior  to  the  ideal 
presentation  of  the  attitude. 

And  if  strength  obliges,  it  is  also  true  that  other  instinctive 
activities  connected  with  condition  worths  acquire  the  sense 
of  obligation.  Sex,  for  instance,  has  its  impulsions  of  an  instinc 
tive  sort,  prior  to  that  reference  of  its  attitudes  and  dispositions 
to  the  self  and  society  which  creates  the  ethical  and  moral  obliga 
tion  proper.  Of  such  instinctive  obligation  Guyau  has  given  a 
good  illustration  in  the  case  of  a  young  girl  who  cast  herself 
out  of  the  window  rather  than  endure  the  embraces  of  a  husband 
who  did  not  call  out  her  love.  Her  state  of  mind  bore  all  the 
characteristic  marks  of  the  feeling  of  remorse,  although  all  the 
acquired  obligations,  personal  and  social,  were  met.  She  had 
obeyed  the  commands  of  her  parents  and  the  demands  of  society, 
but  the  instinctive  obligations  of  sex  triumphed  in  the  form  not 
merely  of  distaste  and  unhappiness,  but  of  a  primary  form  of 
remorse  of  the  most  fundamental  character.  So  too  the  derived 
emotional  attitude  of  shame,  which  represents  the  product  of  a 
long  series  of  racial  inhibitions,  carries  with  it  a  primary  obliga 
tion  of  such  a  character  that,  even  after  reflective  consideration, 
after  the  construction  of  personal  and  social  ideals  which  tend  to 
modify  it,  it  triumphs  in  an  instinctive  obligation  which  cannot 
be  gainsaid. 

An  instructive  characteristic  of  these  obligations,  which  may 
perhaps  serve  to  account  for  their  retention  after  reflection 
comes  in  to  modify  them,  is  their  complementary  worth,  height 
ening  the  worth  of  the  instinctive  activity  itself.  Indeed  one  of 
the  most  remarkable  phenomena  of  the  worth  consciousness  is 
precisely  this  tendency  to  retain  obligations  and  even  to  create 
them.  In  connection  with  this  very  instinct  of  sex,  the  in 
numerable  little  delicacies  of  obligation  with  which  it  hedges 
itself  about,  the  inhibitions  which  it,  so  to  speak,  sets  itself,  often 
of  no  social  significance  and  sub-personal  in  so  far  as  they  are 
not  consciously  referred  to  the  self,  are  so  many  ways  of  instinc- 


Values  of  Simple  Appreciation  2 1 3 

lively  acquiring  new  worth  and  meaning.  Cases,  almost  ab 
normal  in  character,  have  been  pointed  out  (Zola  describes  it  in 
the  case  of  one  of  his  characters)  of  women  who,  after  a  life 
in  which  they  have  abandoned  these  instinctive  inhibitions, 
obligations,  and  shames  of  sex,  have,  when  entering  upon  a  real 
passion,  sought  to  restore  them  with  a  zeal  no  less  than  pathetic. 
The  sense  of  unworthiness  in  the  presence  of  a  true  love  leads 
to  the  development  of  a  conscious  cult  of  modesty  which  the 
pure  woman  who  had  never  felt  the  loss  of  these  values  would 
never  think  of  building  up. 

The  significance  of  these  phenomena  which,  although  they 
have  been  described  in  the  case  of  but  two  fundamental  con- 
ative  tendencies,  are  really  characteristic  of  all,  for  the  instinctive 
creation  of  individual  obligations  is  everywhere  present  in  our 
affective-volitional  life,  lies  in  the  fact  that  this  primary  equiva 
lent  of  obligation,  this  tendency  to  inward  value  movement, 
whereby  the  object  of  immediate  desire  is  set  at  a  distance, 
and  thus  acquires  the  complementary  value  of  the  transgredient 
reference  of  the  disposition  presupposed,  is  a  normal  form  of  value 
movement  in  the  acquirement  of  affective-volitional  meaning. 
It  is  a  special  form  of  that  general  law  of  value  movements,  the 
principle  of  Heterogeneity  of  Ends  according  to  which  conative 
tendency  directed  toward  an  object  develops  new  ends  and  values 
not  foreseen. 

(d)  Modification  of  Economic  Valuation  by  these  Acquired 

Obligations. 

The  close  relation  which  has  been  shown  to  exist  between 
simple  desire  and  instinct  and  the  primary  equivalent  of  obliga 
tion  gives  us  a  point  of  view  from  which  we  may  understand  the 
genetic  relation  between  simple  condition  worths  in  general  and 
ethical  worths,  more  particularly  between  economic  and  ethical 
values.  The  economic  value  of  an  object  is  determined  by  its 
capacity  to  satisfy  desire,  either  intrinsically  by  virtue  of  its  rela 
tion  to  conative  dispositions,  or  instrumentally  through  its  causal 
relation  to  other  objects  which  have  this  intrinsic  capacity. 
The  activities  of  acquisition  and  consumption  which  determine 
the  economic  worth  of  an  object  are  themselves  unethical,  non- 
moral,  but  may  obviously  acquire  ethical  values,  and  become 
moralised.  Certainly  they  may  become  moralised  through  their 
relation  to  personal  and  social  ends  presented  ideally  as  intrinsic- 


214  Valuation:   its  Nature  and  Laws 

ally  valuable  objects  with  which  the  economic  activities  are  in- 
strumentally  related.  But  it  would  also  seem  that  in  the  mere 
processes  of  acquisition  and  consumption  these  condition  worths 
may  insensibly  acquire  an  ethical  meaning.  The  instinctive 
obligations  which  we  have  described  may  become  thus  qualified, 
independently  of  any  consciousness  of  more  remote  ends,  personal 
or  social,  presupposed  in  the  processes  of  mere  acquisition  and 
consumption.  For  such  quasi-ethical  qualifications  there  are, 
as  we  have  seen,  definite  descriptive  terms  such  as  thrift,  enter 
prise,  cleanliness,  etc.  They  represent  dispositions  which,  at 
first  merely  instrumental  to  the  condition  worths  of  acquisition 
and  consumption,  may  later  be  intrinsically  valued  as  personal 
worths,  or  imputed  as  intrinsic  complementary  worths  to  the 
primary  objects  of  desire,  of  acquisition  and  consumption, 
with  the  latter  phenomenon  that  we  are  here  concerned,  and  an 
analysis  of  the  principles  which  underlie  this  imputation  will  dis 
close  the  essential  features  of  the  process  by  which  mere  con 
dition  worths  acquire  the  coefficient  of  obligation  ;  and  secondly, 
the  way  in  which  this  acquired  meaning  modifies  the  laws  which 
govern  judgments  of  economic  value. 

In  our  study  of  the  Laws  of  Valuation,1  it  was  seen  that 
valuation  moves  between  the  two  limits,  or  thresholds,  which 
mark  off  relative  valuation  from  worthlessness  on  the  one  hand, 
and  from  absolute  worth  on  the  other.  The  lower  threshold 
is  that  minimum  of  the  good  which  has  least  importance,  that 
is  the  smallest  increment  the  existence  or  non-existence  of  which 
calls  out  a  modification  of  worth  feeling.  The  upper  threshold, 
or  the  "  existence-minimum,"  is  that  minimum  of  a  good,  without 
capacity  of  substitution,  the  possession  of  which  constitutes  the 
necessary  presupposition  of  all  further  relative  valuation,  and 
which,  therefore,  itself  has  absolute  value.  In  the  sphere  of 
condition  worths,  with  which  we  are  here  concerned,  the  objects 
corresponding  to  these  thresholds  are  quantities  of  a  physical 
object,  the  desirability  of  which  lies  in  its  capacity  to  satisfy 
certain  fundamental  sense  tendencies.  It  is  in  the  modification 
of  these  thresholds  that  the  acquirement  of  the  feeling  of  obliga 
tion  is  seen  to  modify  the  judgments  of  condition  worth. 

It  is  beyond  question,  I  suppose,  that  the  feeling  of  the 
worth  of  an  object  which  constitutes  the  minimum  of  existence, 
and  which  is  without  capacity  of  substitution,  is  quasi-ethically 
qualified,  has  the  acquired  feeling  of  obligation.  Within  limits, 

1  See  chap,  vi,  pp.  148  ff- 


Values  of  Simple  Appreciation  2  1  s 

\J 

which  we  cannot  here  consider,  self-preservation  is  the  first 
law  of  life.  It  is  the  most  primitive  instinctive  obligation 
which  arises  when  the  arrest  of  conative  tendency  is  carried 
far  enough.  There  may  not  be  a  transgredient  reference  of 
the  feeling  to  an  idea  of  the  self  so  explicit  that  the  feeling 

f  personal  obligations  arises,    but   there   is   at   least   a   trans 
gredient  reference  to  the  whole  system  of  ideal  ends  which  are 
founded  upon  the  preservation  of  life.1     Now,  strictly  speaking 
the  feeling  of  the  intrinsic  value  of  the  object,  the  obligatory 
character  of  the  act  directed    toward    the  acquisition  of  the 

>bject,  falls  off  rapidly  after  the  minimum  of  existence  is  passed. 
Larger  quantities  of  the  good  have  only  relative  instrumental 
value,  for  the  reason  that  they  satisfy  only  secondary  and  ac 
quired  tendencies,  and  have  capacity  for  substitution.  But 
is  precisely  at  this  point  that  acquired  complementary  values 
with  their  coefficient  of  obligation  may  come  in  to  modify  this 
law  anc  to  change  the  threshold  of  relative  valuation.  If, 
for  instance,  the  acquired  complementary  value  of  cleanliness 
has  supervened  upon  the  consumption  of  the  raw  material 

f  desire,  and  if  a  corresponding  disposition  is  formed  with  the 
coefficient  of  obligation,  the  minimum  of  existence—  the  quantity 
of  goods  vhich  has  absolute  value,  and  for  which  indefinite 
sacrifice  wll  be  made—  will  be  increased.  Whatever  is  an 
absolutely  necessary  condition  to  a  certain  irreducible  minimum 
of  cleanliness  will  acquire  absolute  value.  And  in  fact,  as  we 
shall  see  lafer,  when  such  a  disposition  is  referred  to  the  person 
ality,  and  ccquires  personal  worth,  its  obligation  may  be  of 
such  a  character  as  to  lead  to  the  risk  of  life  itself.  On  this 
level  of  simple  appreciation  the  acquirement  of  the  coefficient 

t  obligation  is  seen  to  impute  additional  value  to  objects  of 
mere  condition  worth. 

The  acqurement  of  the  feeling  of  obligation  is,  then,  followed 

mputaticn  of  intrinsic  worth  to  the  object.     This  leads  us 

consideration  of  a  second  phenomenon,  namely,  the  modi- 

ication  of  tht  lower  threshold  of  "  final   utility  "  through  the 

imputation   oi  the   feeling   of   obligation   to   the   object.     This 

)menon  nay  be  more  definitely  described  by  saying  that 


0^  SCem;-at  fi^  Sight'  t0  be  an  illust^ion  of  what  Baldwin 

soost  on  of      {       °f    f-  ""I""1;-     The  System  °f  ideal  ends  founded  u?on  the  pre- 
ition  of  self-prservation  is,  of  course,  not  present  as  a  motive  in  any  sense      The 

sLakK/±f  ^,1^  °ne'S  Hfe  '*'  Uke  ^  imPdlent  ™de  °f  "eehng 
speaking,  efyeOtu.     All  that  is  meant  by  this  statement  is  that,  after  arrest  of 

acquircs  an  intrinsic 


2i6  Valuation:   its  Nature  and  Laws 

an  acquired  obligation,  the  result  of  an  inward  value  movement, 
may  make  intrinsically  valuable  a  quantity  of  an  object  whi 
instrumentally  valueless,  i.e.,  below  the  threshold.     The  pursuit 
of  the  minimum  of  existence  develops  an  instinctive  obligate 
in  the  manner  described.     The  effort  itself,  the  economic  virtue 
of    enterprise,    becomes    intrinsically    valuable.     So    also 
the  most  economic  disposal  of  the  goods,   descnbable  as  t 
economic  virtue  of   thrift.     This  acquired   intrinsic  and  c 
plementary  value   is   then   imputed  as  an  additional  value  to 
the  purely  instrumental  values  of  quantities  of  the  good,  *hic 
would,  according  to  the  law  of  Marginal  Utility,  if  it  were  alone 
determinative,  approach  to  worthlessness.    The  illustration  oJ 
small  coin,  instrumentaUy  valueless  but  intrinsically  of  wortl 
described  in  our  study  of  the  concept  of  the  threshold,  is 
case  in  point.     And  in  general  the  very  real -obligation    c  ten 
largely    instinctive    and   without   explicit    ends,    which   impe. 
men  to  acquire  and  conserve  property  after  it  ceases  to  have 
appreciable  instrumental  value  for  them,  is  of  this  nat 

III    ESTHETIC   APPRECIATION   AS  A  SPECIAL   FOR*  OF  THE 
"  MOVEMENT  TOWARD  ACTIVITY-ITS  MODIFICATICN  c 
DITION  WORTHS 

The  distinctive  attribute  of  the  aesthetic  mode  of  experience, 
in  contrast  to  the  ethical,  is  its  immanency.     Beth  « 
are  intrinsic,  but  while  the  ethical  mode  is  appreciatively  « 
scribable   as   a  deepening   of   the   transgredient  reference, 
esthetic  is  seen  to  be  a  deepening  of  the  immsnental  refer- 
ence.      In    our    analysis    of    appreciative  desertions    in 
first  chapter,  we  found  it  to  be  characterised  by  repose  in  ti 
obiect   and  increase   of   expansive   suggestions   o:   the 
From  the  standpoint  of  the  genetic  study  of  vake  movements 
we  subsumed  the  movement  toward  the   esthete  under  th. 
general  type  of  movement  toward  activity.     It  remains  now 
fo  correlate  these  two  points  of  view,  to  show  hew  the  changes 
in   function   and   content   involved   in   the   moment   toward 
activity  are  the  determining  conditions  of   this  repose  in  the 
object  and  expansion  of  feeling:    how  in  fact*  one  at 
repose  and  activity  may  be  combined,  and  how  tiis  ™™binat  ™ 
creates  complementary  worth  which  is   then  mputed 

n  the  case  of  our  study  of  the  ethical  qualification  of 


Values  of  Simple  Appreciation  2,7 

condition  worths,   through   the  acquirement  of  the  obligator 
coefnc.en      so   our  study  rf    ^   ^^  atio™ 

q 


ao 

begm  Wlth  an  attempt  to  define  fte          ho,q  equivakn  s 

of  the  appreciative  description.     And  here,  again    we  find  th, 

feaeTngPo°f    eo"tPreSented-     *"    ^^   ^™"   ™" 
ng,  of  conation,  or  01  cogmtion  ?     The  answers  are  varied 

and,  again  I  am  inclined  to  think,  not  necessarily  contradltorv' 

seTt  V^V^'  ^^fy,  the  .esthetic  attitude  wi?  be' 
seen  to  be  a  mode  definable  in  aU  three  terms.     Let  us  begin 
however,  wifl,  an  examination  of  certain  attempts  to  find  the 

0    T***  aPPredati™  in  f-ling,  as  Determined  by 
arrangement  of  sensational  and  presentational 


ntent 


I.  Analysis  of  the  Esthetic  Mode  of  Appreciation. 
In   each   of   these   attempts    the   question   is    asked    Whv 
do  we  have  this  specific  ordering  or  rlrrangemLt  of  conTent 

Ctl 


ech  ca     th,          e   *StetC   exPerience?      And  m 
ordeTtW  thePy  'S'     uWe  neg'eCt  differen«s  in  terminology 

he  greatest    et't^T  ,     ^  gFeateSt  am°Unt  of  Pleasure  % 
length   of   time   without   satiety.     But   while   all 

eckon  with  this  conceptual  abstraction  of  quantity  of  pleasure 
dement  whTchwl  "^  ""   yS'S'  they  Sh°W'  I  think'  a  common 


rranSha1^  ^/M"W.  A'«.  «"d  .Esthetics,  p.  335 
Groos,  Z?^«/^«r^  Gen^,  Giessen,  19^2"  chap.  ,. 


2 ,8  Valuation:   its  Nature  and  Laws 

pleasure   through   "  playful  illusion."     Here,   again,   we  have 
an  appreciative  differentiation  of  esthetic  pleasures  from  oth  rs 
in  that  both  these  terms,  play  and  illusion,  have  a  funct,ona 
meaning.     In  his  earlier  psychological  works,  wh.ch  were  c< 
cerned  largely  with  the  genesis  of  playful  illusion  as  a  significant 
function,  Uos  was  indined  to  account  for  the  pleasure  of  both 
the  "  play  "  and  "  illusion  "  aspects  by  connecting  it  wit 
biologkal  utility  ,-as  a  pleasure  arising  from  the  mere  activity  c 
these  instincts  themselves.     But  in  his  later  work,'  which  ,s  con 
cerned  more  with  the  analysis  of  the  aesthetic  Attitude  itself 
the  emphasis  is  put  upon  the  rearrangement  of  the      eld .< 
content      Here  the  idea  is  developed  that  the  playful  dwell 
in  "  inner  imitation  "  upon  the  impressions  of  things,  the 
porary  identification  of  the  self  with  other  things  or  persons 
with  its  moment  of  conscious  self-illusion,  are  the 
conditions  which  make  possible  the  fusions  and  compl, 
of  sensational  and  associational  content  out  of  whict ,  compk 
mentary  pleasures  arise.     The  changes  m  content  which  th 
take   place    are   twofold.      There   is,   first,   a  widening  .of   the 
ntd  oPf  experience,  through  inclusion,  in  the  intuit,ve  moment 
of  fusion,  of  the  complementary  associations  ;    and,  S  condly 
an  elimination  of  all  elements  which  are  inharmonious  with  the 
total  experience,  and  which,  by  their  intensity  or  power  to  , 
the  attention,  would  lead  to  judgments  and  value  ™™«** 
in  short,  all  illusion-disturbing  moments      Conversely 
be  added,  an  arrangement  of  content  winch  brm£s  about  th, 
fusion,  and  eliminates   these  disturbing  moments    create 
Aesthetic  illusion.     Here,  again,  the  essential  features  ar =  the 
widening  of  the  field  of  experience,  and  the  repose  in    t,  made 
possible  through  elimination  of  the  elements  which  wo, 

UPAgliSheld  that  the  essential  characteristic  of  the  .sthetic 
experience  is  that  it  combines  in  a  unique  way  the  dish, 
2££  of  activity  and  repose.     A  somewhat  incomplete  con- 


1  Groos,  Der  asthetische  Genuss. 

-  Puffer,  The  Psychology  of  Beauty,  chap.  III. 


Values  of  Simple  Appreciation  219 

equilibrium,  and  if  no  one  breaks  loose,  so  to  speak   to  become 
a  fundamental,  either  in  the  form  of  desire  or  judgment    we 
have  repose  of  desire  and  judgment,  and  consequently  no  value 
movement.     But  this  equilibrium  of  impulses,  in  so  far  as  it  is 
onditioned  by  arrangement  of  content,  is  brought  about  by 
a  diffusion  of  stimulation,  i.e.,  by  a  widened  field  of  well-balanced 
stimulations.     To  the  technical  means  employed  in  the  creation 
this  widened  field  of  diffused  and  balanced  stimulation    we 
have  occasion  to  refer  later.     The  important  point  here  is 
the  definition. 

With  this  sketch  of  the  three  formulations  before  us   it  is  not 
fficult  to  define  the  common  element.     This  is  clearly  the  con 
cept  of  the  widened  ground  of  diffused  stimulation,  the  balance 
[  impulses,  so  that  no  one  shall  constitute  an  illusion-disturbing 
moment  and  lead  to  readjustment  in  a  new  value  movement 
the    consequent    repose   of    conation    in    the    object    and    the 
expansion  of  feeling  which  goes  with  it.     The  ordering    rear 
rangement  of  content  characteristic  of  the  esthetic  experience 
is    therefore,  in  the  service  of  the  deepening,  or  enhancement 
that    fundamental    mode    of    worth    experience    which    is 
appreciatively  described  as  the  immanental  reference    the  ex 
pansive  suggestion  of  the  feeling.     But  this  repose  of  conation 
i  its  expansion  of  feeling,  is,  as  the  preceding  analysis  made 
clear,  conditioned  by  a  characteristic  change  in  the  content  of 
experience,  by  a  reconstruction  of  this  content  in  such  a 
manner  that  the  foreground  of  consciousness  is  taken  up  with 
secondary  and  subsidiary  activities  of  sensation  and  presentation 
play  "  of  impressions  which  inhibits  the  fundamental 
mative  tendency  with  which  they  are  associated.     What  the 
chnique  of  this  reconstruction  of  the  widened  field  of  attention 
tay  be  we  have  yet  to  consider,  but  it  is  apparent  that  it  must 
J  o  i  the  general  nature  of  substitution  of  a  multiplicity  of 
subsidiary  activities  for  the  fundamental  conative  tendency 
Ihis  view  of  the  characteristic  modification  of  the  content 
3  aesthetic  experience  is,  however,  incomplete  without  a 
corresponding  analysis  of  the  aesthetic  as  attitude.     Such  change 
ntent  presupposes  a  change  in  the  presuppositions  of  the 
feelmg-a ttitudc.      Thus   far  we    have    considered   the   aesthetic 
node  only  as  a  mode  of  feeling.     We  must  now  consider  it  in 
its  genetic  relations  to  other  worth  attitudes,  as  a  mode  of  cona 
tion  and  cognition. 


220  Valuation:   its  Nature  and  Laws 


(b)  Analysis  of  the  Cognitive  Presuppositions  of  the  ^Esthetic  Mode 
— Interpolation  in  the  Vital  Series. 

The  answer  to  this  problem  is  possible  only  through  the  em 
ployment  of  the  same  method,  the  attempt  to  locate  the  mode 
in  a  vital  series  of  attitudes,  used  in  the  analysis  of  the  sense  of 
obligation.  When  thus  viewed,  the  aesthetic  is  seen  to  be  not 
desireless  appreciation,  as  it  is  sometimes  held,  nor  feeling  with 
mere  presentations  as  its  presuppositions.  The  aesthetic  atti 
tude  is  seen  to  have  both  conative  and  cognitive  presuppositions. 
The  object  which  now  presents  this  widened  field  is  in  the 
first  place  an  object  of  desire  and  judgment;  the  impulses 
which  are  balanced  in  the  aesthetic  field  were  first  of  all  impulses 
subsidiary  to  some  fundamental  conative  tendency  directed 
toward  the  object.  When  the  movement  toward  activity 
takes  place— when  the  object  as  desired  sinks  into  the  back 
ground,  is  distanced,  the  fundamental  desire  becomes,  it  is 
true,  dispositional,  but  it  is  this  presence  as  a  desire-disposition 
which  gives  the  depth  of  immanental  reference  in  the  feeling 
of  repose — as  we  have  already  seen  in  the  illustration  of  the 
appreciation  of  feminine  beauty.1  The  aesthetic  as  a  feeling 
mode  lies,  then,  midway  between  the  passivity  of  sensation 
feeling  and  explicit  desire,  the  distinction  of  feeling  and  desire 
being  only  relative. 

Nor  is  the  aesthetic  pure  "  presentation  -  feeling  "  without 
reference  to  reality.  Here,  again,  the  method  of  interpolation 
shows  it  to  be  a  mode  of  thought  midway  between  the  judg 
ment  of  existence  and  the  judgment  of  non-existence.  Before 
the  movement  toward  the  aesthetic,  the  object  as  object  of 
desire,  actual  or  possible,  was  the  object  of  explicit  judgments 
of  existence  or  non-existence.  With  the  distancing  of  the  object, 
and  the  transformation  of  explicit  desire  into  feeling,  there  is 
a  change  in  presuppositions.  The  aesthetic  feeling  is  no  longer 
a  judgment -feeling,  neither  is  it  merely  a  "presentation" 
feeling,  but  rather  an  assumption-feeling.  In  aesthetic  feeling 
the  existence  of  the  object  is  always  assumed,  unless  explicit 
judgment  of  non-existence  supplants  the  assumption. 

Here  is  to  be  found  the  element  of  truth  in  the  description 
of  the  esthetic  as  conscious  self -illusion.  This  internally  con 
tradictory  term  is  a  somewhat  bungling  way  of  describing  a 

1  Chap,  ill.,  p.  71 


Values  of  Simple  Appreciation  221 

real  mode  of  experience,  that  intermediate  stage  of  adaptation 
between  explicit  judgment  and  explicit  judgment.  It  is  an 
assumption  of  reality  which  will  suffice  for  the  temporary  repose 
in  the  object,  and  for  the  realisation  of  all  the  values  which 
that  object  may  have  for  conation  when  desire  is  explicit  and 
its  presuppositions  are  existential  judgments.  But  the  illusion 
may  be  said  to  be  conscious,  and  to  be  self-illusion,  only  in  the 
sense  that  the  assumption,  although  possessing  the  coefficient  of 
reality  which  it  has  by  virtue  of  its  actualisation,  in  its  own 
special  way,  of  a  conative  disposition,  is  under  the  control 
of  the  habits  of  judgment  and  implicit  assumptions  created  in 
former  experience,  so  that  ordinarily  any  tendency  of  the  as 
sumption  to  pass  over  into  explicit  judgment  is  inhibited.  It 
is  not  a  conscious  self-illusion  in  the  sense  that  its  control  is 
conditioned  by  a  conscious  reference  to  the  self. 

2.  The.  Origin  of  the  Movement  toward  Activity  and  the  Msthetic 
Attitude — Its  Individual  and  Sub-Social  Character. 

The  preceding  analysis  of  the  structural  and  functional 
modifications  characteristic  of  the  aesthetic  attitude  and  ex 
perience  justifies  its  subsumption  under  the  general  type  of 
value  movement  described  as  movement  toward  activity. 
In  the  first  place,  to  consider  the  second  aspect  first,  there 
is  the  substitution  of  assumption  for  judgment  and  the  dis 
tancing  of  the  object  or  its  detachment,  from  immediate  desire, 
characteristic  of  the  semblant  mode.  With  it,  in  the  second 
place,  comes  the  enhancement  of  worth  through  the  increased 
activity  of  the  subsidiary  tendencies  of  sensation,  presentation, 
etc.,  characteristic  of  the  imaginative  attitude.  With  this 
account  of  the  nature  of  the  aesthetic  mode  comes  the  problem  of 
its  genesis  as  a  mode  of  simple  appreciation.  The  problem  is  in 
reality  twofold  :  (a)  what  are  the  conditions  of  the  movement 
toward  activity  itself  ;  and  (b)  what  is  the  origin  of  the  specific 
form  of  activity,  with  its  quality  of  detachment,  which  character 
ises  the  aesthetic  ? 

(a)  Conditions  of  Movement  toward  Activity  in  its  Pre- /Esthetic 

Form. 

As  the  origin  of  the  inward  movement  toward  the  attitude 
of  obligation  is  thought  by  many  to  be  wholly  social,  so  also 


222  Valuation:   its  Nature  and  Laws 

there  is  a  similar  tendency  to  account  for  the  genesis  of  the 
aesthetic  in  merely  social  terms.  The  study  of  the  beginnings 
of  art  gives  a  certain  plausibility  to  this  view  in  that  much  of 
the  rearrangement,  reconstruction,  of  expression  of  emotions 
and  objects  of  emotion,  seems  to  be  directly  connected  with 
the  motive  of  securing  social  consent  in  emotional  expression, 
and  thus  of  enhancement  of  sympathetic  participation.  The 
element  of  order  introduced  into  crude  emotional  expression 
is  accordingly  conceived  to  have  a  purely  social  origin.  Never 
theless,  while  social  sympathy  may  enhance  aesthetic  worth, 
and  while  the  motive  of  social  consent  may  determine  many 
forms  of  art,  both  the  movement  toward  the  aesthetic 
itself,  and  the  reconstruction  of  content  involved  in  that  move 
ment,  must,  it  seems  to  me,  like  the  movement  to  the  ethical, 
be  sub-social  and  individual  in  their  origin.  As  a  value  move 
ment  in  the  individual,  its  functional  conditions  must  be  looked 
for  in  the  general  conditions  of  value  movement,  arrest  of  the 
fundamental  through  distancing  of  the  object  of  desire,  or 
through  satiety  and  substitution  of  subsidiary  tendencies. 

In  his  valuable  work  on  The  Origins  of  Art T  Him  has  em 
phasised  this  aspect  of  the  origin  of  art.     In  the  first  place, 
he  has  distinguished  in  a  very  satisfactory  way  between  the 
question   of  the   functional   genesis   of   the   aesthetic   psychosis 
itself   and   the  question   of  the  historical  origin   of  particular 
forms   of   artistic   expression.     He   has   seen   clearly   that   the 
various    activities   in   which   the   aesthetic   specifies   itself    (the 
concrete  origins  of  art  through  the  secondary  social  motives 
of  conveying   information,   display,   and  self-exhibition,   erotic 
and  martial  stimulation,  etc.)   do  not  explain  the  antecedent 
emotional  psychosis  which  they  presuppose  and  express.     These 
are   rather   activities,   emotional   and   volitional,   in   which   an 
antecedent  need  of  emotional  expression  specifies  itself.     The 
art  impulse  itself,  or,  in  our  terms,  the  movement  toward  ac 
tivity,  must  be  sought  in  deeper  and  more  general  functional 
causes.     These  he  finds  in  the  tendency  to  the  development  of 
secondary   activities   when   the   primary   is   subject   to   arrest. 
From  the  standpoint  of  description  in  terms  of  pleasure  and  pain, 
art  serves  either  as  the  reliever  of  the  pain  which  comes  from 
thwarted  conative  tendencies,  or  else  as  a  means  of  enhancing 
the  pleasure  of  a  fundamental  already  pleasurable,  and  of  post 
poning  the  moment  of  satiety. 

1  Him,  The  Origins  of  Art,  Macmillan,  1900,  chaps.  I,  iv,  and  vi. 


Values  of  Simple  Appreciation  223 

Space  will  not  permit  a  detailed  study  of  the  facts  of  individual 
and  social  psychology  by  means  of  which  this  view  is  substantiated 
—how  sorrowful  emotions,  and  even  painful  sensations,  may 
really  be  the  relief  of  deep-seated  arrest  of  activity  through  the 
substitution  of  secondary  activities,  and  may  thus  attain  a  posi 
tive  worth  ;  how  in  the  social  expressions  of  grief,  in  the  wailing 
feats  of  certain  savages,  and  in  the  vocero  in  general,  not  only 
is  sorrow  relieved,  but  positive  joy  of  a  new  sort,  joy  in  expression, 
in  activity,  is  generated  ;  how  in  their  orgies  the  Maenads, 
by  noise,  roaring  and  loud  cries,  by  frenetic  dance  and  wild 
actions,  even  by  tortures,  strove  to  preserve  and  recover  the 
faded  sense  of  life  which  ever  baffled  their  exertions.  But  enough 
has  been  said  to  show  that  much  of  the  movement  toward  ac 
tivity,  out  of  which  artistic  activity  is  ultimately  born,  is  due 
to  the  painful  arrest  of  conative  tendency.  The  object  of  the 
fundamental  conative  tendency,  of  desire  and  sorrow,  is  dis 
tanced,  the  full  realisation  of  its  existence  or  non-existence  is 
supplanted  by  a  vague  assumption  which  gives  but  an  undertone 
to  the  whole  experience,  while  the  activity  of  the  subsidiary 
tendencies  usurps  the  foreground  of  consciousness. 

This  is  indeed  one  mode  of  origin  of  the  movement  toward 
activity,  but  it  does  not  include  that  movement  toward 
activity  which  results  in  enhancement  of  the  worth  of  an 
object  already  desirable,  i.e.,  possessing  the  capacity  of  satisfying 
a  fundamental  desire.  The  movement  in  this  case  is  to  be 
understood  also  as  an  adaptation  alter  arrest,  but  here  the  arrest 
of  conative  tendency  appears  at  another  point.  We  have  else 
where  seen  that  when  once  a  fundamental  conative  tendency 
has  through  repeated  arrests  concentrated  about  it  subsidiary 
tendencies,  a  too  immediate  or  exclusive  satisfaction  of  the 
fundamental  leads  to  satiety  in  which  the  subsidiary  tendencies 
are  left  unsatisfied.  The  movement  toward  activity  of  the 
subsidiary  tendencies  has  as  its  motive  in  many  cases  the  dis 
tancing  of  the  object  of  desire  in  the  interest  of  continuity  of 
valuation  of  the  same  object  in  that  to  its  worth  is  added  the 
complementary  value  of  the  secondary  activities. 

Of  this  character  are  the  form-qualities  added  to  the  economic 
activities  of  acquisition  and  consumption  already  discussed  ; 
and  the  artistic  reconstruction  of  objects  of  use  so  that  aesthetic 
repose  in  them  is  possible  seems  to  be,  from  this  point  of  view, 
of  the  same  general  character — the  distancing  of  the  object  as 
an  object  of  utility,  the  feelings  of  instrumental  worth  being, 


224  Valuation:   its  Nature  and  Laws 

as  we  have  seen,  by  their  very  nature,  quickly  subject  to  the 
law  of  limiting  value.  Finally,  the  distancing  of  the  object 
of  sex,  the  movement  toward  mere  activity  in  both  coquetry 
and  idealisation,  whatever  its  biological  origin  may  be,  ap 
pears,  as  individual  value  movement,  to  be  of  the  same  char 
acter.  It  may  be  said,  therefore,  that  the  origin  of  the  movement 
toward  activity  is,  as  has  already  been  shown  in  our  general 
study  of  value  movements,  to  be  found  in  the  arrest  of  funda 
mental  conative  tendency. 

(b)  The  Special  Differentia  of  the  ^Esthetic. 

But  it  may  be  properly  objected  that,  while  this  is  a  true 
account  of  the  origin  of  the  general  type  of  "  movement  toward 
activity,"  that  while  it  is  true  that  the  causes  of  this  movement 
are  sub-social  and  individual,  nevertheless,  these  movements, 
as  such,  are  not  necessarily  movements  to  the  aesthetic  attitude. 
All  movements  to  the  aesthetic  are  movements  toward  activity, 
but  not  all  movements  toward  activity  are  aesthetic.  Granting 
the  inherent  necessity  of  this  movement  toward  activity,  why 
does  the  movement  not  stop  with  crude  and  primitive  motor 
and  emotional  expression — which  would  suffice  to  occupy  the 
foreground  of  consciousness,  to  still  the  pain  of  arrest,  and  to 
enhance  the  faded  sense  of  life,  why  does  the  movement  de 
velop  into  that  ordered  activity  which  characterises  the  aesthetic  ? 
We  have  accounted  for  the  pre-aesthetic  movement  toward 
activity,  out  of  which  the  aesthetic  attitude  itself  is  ultimately 
born,  but  we  have  not  as  yet  shown  why  this  activity  becomes 
aesthetic. 

It  is  clear  that  in  this  question  we  have  the  crux  of  the 
problem,  and,  indeed,  of  all  attempts  to  explain  the  aesthetic. 
Nor  is  the  answer  to  the  question  altogether  easy.  It  is  not 
surprising  that  many  have  finally  looked  for  the  point  of  differ 
ence  between  unaesthetic  free  play  of  emotional  expression  and 
of  activity  of  the  imagination,  and  ordered  aesthetic  activity,  in 
social  rather  than  individual  causes.  But,  important  as  the 
factor  of  community  of  activity  and  social  consent  may  be  in 
fixing  the  forms  of  activity,  and  in  many  cases  perhaps  originating 
them,  it  is  still  possible,  I  think,  to  show  that  this  introduction 
of  the  element  of  order,  this  reconstruction  of  the  content  of 
experience,  is  inherent  in  the  movement  toward  activity  itself 
as  a  stage  in  individual  worth  experience.  If  we  grant  the  truth 


Values  of  Simple  Appreciation  225 

of  our  initial  definition  of  the  aesthetic  mode  as  a  modification  of 
the  immanental  reference  of  repose;  if,  further,  we  grant  that 
this  repose  in  the  object  of  desire,  this  tendency  to  prolong  the 
appreciation  of  the  object  is  fundamental,  and  can  be  achieved 
only  by  detachment  of  the  object  from  immediate  desire,  in 
the  semblant  mode,  the  question  is  practically  answered.  It  is 
precisely  because  this  immanental  reference,  this  repose  in 
the  activity  is  impossible  without  such  ordered,  balanced,  and 
harmonious  activity,  because  without  the  element  of  order, 
the  illusion-disturbing  moments,  the  desires  and  judgments 
spring  again  into  being,  and  lead  to  new  value  movements, 
that  the  movement  toward  activity,  to  become  aesthetic,  must 
include  this  rearrangement  of  activity.  The  element  of  form, 
of  order,  must,  therefore,  be  looked  upon  genetically  as  purely 
instrumental  to  the  acquirement  of  aesthetic  worth,  which  is 
ultimately  the  affective-volitional  meaning  of  content. 

3.     ^Esthetic    Characterisation    of    Activities    and    Objects — The 
Function  of  the  Element  of  Order  in  Creating  ^Esthetic  Values. 

The  role  of  the  element  of  order  in  prolonging  the  appre 
ciation  of  objects,  in  enhancing  the  immanental  worth  of  objects 
through  detachment  from  immediate  desire,  can  best  be  shown 
by  reference  to  artistic  activity,  more  especially  to  the  primitive 
arts  where  the  direct  relation  of  the  art  activity  to  the  funda 
mental  desires  and  instincts  out  of  which  it  emerges  has  not  yet 
been  severed,  where  aesthetic  activity  has  not  yet  become  an 
end  in  itself,  an  independent  object  of  value.  Reconstruction 
of  content,  we  have  seen,  may  be  both  perceptual  and  ideal, 
and  aesthetic  reconstruction  may  be  of  both  types.  Ordering 
of  our  motor  activities,  as  in  the  dance,  ordering  of  our  visual 
or  auditory  experiences,  constitutes  perceptual  reconstruction, 
while  the  aesthetic  characterisation  of  things  or  persons  as,  for 
instance,  in  literature,  would  be  ideal  construction.1 

Let  us  begin,  then,  with  the  most  primitive  and  simple 
form  of  aesthetic  characterisation,  the  perceptual  reconstruction 
of  activities  of  emotional  expression.  Ethnology  shows  us 

1  /Esthetic  reconstruction  should  be  distinguished  from  other  types  of  construction, 
perceptual  and  ideal,  illustrations  of  which  we  have  already  considered.  Perceptual 
reconstruction  which  facilitates  conative  tendency,  ideal  construction  of  objects  and 
dispositions  which,  through  instrumental  judgments,  extends  the  range  of  conation,  may 
also  give  rise  to  new  economic  and  ethical  worths.  But  in  the  case  of  these  types  of 
reconstruction,  conation  is  explicit,  the  worth  moment  is  transgredient ;  while  in  esthetic 
•construction,  conation  is  dispositional  and  the  worth  moment  is  immanental. 


226  Valuation:   its  Nature  and  Laws 

that  primitive  dances  are  invariably  the  reconstructions  of  the 
bodily  expressions  of  concrete  emotional  states  arising  out  of 
definite  concrete  instincts.  By  this  it  is  meant  that  primitive 
dances  are  always  erotic,  martial,  funereal,  or  religious.  Such 
reconstructions  always  have  two  aspects,  the  dramatic  and 
the  rhythmic.  They  show  forth  in  gesture  and  pantomime 
the  motor  activities  in  which  these  instincts  are  expressed, 
and  add  to  these  a  certain  order  and  rhythm.  Both  these  ele 
ments  of  order  involve  a  certain  reconstruction  of  crude  emo 
tional  expression,  and  they  constitute  the  fundamental  elements 
in  all  aesthetic  characterisation.  The  element  of  rhythm,  besides 
its  social  function  of  securing  participation  or  social  consent  in 
movement,  has  the  effect  of  keeping  the  emotion  and  its  ex 
pression,  whether  it  be  martial,  erotic,  or  religious,  on  a  high 
level  for  a  comparatively  long  period,  to  the  exclusion  of  the 
desire  for  the  object.  The  object  is  distanced  and  the  funda 
mental  conation  becomes  dispositional.  This  rhythm,  usually  of 
the  form  of  advance  and  retreat,  of  affirmation  and  arrest  of  ex 
pression,  produces  an  equilibrium  of  impulses  which  prevents 
the  fundamental  tendency  from  breaking  forth  into  overt  action. 

On  the  other  hand,  the  dramatic  elements  of  the  emotional  ex 
pression  are  ordered  in  a  cumulative  dynamic  "  movement-form  " 
which  becomes  fixed  as  the  characteristic  order  of  the  dance. 
It  is  generally,  as  in  the  erotic  or  war  dance,  a  conventionalised 
"movement-form"  of  the  gradual  rise,  and  also  the  decline,  of 
the  fundamental  emotion.  Beginning  with  movements  and 
gestures,  which  are  at  first  rather  suggestive  than  fully  expres 
sive,  it  passes  on  by  gradual  transitions  to  more  and  more  overt 
and  furious  expression,  until  sometimes  the  distinction  between 
art  and  reality  is  finally  lost.  Here,  again,  besides  its  function 
in  securing  social  participation,  this  dramatic  movement-form 
introduces  a  principle  of  serial  order  into  phenomena  by  means 
of  which  there  is  individual  acquirement  of  worth.  Into  the 
aesthetic  experience  is  carried  over  a  conventionalised  form  of 
the  very  activities  which  constitute  the  actual  instinct  from 
which  the  aesthetic  has  arisen  as  a  value  movement  toward 
activity. 

The  significant  and  strictly  formal  aesthetic  factor,  as  the 
above  analysis  shows,  is,  of  course,  the  balance  of  impulses 
which  detaches  the  object  from  immediate  desire.  It  constitutes 
a  special  application  in  the  sphere  of  aesthetic  reconstruction 
of  that  general  principle  of  perceptual  and  ideal  reconstruction 


Values  of  Simple  Appreciation  227 

discussed  in  a  previous  chapter  under  the  head  of  the  law  of 
complementary  values.  But  in  addition  to  this,  there  is  also  the 
application  of  the  principles  of  the  "  total  series  "  and  of  "  end 
feelings/'  characteristic  of  serial  phenomena.  In  the  case 
before  us,  the  dynamic  movement-form  of  the  passions  in  ques 
tion  is  individualised  into  a  total  series,  and  the  gradation  always 
has  reference  to  culmination  in  certain  final  expressions  and 
feelings  which  give  character  to  the  whole. 

The  role  of  these  formal  factors  is  evident ;  but  it  is  im 
portant  to  note  that  the  detachment  of  the  object  is  only  partial, 
that  in  these  primitive  dances,  in  which  the  object  of  conation 
is  still  not  very  remote  from  desire,  and  where  the  fundamental 
desire  or  system  of  desires  is  as  yet  scarcely  dispositional,  the 
'•'  content  "  factor,  the  dynamic  movement-form  of  emotional 
expression,  has  not  reached  the  point  of  aesthetic  reconstruc 
tion  which  it  undergoes  in  more  developed  artistic  activity.  It 
is  still  partly  extra-aesthetic.  When,  however,  in  music  and 
drama,  as  they  become  historically  differentiated  from  the  dance, 
a  still  greater  distance  is  put  between  the  object  and  the  funda 
mental,  the  principle  of  equilibrium  is  introduced  into  the 
movement-forms,  and  the  conative  element  becomes  still  more 
dispositional,  remaining  present,  in  fact,  as  for  instance  in  music, 
only  in  the  form  of  a  dominant  mood.1 

A  word  is  required  concerning  the  perceptual  aesthetic  recon 
struction  of  static  objects  in  which  rhythm  and  dynamic  move 
ment-form  do  not  enter,  i.e.,  aesthetic  perception  as  we  have  it 
in  the  fine  arts.  For  aesthetics  pure  and  simple,  this  is  perhaps 
the  most  important  field  of  analysis,  but  since  our  interest  is 
confined  to  the  problem  of  the  functional  relation  of  the  aesthetic 
to  other  worth  attitudes,  and  since  we  may,  therefore,  expressly 
disclaim  all  desire  for  completeness  in  analysis,  a  mere  sketch 
must  suffice.  Here,  again,  we  have  individuation  of  the  object 
in  the  interest  of  repose.  The  two  significant  factors  are  the 
arrest  and  repose  of  the  fundamental  through  equilibrium  of 
subsidiary  impulses  and  the  individualisation  and  segregation 
or  detachment  of  the  content.  To  consider  the  first  moment 
in  the  light  of  the  results  of  analytical  psychology,  it  is  im 
portant  to  realise  that  here  also  we  are  concerned  with  a  re 
construction  of  activity,  but  in  this  case  not  of  the  motor  ac 
tivity  of  emotional  expression,  but  of  an  inner  activity  of  at 
tention.  Here  the  activity  controlled  is  not  emotional  expression, 

1  See  chap.  V,  pp.  122-3. 


228  Valuation:   its  Nature  and  Laws 

but  attention.  The  movement  toward  activity  in  this  case  is 
toward  what  has  been  described  as  a  "  play  with  impressions," 
wherein  the  objects  of  sensation  and  perception  are  so  ordered 
into  groups  and  "  form-qualities,"  that  the  separate  impulses 
of  attention  are  balanced,  attention  being  distributed  over  a 
field  so  widened  that  no  portion  of  the  presented  whole  calls 
forth  feeling  and  desire  of  such  immediacy  and  intensity  as  to 
break  the  unity  of  the  whole, — to  destroy  the  illusion  and  the 
repose  and  to  lead  to  a  new  value  movement. 

The  principles  of  reconstruction  of  the  aesthetic  field  in  the 
"  non-serial "  arts  are,  mutatis  mutandis,  the  same  as  in  the  serial 
arts,  namely,  the  principles  of  contrast,  of  "  total  series,"  and  of 
"end  feelings";  but  here  these  principles  are  applied  to  the 
activities  of  attention  in  such  a  manner  as  to  make  possible  the 
widened  field  of  attention  and  the  consequent  expansion  of  feeling. 

This  widened  field  of  attention  is  well  illustrated  in  pic 
torial  art.  Here  the  formal  conditions  of  equilibrium  and 
repose  are  to  be  found  in  the  disposition  of  spaces  and  the 
composition  of  the  colour  elements.  The  values  of  these  ele 
ments  of  order  are  in  this  respect  merely  instrumental  to  the 
equilibrium  of  the  attention-activity  sought.  The  colours 
used  and  the  objects  portrayed  have,  of  course,  their  intrinsic 
affective  meaning,  through  association,  but  this  refers  to  pre 
supposed  conative  tendency.  In  so  far  as  the  disposition  of 
the  formal  elements  as  such  is  concerned,  it  is  controlled  by 
special  forms  of  the  fundamental  laws  of  aesthetic  construction 
already  considered.  To  this  end  we  have  the  use  of  colour 
contrasts  and  contrasts  of  masses,  the  principle  of  the  "  series  " 
which  in  this  case  consists  in  the  ordering  of  the  lights  and  shades 
in  a  gradually  diminishing  scale,  with  the  high  light  in  the  centre 
of  attention,  and,  finally,  the  use  of  the  principle  of  "  end  feeling  " 
which  in  this  case  consists  in  such  a  balance  of  elements  that 
the  attention  always  returns  to  a  central  point  of  interest 
equivalent  to  the  tonic  in  musical  composition. 

The  important  point  for  our  study  is,  however,  not  the 
detailed  analysis  of  this  formal  element  of  order,  but  rather 
its  relation  to  the  worth  element  in  the  aesthetic  experience. 
The  object  of  the  worth  feeling  is  always  primarily  the  content 
expressed  by  the  presentations  thus  ordered,  although  there  may 
of  course  be  a  secondary  worth  judgment  upon  the  instrumental 
value  of  the  form  as  such,  its  adequacy  to  represent  or  express 
the  object,  or  to  secure  and  retain  the  aesthetic  repose.  This 


Values  of  Simple  Appreciation  229 

content  is  always  the  object  of  possible  desire  or  aversion. 
That  is,  if  it  were  judged  to  exist  or  not  exist  it  would  call  out 
actual  desire  or  emotion.  But  in  the  aesthetic  state  these  judg 
ments  are  inhibited,  and  for  them  are  substituted  assumptions. 
From  the  standpoint  of  worth  theory,  then,  the  formal  element 
of  order  is  significant  only  as  a  means  of  securing  repose  in 
the  object  (or  content)  which,  when  unaesthetically  experienced 
is  the  object  of  explicit  desire  and  judgment.  What  are  tech 
nically  described  as  the  "  values  "  of  the  particular  elements  in  a 
work  of  art  (a  painting,  for  instance)  are,  therefore,  relative 
instrumental  values,  instrumental,  that  is,  to  the  enhancement 
of  the  primary  aesthetic  worth,  the  repose  or  immanental 
reference. 

Consequently,  when,  by  a  movement  toward  activity,  an 
object  acquires  complementary  aesthetic  values,  it  is  not  to  be 
understood  that  this  acquired  value  is  the  resultant  of  an  addition 
of  the  values  of  the  formal  "  elements,"  but  that  it  is  an  en 
hancement  of  the  immanental  reference  of  the  worth  feeling,  an 
expansion  of  the  feeling  over  the  object  as  a  whole.  The  desire 
for  the  object  of  the  assumption  is  present  dispositionally  and 
actually  in  the  form  of  the  affective  abstract,  sentiment  or  mood. 
The  adequacy  of  the  formal  element  is  determined  by  the  senti 
ment  or  mood,  thus  enhancing  the  worth  or  affective-volitional 
meaning  of  the  object. 


4.    Modification   of   Condition    Worths   through   Acquirement   of 
Complementary  ^Esthetic  Values. 

With  this  account  of  the  nature  and  genesis  of  the  aesthetic 
attitude  as  a  form  of  the  movement  toward  activity,  we  are 
now  in  a  position  to  apply  our  results  to  the  principal  problem 
of  our  study,  that  is,  to  the  determination  of  the  manner  in 
which  the  aesthetic  attitude  modifies  our  primary  judgments  of 
appreciation,  more  especially  our  economic  judgments.  The 
manner  in  which  simple  condition  worths  acquire  quasi-ethical 
qualification  has  already  been  studied.  It  remains  to  indicate 
how  feelings  of  condition  worth  become  aesthetically  qualified, 
and  how  this  qualification  modifies  economic  worth  judgment 
and  its  laws. 

In  order  to  understand  this  problem  rightly,  it  is  necessary 
to  recall  our  distinction  between  the  value  movement  in  simple 


230  Valuation :   its  Nat^tre  and  Laws 

appreciation  and  the  completed  value  movement  to  a  new 
ideal  object,  or  end.  In  the  first  case  the  deepening  of  the 
transgredient  or  immanental  reference  of  the  feeling,  as  the 
case  may  be,  constitutes  merely  a  complementary  value  added 
to  the  primary  object  or  desire.  In  the  completed  value  move 
ment,  on  the  other  hand,  the  original  object  of  desire  is  lost  sight 
of,  and  a  new  object,  to  the  construction  of  which  the  acquired 
meaning  was  germinal,  takes  its  place.  The  object  or  activity 
is  independently  valued.  Now  the  reconstruction  of  objects 
and  activities  which  is,  properly  speaking,  artistic,  and  which 
constitutes  the  aesthetic  in  the  full  sense  of  the  term — such  re 
constructions  as  we  have  described  in  the  preceding  paragraphs, 
are  of  this  conscious  type.  The  complementary  value  of  "  re 
pose  in  activity,"  beauty  in  its  various  modifications,  is  sought 
for  its  own  sake,  and  the  reconstruction  of  the  object  is  instru 
mental  to  this  ideal  end.  But  prior  to  this  stage  of  the  aesthetic, 
there  is  that  phase  of  movement  toward  the  aesthetic,  or  to 
ward  activity,  in  which  the  primary  object  of  desire  remains 
primary,  and  the  secondary  movement  toward  activity  merely 
modifies  the  primary  value.  These  may  properly  be  described 
as  quasi-aesthetic  values. 

These  quasi-aesthetic  values,  like  the  quasi-ethical,  may 
appear  in  connection  with  all  activities,  perceptual  or  conceptual. 
The  solution  of  a  mathematical  problem,  the  performance 
of  a  surgical  operation,  may,  in  this  sense,  be  beautiful,  no  less 
than  the  "  style  "  or  "  manners  "  connected  with  the  more  funda 
mental  and  instinctive  activities,  such  as  eating  or  walking,  or 
what  not.  All  these  acquired  values  may  be  described  as  style,  the 
style  of  life.  In  their  simplest  form,  however,  such  manners,  to 
gether  with  their  values,  appear  in  connection  with  the  funda 
mental  instincts  and  the  acquisition  and  consumption  of  objects. 
Between  any  fundamental  desire  or  instinct  and  its  satisfaction — 
such  as  the  desire  for  food,  raiment,  love,  etc. — intervene  these 
quasi-aesthetic  activities,  the  values  of  which  enter  into  the 
funded  value  of  the  object.  Style  in  the  serving  and  eating  of 
food,  in  the  forms  and  colours,  and  in  the  actual  wearing  of 
dress,  the  graces  and  coquetries  of  love,  all  extend  the  primary 
values  of  the  objects  immensely,  and  give  them  greater  capacity 
for  continuous  valuation. 

It  is  this  extension  of  the  capacity  for  valuation  which  is 
significant.  And  while  any  object,  no  matter  what  its  simple 
condition  worths  may  be,  may  acquire  such  complementary 


Values  of  Simple  Appreciation  2  -» i 

value,  it  is  especially  at  the  threshold  and  limits  of  condition 
worth  that  this  function  of  quasi-aesthetic  values  is  most  ap 
parent.  We  have  already  seen  that  an  object,  or  quantity  of 
an  object,  too  insignificant  either  for  immediate  appreciation 
or  for  re-application,  in  other  words,  below  the  threshold  of 
intrinsic  or  instrumental  value,  may,  as  part  of  an  aesthetically 
individuated  whole,  acquire  the  value  of  the  whole.1  The  same 
phenomenon,  viewed  in  another  way,  means  that  these  quasi- 
aesthetic  values  modify  the  laws  of  satiety  and  dulling  of  sensi 
tivity  for  simple  condition  worths  as  well  as  the  law  of  limiting 
value  as  it  applies  to  instrumental  values.  It  will  not  be  neces 
sary  to  show  again  how  this  law  enters  into  the  unconscious  as 
well  as  the  conscious  technique  of  life,  into  the  fundamental  as 
well  as  the  more  developed  activities.  It  is  sufficient  to  em 
phasise  the  fact  that  it  modifies  in  a  significant  way  all  our 
economic  judgments. 

1  Chap,  vi,  pp.  151-2,  also  p.  175. 


CHAPTER   VIII 

I.  PERSONAL  AND  OVER-INDIVIDUAL  VALUES  :    THEIR  ORIGIN 

AND  NATURE 

i.  Their  "Common  Meaning"  Presupposes  Sympathetic 
Participation  (Einfiihlung) . 

THE  preceding  study  of  "  condition "  worths,  of  the  simple 
appreciation  of  objects,  has  emphasised  the  purely  individual, 
sub-personal,  and  sub-social  character  of  these  feelings.  These 
feeling-attitudes  may  have  both  quasi-ethical  and  quasi-aesthetic 
modifications  without  acquiring  either  the  personal  or  over- 
individual  and  social  reference  which  is  characteristic  of  the 
more  developed  forms  of  both  ethical  and  aesthetic  experience. 
But  it  was  also  maintained  that  these  modifications  of  simple 
appreciation,  these  complementary  values,  contain  the  germ  of 
more  complete  value  movements  to  the  higher  levels  of  per 
sonal  and  impersonal  social  valuation,  movements  which  in 
volve  the  creation  or  ideal  construction  of  new  objects  and 
new  modifications  of  attitude,  the  development,  in  other  words, 
of  new  levels  of  meaning  where  the  feelings,  and  their  corre 
sponding  worth  judgments,  presuppose  new  judgments  and 
assumptions.  In  succeeding  chapters  we  shall  study  personal  and 
impersonal  social  values  in  detail,  defining  and  classifying  their 
objects  and  seeking  to  determine  the  laws  of  valuation  of  these 
objects.  For  the  present,  our  study  is  confined  to  the  genetic 
problem  of  the  origin  of  these  new  objects  and  attitudes,  more 
especially  the  origin  of  the  distinctive  presuppositions  of  these 
feeling-attitudes. 

That  which  personal  and  impersonal  or  over-individual  values 
have  in  common,  and  which  distinguishes  them  from  simple 
condition  worths,  is  the  fact  that  they  presuppose  certain 
processes  of  acquirement  of  meaning  which  simple  appreciation 
does  not.  This  fact  appeared  both  in  the  nature  of  the  objects 
of  such  values  and  in  the  character  of  the  meaning  of  the  feelings 

232 


Personal  and  Over -Individual  Values  233 

as  expressed  in  the  worth  predicates  attributed  to  the  objects.1 
In  the  first  place,  the  objects  were  described  as  ideal  and 
"  founded,"  i.e.,  they  are  ideal  constructions  founded  on  certain 
processes  which  they  presuppose.  In  the  second  place,  the 
meanings  of  these  feelings,  as  expressed  in  corresponding  worth 
predicates,  are  secondary  acquired  meanings  which,  as  was 
shown  in  Chapter  III,  arise  through  explicit  acknowledgment  in 
judgment  of  meanings  and  references  of  the  feelings  of  simple 
appreciation,  and  through  their  characterisation  as  a  personal 
or  social  over-individual  demand. 

When  we  seek  to  define  more  explicitly  what  this  acquired 
meaning  is,  we  find  that  the  meaning  is  "  common  ";  the  feeling 
is  both  individual  and  over-individual  in  its  reference,  and  the 
objects  are  founded  upon  certain  judgments  or  implicit  assump 
tions  that  the  feeling  and  desire  of  the  individual  are  shared  by 
others.  In  all  feelings  of  personal  worth,  it  is  presupposed  that 
the  self  and  its  dispositions  or  qualities  is  the  object  of  worth 
feelings  on  the  part  of  others.  In  all  personal  judgments  upon 
others,  it  is  assumed  that  objects  of  desire  and  feeling  have  a 
common  meaning.  In  all  judgments  of  impersonal  worth,  it  is 
assumed  that  the  individual  is  representing  a  wider  social  con 
sciousness,  which  is  making  the  same  judgments  and  assumptions. 

An  examination  of  the  worth  predicates  in  these  spheres 
illustrates  the  situation  more  fully.  To  consider  personal 
worths  first,  worth  is  imputed  to  expressions  of  feeling 
or  acts  of  will  as  manifesting  certain  dispositions  of  the 
personality.  The  terms  of  this  imputation  are  approval  and 
disapproval,  praise  and  blame,  merit  and  demerit,  accord 
ing  as  the  disposition  exists  or  does  not  exist,  or  ultimately 
according  to  the  amount  of  disposition.  Corresponding  to 
these  judgments  of  imputation  are  certain  demands  upon  the 
person,  as  a  person,  to  possess  and  express  the  dispositions  in 
question.  These  demands,  when  expressed  in  explicit  worth 
judgments,  constitute  judgments  of  obligation.  Such  imputation 
of  praise  and  blame,  and  of  obligation,  may  be  described  as 
ethical  imputation.  There  is  also  an  aesthetic  imputation  which 
finds  expression  in  the  predicates  of  beauty,  nobility,  sublimity 
of  character  and  actions,  closely  related  to  the  ethical  predi 
cates.  All  these  judgments  of  imputation  presuppose  certain 
assumptions  which  we  may  describe  as  constituting  the  common 
"  ideal  of  personality." 

1  Chap.  II,  p.  28  ;  chap.  Ill,  pp.  71-2. 


234  Valuation:   its  Nature  and  Laws 

The  acts  and  dispositions  in  question  may,  however,  be 
judged  from  another  point  of  view.  Value  may  be  imputed  to  an 
act,  not  because  it  expresses  a  personality,  but  because  it  is  in 
strumental  to  certain  social,  over-individual  ends,  and  satisfies 
certain  impersonal  demands.  The  imputation  of  merit  or  de 
merit,  or  of  obligation,  may  in  this  case  be  described  as  im 
personal  imputation,  and  such  imputation  presupposes  certain 
expectations  or  assumptions  with  regard  to  the  participation 
of  the  individual  in  the  social  life,  therefore  constituting  what 
may  be  described  as  an  ideal  of  social  participation. 

If  now  we  describe  the  first  class  of  values  as  the  characterisa 
tion-value,  the  second  as  the  participation-value  of  acts  and  dis 
positions,  we  may  generalise  to  the  extent  of  saying  that  both 
types  of  valuation  presuppose  the  construction  of  ideal  objects, 
the  disposition,  the  person,  the  social  will,  and  the  emergence 
of  new  feelings  or  modifications  of  feeling  which  express  them 
selves  in  these  judgments  of  imputation.  The  expectations  or 
implicit  assumptions  which  are  presupposed  in  all  such  imputa 
tions  of  personal  or  impersonal  value,  are  in  a  greater  or  less 
degree  common  and  shared.  The  question  of  immediate  interest 
is,  accordingly,  the  study  of  the  psychical  processes  of  participa 
tion  in  which  these  new  meanings  emerge.  These  have  been 
variously  described  as  imitation,  sympathetic  projection  or 
Einfiihlung. 

2.  Sympathetic  Participation — Einfiihlung  :  Its  relation  to 
Feeling  and  Simple  Appreciation. 

It  is  a  commonplace  of  present-day  psychology  that  the 
self  and  the  alter  are  ideal  constructions,  the  material  of 
which  are  the  sensations,  ideas,  emotions,  desires  of  the  in 
dividual.  It  is  equally  a  commonplace  that  this  construction 
is  a  social  process  in  which  imitation  and  opposition,  or  contrast, 
are  at  work  as  the  functional,  genetic  causes.  Baldwin  has  in 
broad  outlines  sketched  the  processes  of  imitation  (projection, 
introjection  and  ejection)  by  which  the  dialectic  of  self-con 
sciousness  takes  place.  Royce  has  emphasised  the  importance 
of  contrast  in  the  process,  and  Tarde,  from  the  sociological  point 
of  view,  has  shown  the  equal  importance  of  imitation  and  oppo 
sition  in  the  generation  of  new  content  in  the  individual  and 
society.  But,  so  far  as  I  am  aware,  there  has  been  no  system 
atic  attempt  to  show  in  detail  the  modifications  of  the  con- 


Personal  and  Over- Individual  Vahtes  235 

sciousness  of  value  brought  about  by  this  dialectic  process.  This 
is  due  to  the  fact  that  the  process  has  been  studied  largely  from 
the  point  of  view  of  the  psychology  of  cognition,  the  interest 
being  primarily  in  the  cognitive  content  and  meaning  of  the  con 
structs,  rather  than  in  the  conative  and  affective  side  of  the  pro 
cess.  When  viewed  in  this  latter  aspect,  the  process  is  seen  to  be 
one  of  Einfuhlung,  a  process  of  "  feeling-in,"  in  which  feeling 
attitudes  are  sympathetically  projected  into  another,  re-read 
back  into  the  self,  thus  becoming  the  objects  of  cognition,  of 
judgment  and  assumption,  and  ultimately  of  new  feelings  of 
value.  It  is,  therefore,  in  this  aspect,  a  value  'movement  in  which 
new  values  are  acquired. 

It  is  as  such  a  value  movement,  as  the  continuation  of  pro 
cesses  of  feeling,  of  simple  appreciation,  that  Einfuhlung  is  to 
be  studied,  and,  when  viewed  in  this  light,  it  is  seen  to  be  a 
value  movement  of  the  "  inward  "  type,  where  the  movement 
is  complete.  There  is,  therefore,  no  hard  and  fast  line  between 
feeling  and  "  feeling-in "  (Einfuhlung),  between  appreciation 
and  characterisation.  Activities  of  appreciation  lead  gradually 
and  necessarily  to  characterisation.  In  the  specific  case  of  the 
feeling  of  obligation,  the  felt  impulsion  which  arises  in  the  forma 
tion  of  a  conative  disposition  is  gradually  referred  explicitly  to 
the  disposition  as  a  phase  of  the  self  or  the  alter.  The  vague 
and  uncertain  transgredient  reference  is  made  explicit  in  the  judg 
mental  reference  of  the  disposition  to  the  ideal  construct  of  the  self. 
It  becomes  conscious  obligation.  It  is,  therefore,  as  a  complete 
value  movement,  beginning  with  simple  appreciation  of  the 
emotional  expressions  of  others  and  ending  with  the  characterisa 
tion  of  persons,  that  Einfuhlung  is  to  be  studied.  We  shall 
accordingly  use  the  term  to  designate  the  entire  process  (pro 
jection,  imitation,  and  ejection)  involved  in  the  activities  of 
characterisation  and  participation,  and  shall  consider  it,  more 
over,  in  its  aspect  of  affective-conative  process,1  for  it  is  a 

1  This  broad  use  of  the  term  Einfuhlung  may  perhaps  be  questioned.  It  is  true  that 
the  weight  of  authority  is  in  favour  of  confining  it  to  the  purely  aesthetic  type  of 
personalisation  and  personal  construction  ;  'it  was  to  explain  certain  aesthetic  phenomena 
that  the  concept  was  first  introduced  (Lipps  and  Volkelt).  On  the  other  hand,  the  broader 
use  to  which  the  very  structure  of  the  term  naturally  gives  rise — to  include  all  forms 
of  projection  or  ejection  of  our  own  feeling  into  others — has  been  employed  by  some 
writers,  notably  by  Witasek  in  his  article,  "  Zur  Psychologic  der  cesthetischen  EinfiMung" 
to  which  more  extended  reference  is  made  in  a  later  section  of  this  chapter.  There 
aesthetic  Einfuhlung  is  considered  to  be  only  one  form  of  the  process  and  is  distinguished 
from  ethical.  That  this  broader  conception  is  justified  by  a  genetic  treatment  of  the 
processes  will,  I  think,  become  apparent  as  the  discussion  proceeds.  Baldwin's 
identification  of  Einfuhlung  with  Sembling  is,  I  think,  only  partially  true.  As  will  be 
seen  later,  there  is  a  semblant  mode  at  one  stage  of  the  total  process,  but  it  is  not  the 


236  Valuation:   its  Nature  and  Laws 

term  which  includes  both  feeling  and  conation,  the  relations 
of  which  have  been  determined  in  a  preceding  chapter.  With 
the  change  in  cognitive  presuppositions  through  participation 
there  comes  corresponding  changes  in  feeling,  in  the  aspects 
both  of  qualitative  and  quantitative  meaning,  in  both  quality 
and  degree.  From  these  laws  of  participation  are  later  to  be 
developed  the  laws  governing  the  valuation  of  personal  and 
over-individual  objects. 

II.  THE   PSYCHOLOGY   OF   EINFUHLUNG:     SYMPATHETIC    PRO 
JECTION 

i.  The  Problem — The  Nature  of  the  Projected  Feeling — How  is 
Einfiihlung  Possible  ? 

Projection,  in  the  larger  sense  of  the  term,  is  used  to  de 
scribe  all  externalisation  of  psychical  content.  We  project  the 
spatial  reference  of  touch  and  other  sensations  into  the  objects 
without  us.  We  project  ideas,  concepts,  formed  in  the  inner 
activities  of  comparison  and  differentiation,  as  the  real  grounds, 
external  to  us,  of  our  sensational  experience.  In  like  manner  we 
characterise  ourselves  and  others  by  reading  into  our  immediate 
experiences  of  feeling  conceptual  constructs  of  dispositions, 
relatively  constant,  to  which  these  feelings  are  referred. 

But  it  is  immediately  clear  that  just  as  the  first  type  of  pro 
jection  of  constructs  is  conditioned  by  immediate  perceptual  ex 
perience,  as  these  constructs  are  ideal  objects  founded  upon 
perception,  so,  in  the  second  type,  the  conceptual  constructs  of 
dispositions  are  founded  upon  immediate  experiences  of  feeling,  in 
this  case  upon  the  "  internal  perception,"  the  appreciation  of  the 
feeling-attitudes  of  others.  Immediate  sympathetic  participation 
is,  therefore,  the  condition  of  the  conceptual  construction  of  dis 
positions  of  the  personality  as  objects  of  personal  worth  feel 
ings.  To  understand  the  objects  of  personal  worth  it  is  neces 
sary  to  analyse  the  processes  of  sympathetic  Einfiihlung  on 
which  they  are  founded. 

The  phenomena  in  connection  with  which  sympathetic 
projection  has  been  chiefly  studied  are  those  of  aesthetic  and 
ethical  personalisation.  The  simplest  form  is  found  in  the  read 
ing  into  impersonal  and  often  inorganic  objects  of  the  organic 

whole  of  the  process.  For  a  general  discussion  of  this  term,  see  Baldwin's  Dictionary 
of  Philosophy  and  Psychology,  on  the  topics,  "  Esthetic  Sympathy"  and  "  Einfiihlung" 
in  the  article  on  terminology. 


Personal  and  Over-Individual  Values  237 

sensations,  feelings,  emotions,  and  desires  that  are  really  in  our 
selves.  Of  this  type  the  attribution  of  feelings  of  movement, 
effort,  or  strain,  to  static  spatial  forms,  or  to  successions  in  time 
(the  upward  striving  of  a  pillar  or  tower  in  architecture,  the 
movement  of  a  melody  in  music)  are  illustrations.  Of  sym 
pathetic  Einfuhlung  in  the  case  of  persons,  we  have  illustrations 
in  the  "  feeling-into  "  persons  of  emotions  such  as  anger  and  fear 
on  the  basis  of  expression,  facial  and  bodily,  as  in  the  case  of  the 
aesthetic  appreciation  of  acting,  or  in  the  case  of  more  ethical  judg 
ments  where  we  take  up  practical  attitudes  toward  persons  on  the 
basis  of  this  appreciation.  It  is  out  of  these  immediate  appre 
hensions  and  appreciations  of  an  "inner  life"  beyond  the  self, 
whether  that  inner  life  be  presumed,  assumed,  or  judged  to  exist, 
that  the  ideal  constructions  arise.  Our  first  problem  is,  therefore, 
to  determine  the  nature  and  conditions  of  this  process  of  per 
sonalization,  which,  in  its  cognitive  aspect,  is  apprehension  and, 
in  its  affective-volitional  aspect,  appreciation.  The  psychological 
problem  is  :  How  is  Einfuhlung  possible  ?  The  solution  of  this 
problem  is  the  necessary  preliminary  to  the  study  of  the  feelings  of 
value  which  presuppose  these  processes. 

The  psychological  problem — how  is  Einfuhlung  possible  ? — 
has  given  rise  to  considerable  discussion  in  recent  literature. 
This  discussion  has  occasioned  a  wealth  of  psychological  analysis, 
but  one  in  which  the  theoretical  problems  raised  cannot  be 
said  to  have  reached  a  final  solution.  The  chief  source  of 
difficulty  is  found  in  the  fact  that  the  experience  felt  into  the 
object,  whether  a  thing  or  a  person,  is  at  once  an  experience 
of  the  subject  of  the  sympathetic  projection,  and  a  content  of  the 
object,  i.e.,  apprehended  as  content  of  the  object.  The  feeling 
seems  to  be  both  experience  of  the  subject,  with  its  own  individual 
presuppositions,  and  content  projected  outside  the  subject, 
apprehended  and  appreciated  as  inner  life  of  the  object.  How 
an  inner  life,  other  than  our  own,  whether  real  or  assumed,  can  be 
interpreted  in  terms  of  our  own  experience,  through  projection 
of  our  own  inner  states;  and  what  are  the  changes  which  the 
feeling  of  the  subject  must  undergo,  changes  both  in  content  and 
functional  presuppositions,  in  order  to  be  projected  and  realised 
as  experience  of  the  alter ;  constitutes  one  of  the  most  difficult 
problems  of  psychology. 

We  may  best  consider  this  general  problem  by  taking  up 
in  detail  certain  special  questions  to  which  it  has  given  rise. 
In  the  first  place,  it  is,  as  we  have  seen,  upon  the  basis  of  some 


238  Valuation:   its  Nature  and  Laws 

immediate  appreciation  that  the  conceptual  constructions  of 
character  and  dispositions  are  founded.  It  is  as  intuitively  real 
ised  that  the  feeling  in  another  leads  to  conceptual  characterisa 
tion.  How  shall  we  conceive  this  intuitive  realisation  of  another's 
feeling  ?  If  it  is  by  a  process  of  projection  of  our  own  feelings, 
what  are  the  aspects  of  the  perceived  objects,  whether  things  or 
persons,  which  furnish  the  stimulus  to  this  projection,  and  how 
is  the  fusion  of  the  subject's  feeling  with  the  object  to  be 
understood  psychologically  ? 

In  the  second  place,  what  is  the  character  of  the  projected 
feeling  ?  Is  it  a  "  real "  feeling,  with  the  same  elements 
of  content  and  the  same  presuppositions  which  characterise 
feeling  as  an  immediate  experience,  or  is  there  a  substitution  of 
presentation  for  feeling,  and  a  gradual  change  in  presuppositions  ? 
What,  in  general  terms,  are  the  changes  in  content  and  functional 
presuppositions  which  condition  participation  in  the  successive 
stages  of  the  process  ? 

Finally  we  have  a  third  problem  growing  out  of  the  preceding. 
What  is  the  role  of  feelings  of  participation  in  the  processes  of 
valuation  ?  Is  the  projected  feeling  itself  a  feeling  of  value,  or 
are  the  only  feelings  of  value  those  judgment-feelings  which 
emerge  when  dispositions  in  the  person  are  judged  to  exist. 
This  apparently  somewhat  subtle  question  has  considerable 
bearing  upon  our  view  of  Einfuhlung  as  a  process  of  valua 
tion,  for  it  is  merely  another  aspect  of  the  more  general 
question — whether  assumption-feelings  or  feelings  of  the  imagin 
ation  are  feelings  of  value.1 

2.  Sympathetic  Projection  of  Feeling — Its  Nature  and  Conditions. 

Our  first  problem,  then,  is  to  seek  to  understand  how  the 
individual  intuitively  apprehends  and  appreciates  the  inner  life 
of  an  object  other  than  himself.  This  is  said  to  take  place  through 
a  process  which  is  described,  in  one  aspect,  as  "  inner  imitation," 
in  another,  as  "  projection."  The  subject  is  said  to  enliven  the 
object  by  projecting  or  feeling  into  it  his  own  feeling-content, 
for  which  process  of  "  feeling-in,"  or  inner  imitation,  some  aspect 
or  expression  of  the  object  constitutes  the  stimulus.  For  the 
purely  psychological  analysis  of  this  process,  and  the  structural 
analysis  of  the  content  of  consciousness  which  conditions  the 
process,  it  is  wholly  irrelevant  whether  the  enlivened  object  is  an 

1  Chap.  V,  pp.  137-9. 


Personal  and  Over- Individual  Values  239 

impersonal  thing  or  a  person;  whether  the  assumption  or  judg 
ment  of  the  existence  of  an  inner  life,  which  emerges  in  the 
process,  is  valid  or  not.  Upon  reflection,  the  object  may  be 
known  to  be  impersonal  and  the  assumption  to  be  invalid,  but 
reflection  does  not  affect  this  form  of  inner  perception  any  more 
than  it  modifies  the  illusions  of  external  spatial  perception. 
Here  we  are  concerned  in  the  first  place  merely  with  the  psycho 
logical  processes  involved. 

The  process  of  inner  imitation  or  projection  is  further  charac 
terised  as  a  fusion  or  complication  of  one  type  of  content  with 
another.  The  more  subjective  content,  which  we  describe  as 
feeling  or  desire,  is  said  to  be  complicated  with  more  objective 
content  of  peripheral  origin.  The  stimuli  of  sight  and  sound, 
spatial  forms,  movement-forms,  gesture,  vocal  expression,  etc., 
act  as  cues  for  experiences  of  feeling  and  organic  sensations. 
The  latter  constitute  the  inner  imitation,  as  distinguished  from 
muscular  imitation,  and  these  contents,  being  fused  with  the 
peripheral  content,  become  objectified — i.e.,  acquire  the  objective 
reference  of  that  content  and  are  apprehended  as  qualities  of  the 
object.  We  may,  therefore,  define  the  feeling  in  inner  imitation 
as  the  induced  feeling  and  the  perceptual  contents  as  the  inducing 
conditions. 

In  order  to  understand  how  this  fusion  is  possible,  we  must  get 
the  phenomena  more  definitely  before  us.  And  in  the  first  place, 
it  must  be  observed  that  the  appreciation  of  the  projected  feeling 
as  a  quality  of  the  object  includes  two  relatively  distinct  situa 
tions,  which  may  be  described  as  a  more  emotional  appreciation 
and  as  an  intuitive  apprehension.  In  aesthetic  Einfuhlung,  for  in 
stance,  two  distinct  situations  are  possible.  In  looking  at  a  Gothic 
tower  we  may,  either  merely  see  the  pinnacles  striving  upward,  or 
be  ourselves  actually  emotionally  elevated,  i.e.,  we  may  have  the 
actual  organic  sensations  attendant  upon  an  inner  motor  tendency. 
In  the  latter  case  there  is  a  personal  participation  in  the  upward 
urge  of  the  pinnacles,  while  in  the  former  we  have  a  projection 
of  the  movement  in  terms  of  representation.  In  like  manner 
we  may  realise  vividly  the  affective  state,  with  all  its  worth  sug 
gestions,  i.e.,  anger  or  fear,  in  another  personality  without  shar 
ing  in  the  actual  organic  sensations  which  make  up  its  content  ; 
or,  again,  when  the  necessary  presuppositions  are  present,  we 
may  experience  sympathetically  the  organic  sensations  likewise. 
And,  it  is  important  to  recognise  that  when  we  thus  vividly 
"  see  "  or  realise  the  upward  urge  of  the  pinnacles,  or  the  fear 


240  Valuation :   its  Nature  and  Laws 

and  anger  of  our  fellows,  we  do  not  simply  think  them  in  con 
ceptually,  but  we  also  in  some  sense  see  them.  It  is,  as  Witasek 
insists,  an  intuitive  (anschauliche)  representation. 

When  we  analyse  these  two  situations,  we  find  that  they 
differ  in  two  important  respects,  in  the  nature  of  the  feeling- 
content  projected  and  in  the  nature  of  the  presuppositions  of 
the  feeling.  We  shall  see  that  between  the  stage  of  immediate 
emotional  participation,  organic  sympathy,  and  the  more  in 
tellectual  intuitive  apprehension  there  is  a  change  both  in  content 
and  cognitive  presuppositions.  It  is  with  the  first  point  that 
we  are  here  concerned ;  the  second  will  be  considered  later. 

In  the  case  of  the  more  emotional  participation,  full  organic 
sympathy,  there  is  a  fusion  of  the  emotion  or  desire,  with  all 
its  subjective  meaning,  with  the  more  objective  perceptual 
or  ideal  content.  This  fusion  we  have  no  difficulty  in  under 
standing  when  we  recall  our  analysis  of  feeling  as  the  "  meaning  " 
of  a  special  form  of  sensitivity,  motor  and  organic.  This  mean 
ing,  we  have  seen,  is  embodied  in  certain  form-qualities  of  the 
sensitivity,  and  when  this  subjective  sensitivity  is  fused  with 
the  more  objective,  under  certain  conditions  presently  to  be 
developed,  the  subjective  meaning  is  felt  into  the  object.  On 
the  other  hand,  the  understanding  of  the  more  intuitive  appre 
hension  of  the  projected  feeling  requires  the  application  of  that 
other  fact,  developed  in  our  analysis  of  feeling,  namely,  that  the 
form  of  combination  of  the  elements  may  be  abstracted  from  the 
elements  and  pass  over  into  presentation.  In  this  way  that  which 
is  peculiarly  psychical,  i.e.,  feeling  and  will,  may  be  presented, 
and,  as  in  the  phenomenon  before  us,  the  projected  feeling  may 
be  intuitively  apprehended  as  a  quality  of  the  object.1 

And  now  appears  a  further  fact  of  importance  to  which  Groos 
has  called  attention,2  and  which  we  shall  have  occasion  to  con 
sider  more  fully  later.  Organic  sympathy,  Einfiihlung  as  well 
as  feeling,  is  subject  to  the  law  of  dulling  of  sensitivity  with 
repetition.  Repetition  may  deaden  the  organic  sympathy, 
damp  down  the  intensity  of  the  organic  resonance  until  it  finally 

1  This  is  developed  more  fully  in  chap,  iv,  p.  103.  Of  the  four  distinguishable  aspects  of 
a  given  affective  attitude,  its  positive  or  negative  direction,  its  presuppositions  (judgment 
or  assumption),  the  form  quality  of  its  elements  and  the  intensity  of  its  resonance,  only 
the  form-quality  is  the  object  of  intuitive  presentation.  The  other  aspects  may  become 
the  objects  of  judgment,  they  may  be  conceptually  represented.  The  pleasantness  or  un 
pleasantness,  with  its  intensity,  and  the  peculiar  individual  presuppositions  may  become 
the  objects  of  judgment  but  not  of  immediate  presentation.  They  belong  to  the  unpre 
sentable  side  of  our  experience. 

8  Der  asthetische  Genuss,  Giessen,  1902,  pp.  1 86  ff. 


Personal  and  Over- Individual  Values  241 

disappears,  but  the  intuitive  realisation  of  the  feeling  or  desire 
projected,  and  of  its  meaning,  still  remains.  We  no  longer  feel 
the  upward  urge  of  the  pinnacles,  or  the  organic  resonance  of 
the  angry  man,  but  we  apprehend  them.  We  have  here,  in 
sympathetic  participation,  a  phenomenon  similar  to  that  ob 
served  in  the  case  of  simple  feeling — showing  the  genetic 
relation  between  simple  appreciation  and  participation,  namely, 
that  while  accommodation  through  repetition  dampens  the 
intensity  of  the  hedonic  and  sensational  aspect  of  the  total 
feeling-attitude,  nevertheless,  with  the  repetition  there  emerges 
the  affective  -  volitional  meaning,  the  dynamic  and  expansive 
suggestions  of  the  feeling  attitude.  This  phenomenon  on  the 
level  of  Einfuhlung,  corresponds  to  facts  of  the  vital  series  where 
emotion  and  passion  pass  over  into  sentiment,  mood,  and 
"  affective  sign." 

Accordingly,  a  certain  change  in  feeling,  as  content,  condi 
tions  the  process  of  projection  whereby  the  projected  feeling  is 
apprehended  and  appreciated  as  a  quality  of  the  object.  The 
two  stages  of  our  illustration  indicate  phases  of  this  modifi 
cation.  The  further  development  of  the  process  by  which 
a  subjective  feeling  becomes  object,  requires  the  study  of  the 
change  in  cognitive  presuppositions  involved  in  the  process  of 
projection.  Before  considering  this  aspect  of  the  problem,  we 
must  glance  at  the  inducing  conditions  of  affective  projection, 
those  aspects  of  perceptual  content  which  constitute  the  stimuli 
for  this  projection. 

The  Inducing  Conditions  of  Affective  Projection. 

The  form- qualities  of  objects,  whether  the  objects  be  persons 
or  things,  whether  the  form-qualities  be  static  or  dynamic, 
are  the  inducing  grounds  of  such  intuitive  affective  projection. 
It  is  with  the  inducing  conditions  in  persons  that  we  are  chiefly 
concerned,  but  these  will  be  best  understood  in  the  light  of 
an  analysis  of  the  simpler  phenomena  of  Einfuhlung  as  involved 
in  the  aesthetic  characterisation  of  things. 

The  inducing  conditions  of  all  affective  projection  are,  in 
the  beginning,  perceptual  form-qualities  determined  by  re 
lations  of  quality,  intensity,  duration,  or  extension  of  simple 
elements.  Rise  or  fall  of  melody  in  the  tonal  scale,  increase 
or  decrease  of  relative  intensities,  light  and  shade,  loudness  or 
softness  of  sounds,  gradual  increase  or  decrease  of  rapidity  of 


242  Valuation :   its  Nature  and  Laws 

tempo  induce  affective  states  with  dynamic  suggestions  which 
are  then  projected  as  tertiary  qualities  of  the  object.  The 
emotions  or  moods  induced  by  the  gradual  increase  of  light  at 
sunrise,  or  decrease  at  sunset,  the  rising  shriek  of  the  wind 
are  illustrations.  To  these  must  be  added  the  dynamic  sug 
gestions  of  qualitative,  intensive,  and  other  contrasts,  in  which 
one  member  of  the  contrasting  pair  is  emphasised  through  the 
opposition  of  the  other;  they  become  the  inducing  grounds, 
especially  in  art,  for  the  suggestion  of  affective  -  volitional 
meanings,  the  affirmation  and  arrest  of  conative  tendencies. 
Rhythmic  time  relations,  both  in  nature  and  art,  may  induce 
various  forms  of  the  two  fundamental  modifications  of  worth 
feeling,  the  dynamic  and  expansive,  as  the  peaceful  murmur 
of  the  brook  or  the  angry  lash  of  the  waves.  Finally,  space 
forms,  such  as  the  upward  urge  of  the  pinnacle,  or  the  dim 
distances  of  perspective,  induce  emotions  or  moods  of  the  dyn 
amic  transgredient  character,  while  other  forms  induce  expan 
sive  feelings  of  repose. 

When  the  objects  of  Einfuhlung  are  persons,  the  inducing 
conditions  are  of  the  same  general  character, — form-qualities 
of  perception.  The  flush  that  mantles  the  face  has  a 
different  emotional  meaning,  according  as  its  rise  is  sudden 
or  gradual.  Muscular  expression  of  the  body,  of  the  face,  or 
even  of  the  eyes  alone,  emphasis  in  speech,  either  of  stress  or 
pitch,  modulation  of  the  voice,  etc.,  all  these  are  in  the  first 
place  form-qualities,  i.e.,  relatively  permanent  relations  of  quality, 
intensity,  or  duration,  among  the  sense  elements.  They  are 
expressive,  that  is,  they  have  meaning,  because  with  them 
fuses  the  movement  form  of  the  emotional  attitude. 

In  all  these  cases  the  psychical  process  may  be  described 
as  a  fusion  of  an  inward  with  the  external  movement  form, 
emotions  being,  as  we  have  seen,  in  their  aspect  of  content, 
movement  forms  of  the  more  subjective  aspect  of  sensational 
content,  muscular  and  organic  sensations,  genetically  residual 
traces,  vestiges  of  form  remaining  over  from  motor  attitudes. 
But  it  must  be  observed  that  while  in  many  cases  we  find  an 
actual  fusion  of  the  muscular  and  organic  sensations  with  the 
external  forms,  this  is  not  a  necessary  prerequisite  of  intuitive 
realisation  of  the  objectified  emotion.  As  we  have  already 
observed  in  Groos's  illustration,  we  may  see  the  upward  urge  of 
the  pinnacle  without  feeling  it  in  the  sense  of  having  the  feelings 
of  effort,  the  reason  being  that  the  essential  element  of  the 


Personal  and  Over- Individual  Values          243 

emotional  attitude  is  the  movement  -  form  which  may,  with 
repetition,  be  abstracted  from  the  organic  and  muscular  sensa 
tions  on  which  it  was  primarily  founded,  and  transposed  to 
another  series  or  group  of  sensations.  The  external  phenomena 
into  which  the  intuited  feeling  is  projected  have  acquired  this 
affective-volitional  meaning. 

It  is  also  important  to  emphasise  the  fact  that,  while  the 
inducing  conditions  are  primarily  perceptual  form-qualities,  they 
are  not  exclusively  so.  Especially  in  the  case  where  the  objects 
are  persons,  certain  unities  and  continuities  of  ideational  activity 
become  the  inducing  grounds  of  sympathetic  projection.  Trains 
of  ideas,  their  types  of  combination  or  association,  acquire  definite 
movement-forms  which  are  expressive  of  feeling  attitude,  more 
especially  those  differences  in  attitude  which  we  describe  as 
temperamental.1  Without  doubt,  emotions  and  moods,  and 
the  dispositions  which  underlie  them,  frequently  determine 
the  associational  processes,  both  as  to  what  images  shall  be 
called  up,  and  as  to  the  manner  in  which  they  shall  be  or 
dered.  We  cannot  be  far  wrong,  then,  in  describing  such  image 
continuities  as  movement-forms  of  thought,  since  in  them  is 
embodied  some  form  of  affective-volitional  meaning.  And  such 
types  of  thought  become  the  basis  for  intuitive  projection  of 
feeling  attitude,  constituting  for  the  observer  emotional  ex 
pression,  just  as  truly  as  do  the  perceptual  movement-forms. 
For  when  through  the  expression  of  his  thoughts  I  realise  the 
affective-volitional  attitude  of  a  poet,  or  of  my  friend,  it  is  not 
by  a  process  of  inference,  a  man  having  such  thoughts,  such  a 
type  of  mind,  must  have  such  and  such  feelings  or  feeling  dis 
positions,  but  immediately  through  sympathetic  participation. 
\ 

1  Herein  lies  the  great  importance  of  similarity  or  difference  of  temperament  as  con 
ditions  of  sympathetic  projection,  an  importance  which  we  can,  however,  merely  suggest 
and  not  develop.  Differences  in  temperament  represent  differences  in  the  capacity  of 
individuals  for  experiencing  different  types  of  emotional  attitude,  and  ultimately  of 
emotional  expression.  They  have,  indeed,  been  classified  on  the  basis  of  degree  of  im 
mediacy  and  intensity  of  sensori-  and  ideo-motor  response,  in  impulse  and  emotion,  that 
is,  on  the  basis  of  the  functional  relation  of  systematic  affirmation  and  arrest  of  conative 
tendency.  It  is  not  difficult  to  see,  therefore,  that  sympathetic  intuitive  realisation  of 
another  man's  attitude  is  largely  dependent  upon  the  temperamental  equipment  of  the 
intuiting  subject.  It  is  equally  apparent  that  worth  judgments  which  have  their 
grounds  in  this  sympathetic  participation,  and  which  we  may,  therefore,  describe  as 
emotional  and  personal  imputation,  will  vary  in  a  significant  manner  from  those  more 
impersonal  judgments  which  are  intellectual  in  character  and  have  their  grounds  in  con 
ceptual  judgment  and  inference.  From  this  difference  important  consequences  will 
follow  for  the  later  studies. 


244  Valuation :   its  Nature  and  Laws 

3.  The  Presuppositions  of  Feelings  of  Participation, 
(a)  In  the  First  Stage  of  Sympathetic  Projection  the  Presuppositions 

are  Presumptions. 

The  second  of  these  general  problems  raises  the  question 
of  the  nature  of  the  presuppositions  of  projected  feelings.     It 
has    been    maintained    that    there    is    an    ultimate   difference 
between  feeling  as  a  simple  condition    of    the   individual    and 
feeling  as  projected  or  felt  into  another,  and  that,  m  order  to 
be  projected,  feeling    must    be    presented  or  imagined, 
presuppositions  of  an  actual  feeling,  as  a  unique  expenence  of 
the  individual  are,   it  is  said,   sensations,   organic   tendencies, 
and  judgments,  of  the  individual  alone,  which  cannot  be  shared. 
Therefore  the  projected  feeling  is  not  an  actual  feeling,  but  a 
presented  one.     In  the  actual  situation  of  sympathetic  Emfuh- 
lung  it  is  not  the  projected  feeling  which  is  real,  but  merely 
the  secondary  feeling  which  follows  upon  judgment  about  the 
disposition  of  the  person  into  whom  the  feeling  is  projected. 
The  projected  feeling  as  such  is  an  imagined  feeling. 

Now  that  in  the  course  of  the  total  process  of  participate 
there   appears    a    stage    where    the    feeling    is    imagined    and 
presented,    and   that   this   is   a   significant   stage    in   the   for 
mation  of   personal  ideals,  is  undoubtedly  true.     There  arise 
a    distinction   between    the    feeling    as    immediately    felt 
as  projected,  due  to  a  differentiation  of   presuppositions.     But 
this   is   not   characteristic   of    the   first   stage    of    the  process. 
At  this  point   the  feeling  has  as  its  presupposition  a  simple 
presumption    of    existence    determined    by    organic    imitation. 
It  has  already  been  pointed  out  that,  when  genetically  viewed, 
the  distinction  between  feeling  and  Einfiihlung  is  not  absolute. 
The  latter  is  but  a  continuation  of  the  processes  of  simple  ap 
preciation    and  we  find  upon  closer  study  that  in  its  simplest 
form    as  mere  organic  sympathy,  it  shows  no  such  distinct 
between  the  feeling  as  felt  and  the  feeling  as  projected, 
are   good  reasons   for  maintaining   that   in   elemental   organic 
sympathy  there  is  no  presentation  of  the  feeling,  and  no  differ 
entiation  between  the  presuppositions  of  the  feeling  as  a  con 
ditionof  the  subject  and  the  feeling  as  a  quality  of  the  object 
It  is  hard,  for  instance,  to  believe  that  the  feelings  of  eff 
experienced  through   organic  sympathy   are  in   any  way 
tinguishable  from  the  same  feelings  when  they  have  their  origin 


Personal  and  Over-Individual  Values          245 

solely  in  the  individual,  and  still  harder  to  believe  that  there 
is  any  difference  between  the  feelings  of  an  infant  when  he 
weeps  in  organic  sympathy,  and  when  he  weeps  because  of 
some  stimulus  which  has  its  origin  wholly  within  his  own 
organism. 

In  the  first  stage  of  sympathetic  projection,  we  may  there 
fore  conclude,  the  presupposition  of  the  feeling  is  a  simple 
presumption  of  existence  of  an  inner  life — not,  it  should,  how 
ever,  be  observed,  of  an  inner  life  as  definitely  localised  either 
in  the  self  or  the  alter.  Negatively  expressed,  there  is  no  dis 
tinction  between  the  feeling  as  individual  experience  and  as 
projected,  no  distinction  between  the  presuppositions  in  the 
two  cases.  In  this  organic  sympathy  we  have  the  germ  of  a 
common  meaning  later  to  develop  into  feelings  of  participation, 
but  as  yet  scarcely  distinguishable  from  the  "  condition " 
feelings  of  simple  appreciation. 

(6)  The  Rise  of  Assumption  Feelings  and  Emergence  of  Distinction 
between  Presuppositions — "  Sembling." 

Nevertheless,  such  participation  in  the  feeling  of  others, 
where  there  is  lacking  the  sense  of  distinction  in  presuppositions, 
is  undeniably  limited  to  the  most  rudimentary  form  of  organic 
sympathy,  on  the  sensational  and  perceptual  level.  All  that  we 
can  infer  from  these  facts  is  that  the  distinction  between  real 
and  imagined  or  presented  feeling,  which  later  becomes  of  im 
portance,  is,  at  least  genetically,  not  ultimate.  But  as  soon  as 
the  instinctive  organic  sympathy  is  in  the  least  degree  modified 
by  arrest,  there  is  readaptation  in  conative  process,  a  change 
in  presuppositions  takes  place,  and  with  it,  we  shall  find,  a 
change  in  the  content  of  projection.  This  change  consists 
in  the  gradual  substitution  of  explicit  assumption  for  presump 
tion,  and  with  such  substitution  emerges  a  distinction  between 
the  feeling  as  merely  felt  and  as  projected. 

The  processes  by  which  the  subject  comes  to  assume  or 
imagine  the  existence  of  an  inner  life  in  the  objects  without 
him  are  in  principle  the  same  as  those  which  condition,  in  its 
most  general  aspect,  the  passage  from  presumption  to  assump 
tion.1  All  feelings  of  value  presuppose  at  least  the  presumption 
of  existence  of  the  object.  When  such  dispositions  to  desire 
or  feel  in  a  certain  way  have  been  created  by  relatively  external 

1  Cf.  chap.  II,  p.  51  f. 


246  Variation  :   its  Nature  and  Laws 

and  objective  conditions,  the  dispositions  thus  formed  become 
a  factor  of  subjective  control,  giving  rise  to  assumptions  of  exist 
ence.  In  the  case  of  organic  sympathy,  and  its  sympathetic 
projection,  there  are  created  certain  dispositions  to  participate 
which  go  beyond  the  limits  of  organic  sympathy,  and  thus  give 
rise  to  Einfuhlung  of  the  playful  semblant  type,  with  its  mere 
assumption  of  existence.  This  may  remain  mere  aesthetic 
appreciation,  or  the  assumption  may  develop  into  judgments 
of  existence  with  their  accompanying  ethical  feelings  of  a  more 
serious  character. 

In  this  passage  from  simple  presumption  to  assumption 
there  is,  therefore,  a  significant  reconstruction  of  the  situation. 
The  subject  now  explicitly  assumes  the  existence  of  the  feeling 
in  the  alter.  In  the  preceding  stage  no  such  explicit  assump 
tion  is  made,  for  the  reason  that  there  is  no  distinction  between 
inner  and  outer  control.  But  to  assume  explicitly  the  existence 
of  the  feeling  is  to  assume  its  necessary  presuppositions  in  the 
alter.  These  presuppositions,  however,  as  well  as  the  hedonic 
subjective  aspect,  cannot,  as  we  have  seen,  be  immediately 
or  intuitively  projected.  They  can  only  be  ejected  as  conceptual 
constructions.  In  the  attitude  of  assumption,  therefore,  we  have 
the  beginning  of  that  ideal  reconstruction  of  the  experiences 
of  immediate  feeling  which  conditions  their  ejection  into  the 
alter  and  his  characterisation  as  a  person. 

Certain  important  changes  in  the  feeling  of  the  subject 
accompany  this  change  from  presumption  to  assumption,  and 
condition  the  further  developments  of  ideal  construction  and 
characterisation.  The  feeling  has  become  an  assumption-feeling, 
and  this,  we  have  seen  from  our  preceding  studies,  involves 
certain  characteristic  changes  in  the  feeling.  In  the  first  place 
there  appear  those  changes  which  we  described  as  abstraction 
and  generalisation.  On  the  functional  side  the  process  was  seen 
to  be  one  of  abstraction  or  detachment  of  the  feeling  from  in 
dividual  presuppositions.1  The  "  feeling-in"  of  an  attitude  into 
another,  with  the  assumption  of  presuppositions  different  from 
those  of  one's  own  feeling,  gives  to  the  feeling  a  quasi-general 
meaning,  a  schematic  character,  which  raises  it  out  of  the 
sphere  of  simple  subjective  appreciation,  and  starts  it  upon  a 
new  path  of  objective  meaning. 

In  the  second  place,  in  the  assumption-feeling  we  found  the 
beginning  of  a  differentiation  of  certain  aspects  of  the  total  feel- 

1  Chap.  V,  pp.  131-3. 


Personal  and  Over- Individual  Values          247 

ing-attitude — of  the  individual  unpresentable  aspect  from  the 
presentable  form-qualities  of  the  feeling.  With  this  differentia 
tion  the  feeling  acquires  recognitive  and  generic  meaning.  These 
characteristics  of  assumption-feelings  which  were  developed 
in  the  abstract  analysis  of  the  earlier  chapter,  are  now  seen  to 
get  an  additional  significance  in  the  processes  of  Einfiihlung. 
On  the  one  hand  the  rise  of  assumption-feelings,  feelings  of 
the  imagination,  with  accompanying  changes  in  content  and 
function,  is  the  condition  of  the  extension  of  participation  beyond 
the  range  of  the  simple  appreciations  of  organic  sympathy. 
On  the  other  hand,  the  processes  of  Einfiihlung  continue  the 
processes  of  generalisation  and  obj  ectification  of  feeling-attitudes 
already  begun.  It  is  to  this  latter  aspect,  and  its  significance 
for  the  further  processes  of  characterisation  and  ideal  construc 
tion,  that  we  must  now  turn. 

These  two  characteristics  of  the  assumption-feeling,  or 
feeling  of  the  imagination,  its  relative  independence  of  in 
dividual  presuppositions,  and  its  differentiation  of  aspects 
of  the  total  attitude,  lead  to  important  progressions  or 
value  movements  in  the  characterisation  of  the  self  and  the 
alter. 

In  the  first  place,  such  a  feeling  has  acquired  a  dual 
character  and  function.  It  is  at  the  same  time  both  subjective 
feeling  and  objective  presentation  ;  it  has  both  an  individual 
and  an  over-individual  reference.  It  is,  so  to  speak,  in  a  state 
of  unstable  equilibrium  preliminary  to  new  differentiations 
and  determinations.  This  point  must  be  emphasised  if  we  are 
to  make  clear  the  nature  of  this  transition  stage  in  the  total 
process  of  Einfiihlung,  for  it  is  closely  connected  with  a  ques 
tion  to  be  raised  presently — whether  the  projected  content 
is  a  "  real "  feeling,  or  is  merely  imagined.  To  say  that 
the  sembled  feeling  is  both  subjective  feeling  and  objective 
presentation  seems  at  first  sight  paradoxical.  But,  according 
to  the  view  already  developed,  feeling  is  the  subjective  meaning 
of  a  specific  kind  of  content,  a  meaning  which  may  be  em 
bodied  in  organic  sensations,  or  may  be  transposed  to  more 
objective  content  of  peripheral  origin,  when  the  feeling  is  said 
to  be  intuited.  When,  therefore,  it  is  held  that  the  sembled 
or  presented  feeling  has  a  dual  reference,  both  subjective 
and  objective,  it  is  meant  that  it  stands  in  a  representative 
capacity  for  the  individual  feeling,  with  its  uniquely  individual 
presuppositions,  from  which  it  has  been  abstracted  and  into 


248  Valuation :   its  Nature  and  Laws 

which  it  may  be  re-converted  on  occasion,  and  that  at  the  same 
time  it  has  an  objective  over-individual  reference. 

This  dual  character  develops  further  the  "  common  mean 
ing  "  already  implicitly  present  in  organic  sympathy.  In 
organic  sympathy,  presuppositions  are  not  distinguished;  but 
with  the  abstraction  from  individual  presuppositions  which 
comes  with  the  assumption  attitude,  the  feeling  acquires  a 
schematic  character,  which  permits  it  to  be  read  back  and  forth 
from  the  self  to  the  alter.  By  its  schematic  character  is  under 
stood  precisely  this  abstraction  from  uniquely  individual  pre 
suppositions  and  from  organic  sensation  content,  the  charac 
teristic  which  makes  possible  its  dual  reference,  and  makes 
it,  to  use  Baldwin's  terms,  a  "  Person-Project,"  which  may, 
in  the  further  development  of  the  process,  be  identified  in  ex 
plicit  judgment  with  either  the  self  or  the  alter. 

Of  chief  importance  in  the  rise  of  the  objective  and  "  com 
mon  "  meaning  of  feeling-attitude,  is  the  fact  that  the  schematic 
feeling  of  the  imagination  may  acquire  recognitive  and  generic 
meanings.  With  the  intuitive  projection  of  the  feeling  into  the 
"  other,"  appears  the  conscious  recognition  of  the  feeling  as 
one's  own.  It  has  already  been  shown  in  another  connection  T 
that  the  condition  of  acquirement  of  recognitive  meaning  on 
the  part  of  a  feeling  is  that  it  shall  become  an  assumption 
feeling.  It  is  not  the  unique,  individual  aspect  of  the  feeling 
which  is  recognised,  but  the  schematic  movement-form  of  the 
feeling  of  the  imagination.  In  the  semblant  mode  of  Einfuhl- 
ung  we  have  precisely  the  condition  necessary  for  the  objecti- 
fication  and  recognition  of  feeling-attitude,  and  for  the  reading 
back  of  the  feeling,  in  terms  of  idea,  into  the  self.  Here  also 
arise  those  further  differences  in  the  meaning  of  the  feeling, 
between  feeling  as  passive  pleasantness-unpleasantness  and 
as  dynamic  and  expansive  movement-forms,  with  their  reference 
to  conative  dispositions, — those  differences  in  feeling  which  lead 
on  to  ideal  construction. 

(c)  Feelings  of  Participation  as  Judgment- Feelings — Presuppo 
sitions  as  objects  of  Judgment — Conceptual  Reconstruction 
of  the  Inner  Life  in  Terms  of  Dispositions. 

The  rise  of  assumption-feelings,  and  the  semblant  mode 
in  participation,  is  the  condition  of  the  acquirement  of  certain 

1  Chap,  v,  p.  117  ff. 


Personal  and  Over- Individual  Values          249 

meanings  and  of  further  development  in  the  characterisation 
of  persons.  It  has  already  been  shown  that,  while  this  stage  of 
assuming  or  sembling  of  an  inner  life  in  other  things  and  persons 
may  remain  a  mere  assumption,  an  intrinsic  appreciation  with 
the  immanental  values  of  the  aesthetic,  it  may  also  lead  on  to 
a  further  stage  in  characterisation.  The  intuitive  schematic 
feeling  "  project  "  may  be  merely  instrumental  to  a  stage  where 
the  vaguely  assumed  inner  life  is  explicitly  acknowledged  in 
acts  of  judgment.  This  stage  we  may  describe  as  the  ejection 
of  the  presuppositions  of  the  feeling  in  the  form  of  conceptual 
construction  of  dispositions. 

When  this  stage  of  the  process  is  reached,  the  conditions 
of  the  complete  inward  value  movement,  toward  the  valua 
tion  of  the  disposition,  are  given.  In  the  place  of  the  mere 
presumption  or  assumption  of  the  existence  of  a  dispositional 
correlate  for  the  projected  feeling,  judgment  develops,  and 
judgmental  habit.  It  is  these  judgments  of  existence  or  non- 
existence  of  dispositions,  in  the  self  or  the  alter,  which  consti 
tute  the  presuppositions  of  those  feelings  of  value  which  we 
have  described  as  personal  and  impersonal  (social),  and  which 
find  expression  in  those  judgments  of  obligation  and  imputation 
described  in  earlier  paragraphs  of  this  chapter.  Repetition 
of  judgment,  the  formation  of  judgment-habit,  gradually  creates 
certain  implicit  assumptions  or  expectations,  with  which  the 
subject  making  the  worth  judgment  comes  to  the  objects  of 
judgment,  the  acts  and  dispositions  of  persons. 

4-  Einfilhlung  as  a  Process  of  Valuation—The  Nature  of  the 
"  Feelings  of  Value  "  involved—Value  Movements  in  Sym 
pathetic  Participation  and  Characterisation. 

(a)  The  Projected  Feelings  are  "Real  "  Feelings. 

We  have  now  traced  genetically  the  gradual  change  in 
tive  presuppositions,  together  with  the  corresponding 
changes  in  feeling,  which  characterise  the  total  process  of  par 
ticipation  as  affective-conative  process.  The  purely  psychological 
study  of  changes  in  content  and  functional  presuppositions 
which  condition  the  acquirement  of  these  personal  and  over 
individual  meanings,  requires  to  be  supplemented  at  certain 
points  by  a  more  specific  inquiry  into  the  role  of  the  feelings 
of  participation  in  the  processes  of  valuation. 

This  inquiry  is  obviously  fundamental  to  the  study  of  the 


2  50  Valuation  :   its  Nature  and  Laws 

consciousness  of  personal  and  social,  over-individual  values, 
for  these  feelings  of  value  depend  upon  Einfiihlung,  and  one  of 
the  most  important  succeeding  studies  will  be  concerned  with 
the  application  of  the  laws  of  valuation  to  these  feelings.  But 
as  a  preliminary  to  this  study  certain  questions  arise  which 
belong  properly  to  this  psychological  analysis.  The  projected 
feeling,  with  its  individual  and  over-individual  reference,  is 
both  subjective  feeling  and  objective  presentation.  In  addition 
to  its  individual  presuppositions,  as  subjective  feeling,  it  pre 
supposes  presumption,  assumption,  or  judgment  of  existence 
beyond  the  subject,  in  another.  Is  the  feeling  when  projected 
a  real  feeling  of  the  individual,  and,  therefore,  a  feeling  of 
value  ? 

In  the  first  place,  the  question  has  arisen  whether  feelings, 
when  they  lose  their  purely  individual  character,  are  still 
feelings,  or  whether,  with  the  change  in  functional  presupposi 
tions  and  content  described,  they  do  not  cease  to  be  feelings 
and  become  presentations.  This  question,  which  is  but  a 
special  aspect  of  the  larger  problem  which  has  presented  itself 
at  different  points,  has  received  attention  largely  in  connection 
with  the  study  of  Einfiihlung  in  its  aesthetic  aspect.  In  this 
discussion  it  has  been  maintained,  on  the  one  hand,  that  only 
presented  feelings  could  be  projected,  while  on  the  other  hand 
it  has  been  insisted  that  the  projected  feelings  are  real  feelings. 

The  question  whether  the  projected  feeling  is  an  actual 
feeling  or  a  presentation  of  a  feeling,  involves  a  further  question. 
Is  the  immediate  apprehension  and  appreciation  of  the  feeling 
of  another,  with  its  presumption  or  assumption  of  existence,  a 
feeling  of  value,  or  do  feelings  of  value  arise  only  when  secondary 
judgments  as  to  the  existence  or  non-existence  of  real  feelings 
or  feeling-dispositions  corresponding  to  the  presented  feeling 
are  passed  ?  We  have  already  maintained  that  the  projected 
feeling  is  a  real  feeling  of  value,  with  presumption  or  assumption 
of  existence.  It  remains  now  merely  to  justify  this  position 
by  a  critical  consideration  of  the  psychological  questions  in 
volved.  The  entire  dispute  may  be  referred,  I  think,  to  an  in 
adequate  conception  of  the  nature  of  feeling,  and  to  an  un- 
genetic  abstraction  of  the  two  terms, — function  and  content. 

Witasek 1   maintains   that   only   presented   feelings    can    be 

1  Witasek,  Zur  psychologischen  Analyse  der  resthetischen  Einfiihlung,  Zeitschrift  fur 
Psychologic  und  Physiologic,  etc.,  XXV,  1901.  Also  Allgemeine  ^sthetik,  Leipzig, 
1904,  pp.  114  and  107. 


Personal  and  Over-Individual  Values          2  5 1 

projected.     For,  in  the  first  place,  actual  feelings  have  certain 
presuppositions,  sensational,  organic,  conative,  and  judgmental, 
which  belong  to  the  individual  alone.     Without  these  the  real 
feeling  cannot  exist.     But  these  presuppositions  are  different  in 
the  individual  who  projects  and  the  other  into  whom  the  feeling 
is  projected.  The  presuppositions  cannot  themselves  be  projected. 
While  the  projected  feelings  are  imagined  or  presented,  never 
theless,  actual  feelings  of  sympathy,  feelings  of  participation,  may 
arise  when  the  subject  affirms  the  existence  of  dispositions  cor 
responding  to  these  presented  feelings.    Groos,1  on  the  other  hand, 
maintains  the  real  character  of  the  projected  feelings  themselves, 
on  the  ground  that  organic  sensations  are  present,  thereby  making 
them  the  criterion  of  real  feeling,  at  the  same  time  recognising 
that  there  may  be  projection  of  feeling  in  terms  of  pure  presenta 
tion.     Thus,  I  may  be  conscious  of  a  man's  anger  or  fear  without 
experiencing  the  organic  disturbance  which  is  the  basis  of  that 
actual    emotion,   and,   again,   I    may    have    that    presentation 
plus  the  organic  disturbance,  when  I  actually  feel  the  experience 
with  him.     In  both  cases  we  have  Einfiihlung,  but  in  one  case 
the  feeling  is  actual,  while  in  the  other  it  is  presented. 

It  is  apparent  that  in  the  case  of  both  these  views  we  are 
presented  with  but   a  partial   aspect   of   the   truth.     Witasek 
finds   "real"  feelings  of  participation  only  on  the  higher  level 
of  ejection  where  dispositional  presuppositions   are  judged  to 
exist.      They    are,    therefore,    judgment-feelings.      Groos    sees 
in  the  primitive   organic  sympathy,   with  its  presumption  of 
reality,  an  experience  of  real  feeling,  but  denies  reality  to  the 
stage   of    imaginative    projection,   because    of   the   absence   of 
organic   sensations.      In    neither   of    these   views   is    the    true 
criterion  of  real  feeling  given.     With  respect  to   the  content 
aspect    of    feeling-attitude,  which    Groos    has    in    mind   when 
he    makes    the    presence    of    organic    sensations    the    criterion 
of  real  feeling,  we  have  already  seen  in  our  analysis  of  feeling  2 
that,  while  feeling  is  primarily  the  embodied  meaning  of  certain 
forms  of  sensitivity,   organic  and  motor,  it  is  not  dependent 
upon  that  content  for  its  meaning,  but  may  be  transferred  as 
form-quality  to  other  content,  relatively  more  objective,  without 
losing   its   subjective   reference    and   meaning.     On    the   other 
hand,  the  same  analysis  of  feeling  has  shown  us  that  the  as 
sumption  which  underlies  the  view  of  Witasek — that  the  dis 
tinction  between  feeling  and  presentation  is  ultimate — will  not 

1  Groos,  Der  asthetische  Genuss,  p.  209.  2  Chap,  iv,  pp.  100-3. 


252  Valuation:   its  Nature  and  Laws 

bear  examination.  If  such  a  view  is  taken  there  is,  of  course, 
no  question  that  the  coefficient  of  reality  is  lacking  to  the  pro 
jected  and  intuitively  presented  feeling-attitude,  but  this  view 
is  without  foundation. 

The  mistake  at  the  root  of  both  these  misconceptions  of  the 
situation  is,  as  has  already  been  suggested,  the  neglect  of  the 
genetic  point  of  view,  the  failure  to  recognise  that  the  imaginative 
projection,  with  its  quasi-presentational  content,  with  recog- 
nitive  and  generic  meaning,  is  but  a  transition  stage  between 
the  simple  appreciations  of  organic  sympathy  and  the  more 
developed  feelings  of  participation  which  come  with  judgment 
and  judgment-habit,  or  implicit  assumption — an  intermediate 
stage  in  which  the  "  psychical  "  attitude  is  intuitively  presented, 
recognised,  and  referred  in  acts  of  judgment  to  the  self  and  the 
alter.  As  such,  it  has  all  the  meaning  of  the  primary  feeling 
for  which  it  stands,  and  into  which  it  can  be  again  converted 
by  reinstatement  of  individual  presuppositions.  As  such,  it  is 
also  representative  and  anticipatory  of  the  later  feelings  of  par 
ticipation,  when  presumption  and  assumption  pass  over  into 
explicit  judgment.  What  has  been  said,  in  more  general  terms, 
with  reference  to  the  real  character  of  the  affective  abstract  and 
affective  sign  *  and  its  role  as  real  feeling  in  the  processes  of 
valuation,  may,  therefore,  now  be  repeated  and  again  emphasised 
in  connection  with  its  function  in  the  activities  of  sympathetic 
participation.  The  projected  feeling  is  a  feeling  of  value, 
whether  that  over-individual  reference  consist  in  a  presumption, 
assumption,  or  judgment  of  its  existence  in  another. 

(b)   Value  Movements  in  Participation  and  Characterisation. 

The  conclusion  of  the  preceding  paragraphs — that  the 
distinction  between  the  feeling  as  individual  and  as  projected 
and  shared  is  merely  genetic  and  relative,  enables  us  finally 
to  see  the  genetic  relations  between  the  different  feelings  of 
participation  and  the  different  cognitive  attitudes  in  Einfiihlung 
and  characterisation  of  persons.  As  in  simple  appreciation  of 
objects,  so  in  characterisation  of  things  and  persons,  there 
are  value  movements,  in  the  latter  case  from  organic  sympathy 
to  ethical  and  aesthetic  participation  and  characterisation. 

The  difference  between  ethical  and  aesthetic  participation  is  not 
a  difference  in  affective  content  of  projection,  but  in  cognitive 

1  Chap,  v,  pp.  137-9. 


Personal  and  Over-Individtial  Values          253 

presuppositions.  In  aesthetic  Einfiihlung,  the  presuppositions 
are  assumptions,  in  ethical,  judgments.  In  both  cases  the 
feeling  as  psychical  content  has  been  intuitively  projected  on 
the  basis  of  the  inducing  conditions  described,  and  with  this 
has  developed  the  conceptual  construction  of  dispositions 
which  are  either  judged  or  assumed  to  exist  or  not  exist.  It 
is  the  feelings  following  upon  these  judgments  or  assumptions 
which  are  different  in  the  two  cases.  The  ethical  feelings 
are  those  of  judgment,  the  aesthetic  those  of  assumption.1  In 
the  case  of  a  character  of  fiction  or  of  the  drama,  as,  for 
instance,  Lear,  we  realise  the  despair  and  fury  just  as  really  as 
in  real  life,  sometimes  more  clearly  and  vividly,  and  we  have 
a  certain  kind  of  sympathy  arising  from  a  temporary  assump 
tion  of  the  existence  of  actual  presuppositions  of  the  feelings. 
In  ethical  participation,  on  the  other  hand,  we  have  the  in 
tuitive  realisation  of  the  feelings,  but  the  presuppositions  are 
judged  to  be  real. 


III.  THE  DISTINCTION  BETWEEN  PERSONAL  AND  IMPERSONAL 
FEELINGS  OF  PARTICIPATION 

i.  Intensive  and  Extensive  Projection. 

The  preceding  study  of  the  processes  of  Einfiihlung,  of  the 
origin  of  feelings  of  participation  with  their  ideal  objects,  has 
failed  to  take  account  of  one  difference  in  acquired  meaning 
which  emerges  very  early  in  the  process,  namely,  the  distinction 
between  personal  and  impersonal  feelings  of  value.  All  feelings 
of  participation  have  an  over-individual  reference,  have  a 
common  meaning,  but  this  common  meaning  may  be  specifically 
qualified  as  personal  or  impersonal.  In  the  first  chapter  it  was 
pointed  out  that  the  same  object,  the  disposition  to  act  or  feel 

1  This  analytical  distinction,  true  as  it  is,  may  be  pressed  too  far  if  in  calling  the 
assumption-feelings  of  aesthetic  sympathetic  participation,  Sc/ieingefiihle,  and  the 
judgment  feelings  of  the  ethical,  real  feelings,  the  aesthetic  are  banished  from  the  realm 
of  worth  feelings.  We  have  already  in  our  introductory  definitions  included  the 
assumption-feelings  among  the  worth  phenomena,  a  procedure  the  importance  of  which 
will  become  especially  apparent  in  the  following  study  of  the  activities  and  values  of 
characterisation.  In  the  actual  process  of  characterisation  of  the  person  and  the  im 
puted  worths  which  arise  in  that  process,  we  shall  find  the  a-sthetic  participation  worths 
entering  as  actual  determinants  in  ethical  valuation.  Without  going  into  greater  detail 
in  the  present  connection,  it  will  be  sufficient  to  point  out  that  idealisation  of  character 
plays  an  important  role  in  ethical  judgments,  and  that  these  ideals  are  realised  in  aesthetic 
moments  when  assumptions  take  the  place  of  judgments.  The  counsels  of  perfection 
whispered  to  our  souls  in  these  moments  colour  our  ethical  judgments  in  the  real  life 
that  follows. 


254  Valuation:   Us  Nature  and  Laws 

in  a  certain  way,  might  be  judged  from  two  different  stand 
points  the  personal  and  the  impersonal.  The  feelings  of  value 
in  these  cases  are  qualified  by  different  acquired  meanings  and 
references.  The  subject  of  the  judgment  represents  two  different 
selves  or  perhaps  better,  two  different  attitudes  of  the  sel: 
Our  problem  is  to  account  for  this  differentiation  of  meaning 

and  attitude. 

This  difference  in  attitude  is  fundamental,1  and  has  its  r 
in  the  most  elementary  conditions  and  processes  of  sympathetic 
participation.  Prior  to  all  reflective  judgment,  to  conscious  judg 
ment  regarding  the  instrumental  value  of  a  quality  or  disposition 
for  the  purposes  of  the  self  or  others,  and,  therefore,  prior  to 
the  reflective  distinction  between  egoism  and  altruism,  there 
is  a  felt  difference  between  the  demand  which  is  more  intensive 
and  personal,  and  that  which  is  extensive  and  social,  between 
the  feeling  of  approval  or  disapproval  which  is  more  inward  and 
personal  in  origin  and  reference  and  that  which  is  more  external 
and  impersonal.  These  distinctions  go  back  to  a  difference  in 
presuppositions  present  in  the  most  elementary  forms  c 

pathy. 

The  fundamental  difference  we  have  in  mind  may 
conveniently  characterised  as  the  distinction  between  intensive 
and  extensive  sympathy.  By  intensive  sympathy  and  sym 
pathetic  participation,  we  understand,  in  the  first  place,  that 
form  of  sympathy  in  which  the  processes  of  imitation,  imagina 
tive  projection,  and  ideal  construction,  are  confined  to  the  ego 
and  a  single  alter.  Where  the  conditions  of  such  relations  are 
realised  repeatedly,  the  common  content  with  its  common  mean 
ing  becomes  markedly  individuated  and  personal.  By  exten 
sive  sympathy,  on  the  other  hand,  we  understand  that  form  in 
which  a  larger  number  of  individuals  are  included  in  the  range 
of  sympathetic  projection,  where  the  individual  participates  11 
the  feeling  and  conation  of  social  or  racial  groups.  Between 
the  extremes  of  intensive  and  extensive  sympathy,  there  are, 
however,  innumerable  intermediate  stages,  and  the  terms 
obviously  have  merely  a  relative  significance. 


Personal  and  Over- Individual  Values          255 

This  relative  significance  is,  nevertheless,  important,  for  the 
reason  that  the  more  extensive  the  range  of  participation,  the 
more  unindividuated  and  impersonal  the  common  meaning  is. 
The  grounds  for  this  we  shall  seek  to  show  in  detail  presently, 
but  it  is  possible  to  see  without  further  analysis  that  the  more 
extensive  the  range  of  projection,  the  more  generic  and  racial 
must  be  the  feeling-attitude,  and  the  less  completely  will  it 
be  identified  with  a  single  personality.  If  this  is  so,  we  should 
properly  expect  to  find  the  implicit  assumptions  and  the  dis 
positions  which  they  express,  differing  widely  according  to  the 
intensive  or  extensive  character  of  the  processes  of  participation 
in  which  they  are  formed. 

2.  The  Distinction  due  primarily  to  Differences  in  Inducing 

Conditions. 

The  inducing  conditions  of  affective  projection  in  general, 
considered  without  reference  to  the  later  distinction  between 
intensive  and  extensive  projection,  have  already  been  described. 
Organic  imitation  of  movement- forms,  perceptual  or  ideal, 
is  the  basis  of  sympathetic  projection.  When  the  objects 
of  Einfuhlung  are  persons,  the  perceptual  movement-forms 
are  bodily  expression  of  various  kinds,  and  those  phases 
of  motor  expression  which  appear  in  the  qualities  of  thought 
and  speech.  For  the  most  elementary  organic  sympathy, 
similarity  of  motor  attitude,  of  expression  in  general,  is  a  neces 
sary  pre-requisite ;  and  for  imaginative  projection,  certain  simil 
arities  of  attitude  or  temperament,  which,  as  we  have  seen, 
are  reducible  to  similarities  in  the  functional  relations  of 
affirmation  and  arrest  of  conative  tendency,  are  equally 
necessary. 

Taking  attitude  in  the  sense  defined,  it  is  clear  that 
within  any  social  group,  whether  large  or  small,  there  will  be 
various  differentiations  of  attitude.  In  the  first  place,  to  con 
sider  the  most  general  and  abstract  difference  first,  individuals 
are  found  to  differ  in  the  degree  to  which  they  are  predominantly 
affirmative  or  negative  in  type.  Between  the  two  limits — total 
absence  of  inhibition,  and  predominant  arrest,  such  as  we  find 
in  ascetics,  there  will  be  indefinite  variation.  At  present  we 
are  not  concerned  with  the  causes  of  such  variation — whether 
sub-social,  biological,  and  economic,  but  merely  with  the 
fact  of  variation.  This  fact  being  recognised,  it  is  further  seen 


256  Valuation:   its  Nature  and  Laws 

that  within  any  organised  society,  or  even  group  within  a  larger 
society  wherever  there  has  been  isolation  enough  to  make  sue 
a  group  or  society  relatively  homogeneous,  there  is  a  tendency 
for  these  variations  to  arrange  themselves  about  a  norm   t 
more   individual  variations,  either  in  excess  or   defect  of 
normal  being  eliminated  or  lost  in  the  group  attitude. 

More    specific    variations    in    attitudes    and    emotional 
are  determined  by  the  character  of  the  fundamental  conative 
tendency  about  which  the  other  tendencies  are  systematised, 
and  which  in  consequence  has  the  greatest  arresting  and  organ 
ising  influence  upon  other  tendencies.     Thus,  in  every  assemblage 
of  individuals,  sympathetic  participation  is  determined  by  what 
we  may  describe  as   similarity  or  differentiation  of    Crests, 
which  colour  in  a  noticeable  way  all  emotional  attitudes.     With 
out  pretending  to  an  exhaustive  treatment  of  the  types  and 
causes   of   this   differentiation   and   its   effect   upon   emoti< 
attitude    the   possibilities   being   really  inexhaustible,  we   may 
content  'ourselves  with  reference  to  two  lines  of  differentiation 
which  have  attracted  special  attention,  namely,  differences  c 
function  arising  out  of   differences  of  sex  and  of  employment. 
Recent  studies  in  social  psychology,  especially  those  of  Bucher 
and  Veblen,2  have  made  much  of  the  effect  of  differentiation  of 
fundamental  functions  upon  secondary  or   derived  sentimer 
and    emotions.     The    centre   of    affirmation  and  arrest  vanes 
so  greatly,   and  becomes  so  fixed  with  generations   of  soc 
heredity,  that  the  sentiments  and  ideals  of  one  class  are  wit 
difficulty  realisable  by  another. 

When  we  turn   to  the  fundamental  differentiation   of  sex 
function,  the  effect  of  organic  function  in  determining  dil 
ence  in  derived  sentiments  and   emotions  is  still   more  promi 
nent      In    addition   to   the    generic    difference  which   appears 
in   the   fact  that  to  the  man  affirmation  is  more  of   a  hat 
than  in  the  case  of  the  woman,  there  are  the  specific  differenc 
which  arise  from  the  different  locus  of  the  arrest  in  the  two 
cases       Fundamental    differences    in    feeling-attitude    and 
dispositional  capacity  for  given  sentiments  and  emotions,  se1 
certain  definite  limits  to  sympathetic  participation,  and  ul 
mately,  when  ideal  construction  of  dispositions  has  taken  pi; 
to  differentiation  in  ideals  of  obligation  and  virtue. 


Leipzig,  1890. 


Personal  and  Over- Individual  Values          257 

From  this  brief  sketch,  it  may  be  easily  seen  that  there  are 
causes  at  work  differentiating  and  establishing,  in  larger  and 
narrower  groups,  habitual  attitudes  and  expressions  which 
form  the  inducing  conditions  of  sympathetic  participation. 
These  causes  are  doubtless  both  social  and  sub-social,  the  sub- 
social  factors  being  biological  and  psychological  in  the  narrower 
individual  sense.  But  with  the  special  character  of  these 
causes  we  are  not  immediately  concerned.  Our  problem  is 
rather  this  :  granted  the  existence  of  such  differentiations  of 
attitude,  how  do  they  affect  sympathetic  participation  and  the 
feelings  of  value  which  emerge  in  these  processes  ? 

3.    The  Distinction  Develops  with  Progress  from  Organic 
Sympathy  to  Ideal  Construction. 

The  most  elementary  differentiation  of  the  personal  and 
impersonal  reference  of  "  common  feeling  "  occurs  in  organic 
sympathy,  the  first  stage  of  sympathetic  participation.  It 
becomes  more  marked  in  the  succeeding  stages  of  imaginative 
projection,  or  sembling,  and  of  ideal  construction  and  judgment. 
In  organic  sympathy  itself,  prior  to  differentiation  of  presuppo 
sitions,  the  sympathy  may  have  an  individual  or  social  reference. 
Group  passions  and  emotions  are  as  fully  realisable  sympathetic 
ally  as  are  individual.  We  may  well  believe  that  what  is  called 
contagion  of  emotion  is  more  completely  realised  on  primitive 
than  on  the  higher  social  levels.  The  individual  subject  to 
this  contagion  apprehends  immediately,  through  inner  imitation, 
the  emotional  attitude  thus  expressed,  and  projects  it  with 
presumption  of  its  existence  beyond  the  self.  It  is  a  common 
feeling,  and  for  the  participant  has  a  common  meaning,  but  it 
is  not — and  this  is  the  important  point — localised  in  an  individual 
and  then  read  back  into  the  self.  It  remains  external  and  im 
personal  in  its  reference,  as  an  over-individual,  impersonal 
demand. 

The  differentiation  of  personal  and  impersonal  reference  of 
the  common  meaning  becomes  still  more  marked  when  sym 
pathetic  participation  is  extended  beyond  organic  sympathy, 
through  sembling  or  feelings  of  the  imagination.  The  inducing 
conditions  of  organic  sympathy  are  similarities  of  motor  atti 
tude  and  expression,  and  the  limits  of  such  participation  are 
definitely  set  by  biological  conditions  which  are  sub-social 
and  psychological.  Beyond  this  point  the  projection  becomes 


258  Valuation:   its  Nature  and  Laws 

imaginative,  and  the  feeling  of  participation  is  an  assumption- 
feeling.  The  condition  of  such  imaginative  projection,  we 
have  already  seen,  is  abstraction  of  the  feeling  from  its  individual 
presuppositions  and  from  its  organic  accompaniments.  The 
significance  of  this  fact  for  our  study  of  the  modification  of 
the  "  common  meaning  "  of  feeling  through  extensive  projection, 
is  far-reaching.  It  follows  that  the  more  extensive  the  range 
of  sympathetic  projection,  the  more  the  individuals  vary  in 
expression  and  attitude,  the  more  complete  this  process  of 
abstraction  must  be,  and  the  less  personal  will  be  the  reference 
of  the  feeling. 

This  consequence  may  be  shown  in  two  ways.     In  the  first 
place,  only  certain  limited  classes  of  feeling-attitudes  are  sus 
ceptible  of  extensive  projection,  and  these  are  precisely  those 
attitudes  which  are  least  individualised,  most  completely  generic 
or  racial.    Within  narrower  groups,  the  derived  and  more  highly 
differentiated  emotional  attitudes  may  be  intuitively  appreciated, 
but  with  the  increase  of  extent  of  the  projection,  the  assump 
tion  becomes  more  and  more  uncertain  and  the  characterisa 
tion  more  and  more  fanciful ;  it  is  only  the  primitive  condition 
worths    which    are    susceptible    of    very    extensive   projection. 
I    hear    of    some    act    of   heroism    on    the  part    of    a   savage 
or  an  Oriental.     Up  to  a  certain  point  I  can  intuitively  realise 
his  attitude,  especially  if  it  is  an  act  of  heroism  in  connection 
with  the  fundamental  motives  of  family  or  state,  but  even  then, 
as  a  result   of  racial   differentiations   of  habits   of   expression 
and  thought,  only  incompletely,  and  in  the  case  of  more  differ 
entiated  attitudes,  not  at  all.     This  is  especially  true  in  the 
case  of  the  nuances  of  his  sense  of   honour.     Into  the  back 
ground  of  presuppositions  or  implicit  assumptions  which  deter 
mine  his  feelings  and  judgments,  I  can  penetrate  only  with  the 
greatest  difficulty,  for  they  have  been  created  by  activities  of 
sympathetic  participation  and  consequent  imaginative  and  ideal 
construction  differing  in  important  respects  from  those  which 
have  determined  my  own.     Social  selection  and  differentiation 
have  worked  differently  in  the  two  cases.     With  the  nature  and 
laws  of  this  selection  we  shall  be  concerned  in  another  connec 
tion ;  here  it  is  important  merely  to  note  the  fact  that  beyond 
purely  organic  sympathy,  sympathetic  participation  is  distinctly 
limited. 

In  view  of  these  facts,  it  can  be  easily  seen  that  the  judg 
ments  and  implicit  assumptions  of  existence  or  non-existence 


Personal  and  Over-Individual  Values          259 

of  more  extensive  social  dispositions  and  demands,  rest  less  and 
less  upon  immediate  organic  sympathy  and  intuitive  projection, 
and  more  and  more  upon  abstract  inference  and  judgment,  and 
that,  consequently,  the  feelings  of  participation  in  this  case 
become  progressively  more  and  more  impersonal.  It  may  be 
also  readily  seen  that  the  implicit  assumptions  or  expectations 
thus  generated  will  vary  widely,  both  in  the  character  of  the 
attitudes  or  dispositions  demanded  and  in  the  amount  expected, 
from  the  expectations  or  demands  generated  in  more  immediate 
personal  intercourse.  In  these  facts  we  shall  find  the  explana 
tion  of  those  differences  in  personal  and  social  obligation,  per 
sonal  and  impersonal  imputation,  reference  to  which  has  been 
made  at  the  beginning  of  this  chapter.  The  further  develop 
ment  of  these  differences  and  their  explanation  belong  to  suc 
ceeding  chapters. 


CHAPTER    IX 

I.  PERSONAL  WORTHS— THE  VALUES  OF  CHARACTERISATION 
OF  THE  PERSON 

i.  Definition :   the  Personal  Attitude  in  Valuation. 
THE   values    described   as   personal   constitute    a   well-marked 
sphere    of   meanings   in   our   worth   experience.     Between   the 
satisfaction  of  a  simple  sense-tendency,  and  its  corresponding 
condition  worth,  and  the  satisfaction  of  a  demand  more  deeply 
rooted  in  the  personality,  there  is  a  difference  which  is  immedi 
ately   appreciated.     No   less   clear   is   the   distinction   between 
the  value  of  an  act  as  a  quality  or  expression  of  the  personality 
and  as  merely  a  means  to  other  ends,  either  individual  or  social. 
If  we  had  occasion  to  criticise  the  statement  of  Lipps  that 
"  every    pleasure    is    conditioned    by   a   personality   worth," 
it  is,  nevertheless,  true  that  a  certain  class  of  worth  feelings 
presuppose  explicit  reference  to   the  ideal  or  concept  of 
person.     These  feelings   are   described   as   feelings   of  personal 

worth. 

The  objects  of  such  feelings  are  primarily  qualities  c 
positions   of  the  person.     It  is  true  that  physical  objects   of 
condition  worth  may  acquire  a  complementary  personal  value, 
but  this  is  possible  only  when  they  are  related  to  the  primary 
objects  through  secondary  associations  or  instrumental  judg 
ments      It  is  with  the  primary   objects   of  intrinsic  personal 
worth  that   we   are  here  concerned.     These  objects,  we  have 
seen,   are   ideal   constructions   developed   in    the   processes    of 
sympathetic  participation.     Through  the  processes  of  Emfuhl- 
ung      the     feeling     of    the     individual     acquires    a    common 
over-individual    meaning.     The    disposition    corresponding 
this   feeling,   now  presumed,   assumed,   or  judged   to   exist   in 
another,  has  acquired  a  new  meaning  and  value  through  1 
very  fact  of  participation,  of  its  being  shared.     But  wher 


1  Chap.  II,  pp.  5° 
260 


Personal  Worths  261 

disposition  is  thus  cognised,  and  has  this  subjective  "  participa 
tion-value,"  it  may  be  further  intrinsically  and  independently 
valued  as  an  expression  of  the  person,  the  self,  or  the  alter. 
This  is  its  personal  value  or  value  for  characterisation. 

The  attitude  toward  personal  worths,  of  the  self  or  the  alter, 
is  one  which  we  describe  in  terms  of  respect  or  admiration,  with 
their  opposites,  contempt  and  disdain.  We  clearly  distinguish 
self-respect  from  self-complacency,  and  self-disdain  from  self- 
pity  ;  and  respect  and  disdain  for  others  from  mere  liking  or 
disliking  and  pity.  The  reason  for  this  distinction  is  to  be  found 
in  the  fact  that  the  first  class  of  predicates  represents  personal 
worths,  the  second  condition  worths.  Pity,  like,  and  dislike 
are  simple  modifications  of  feelings  of  condition  worth,  adequately 
described  in  terms  of  pleasantness  and  unpleasantness.  It  is 
to  these  latter  attitudes  that  the  distinction  between  egoism 
and  altruism  alone  applies,  for  it  is  only  when  we  take  the  sub 
jective  feeling  as  end,  to  which  the  qualities  of  the  person  are 
related  as  means,  only  when  we  form  the  concept  of  our  own  or 
another's  happiness,  that  the  conflict  between  egoism  and  al 
truism  arises.  It  is  merely  this  subjective  aspect  of  "  condition 
worth  "  that  has  no  common  meaning,  and  that  cannot  be  shared. 
Self-complacency  and  self-pity  are,  properly  speaking,  egoistic, 
self-respect  and  self-disdain  are  not.  Pity  for  the  unhappiness 
of  another,  or  satisfaction  in  his  pleasure,  is  altruistic  ;  respect 
and  disdain  for  another  are  neither  egoistic  nor  altruistic.  The 
intrinsic  valuation  of  personality  transcends  the  distinction.1 

Equally  clear  is  the  distinction  between  the  personal  and  the 
impersonal  attitude  in  valuation.  Within  certain  limits,  which 
will  be  defined  as  our  study  proceeds,  respect  and  disrespect, 
admiration  and  disdain,  may  be  independent  of  the  moral  judg 
ment  of  good  and  bad.  Just  as  the  personal  attitude  transcends 
the  distinction  between  egoism  and  altruism,  so,  at  points,  it 
transcends  the  distinction  of  moral  goodness  and  badness  with 
which  they  are  closely  connected. 

The  preceding  study   of  feelings   of  personal  worth,   their 

1  The  feelings  of  hate  and  envy  which  are  so  often  the  accompaniments  of  the  re 
cognition  of  personal  worths  are  but  proofs  of  this  position.  We  cannot  hate  or  envy 
a  man  because  of  his  personal  worth.  This  is  an  acquired  worth  which  is  intrinsically 
recognised  as  good  and  desirable  for  us.  It  is  only  when  we  turn  our  eyes  to  the  hedonic 
accompaniments,  the  condition  worths  which  we  infer  to  accompany  the  personal  worth 
or  with  which  we  conceive  the  personal  worth  to  be  instrumentally  connected,  that  we 
feel  envy  and  hate.  Thus,  doubtless,  the  illiterate  Athenian  when  he  hated  to  hear 
Aristides  «z//«f  just,  was  always  thinking  of  the  rewards  of  the  virtue.  To  envy  a  man 
because  of  his  beauty,  strength,  or  virtue  is  essentially  a  backward  value  movement. 


262  Valuation:   its  Nature  and  Laws 

objects  and  the  qualities  predicated  of  the  objects,  makes  clear 
the  more  important  characteristics  of  these  feelings.  Their 
meanings  are  acquired  meanings,  and  their  objects  ideal  con 
structs.  Both  the  feelings  and  their  objects  presuppose  the 
development  of  a  certain  sphere  of  meaning  within  which  the 
judgments  of  characterisation  are  passed,  a  sphere  or  level  of 
meaning  which,  while  it  develops  by  certain  processes  out  of 
the  lower  level  of  feelings  of  condition  worth,  is  now  contrasted 
with  that  lower  level. 

Analysing  these  presuppositions  in  more  detail,  we  find 
that  they  are  certain  implicit  assumptions  with  which  the 
subject  of  the  feeling  comes  to  the  object.  These  assumptions, 
as  we  have  already  seen  in  the  studies  of  the  preceding  chapter, 
are  of  the  nature  of  expectations  of  the  existence  or  non-existence 
of  certain  dispositions,  certain  tendencies  to  desire,  feel,  and  act, 
on  the  part  of  the  self  and  the  alter.  These  assumptions,  when 
made  as  a  demand  upon  the  self,  give  rise  to  feelings  of  personal 
obligation ;  when  made  as  a  demand  upon  the  alter,  they  con 
dition  our  judgments  of  praise  and  blame.  In  general,  then,  the 
presupposition  of  the  sphere  of  meanings  described  as  personal 
worths  is  the  ideal  construct  of  the  'person,  individuated  as  the 
self  or  the  alter,  to  which  the  objects  of  the  feelings  are  referred. 

2.  The  Idea  or  Ideal  of  Personality :  Its  Meaning — How  assumed 
in  all  Judgments  of  Personal  Worth. 

In  saying  that  the  existence  of  the  person  is  assumed  im 
plicitly  in  all  feelings  and  judgments  of  personal  worth,  we 
have  introduced  the  concept  of  personality,  and  in  doing  so 
we  must  proceed  with  caution  if  we  are  not  to  go  beyond  the 
limits  of  psychological  method.  Whatever  may  be  said  as  to 
the  ultimate  metaphysical  reality  of  the  self,  it  is  not,  strictly 
speaking,  an  object  of  immediate  experience,  not  an  object  of 
perception,  nor,  on  the  side  of  feeling,  of  simple  appreciation, 
but  is  rather  a  construct  of  a  higher  order  built  up  upon  im 
mediate  perceptions  and  appreciations.  The  self  is  not  first 
there  as  an  object  and  then  characterised,  but  is  rather  an 
object  which  is  constructed  and  individuated  in  the  very  pro 
cesses  of  characterisation.  Constructed  first  for  practical  pur 
poses,  as  a  concept  for  the  regulation  of  our  expectations  of 
sympathetic  participation,  it  becomes  individuated  as  an  object 
with  intrinsic  value  and  meaning,  to  which  obligations,  responsi- 


Personal  Worths  263 

bilities,  merits,  and  demerits  may  be  imputed,  concepts  which 
stand  for  certain  acquired  meanings  of  feeling.  It  is,  first  of  all, 
a  worth  construction,  and  only  secondarily  an  object  of  knowl 
edge. 

Such  a  self  is  always  assumed  in  feelings  and  judgments  of 
personal  worth.  And  the  self  thus  assumed  is,  in  the  first  place, 
a  generic  and  ideal  self.  To  be  more  explicit,  the  ideal  self  is 
at  the  same  time  personal  and  over-individual.  The  specific 
disposition  assumed  has  acquired  an  additional  meaning  through 
reference  to  the  concept  of  the  person,  but  the  concept  of  the 
person  is  at  first  schematic,  and  not  individuated  into  the  ego 
and  the  alter  as  centres  of  unique  interests  and  inner  life.  In 
order  to  understand  this,  it  is  necessary  to  recall  the  conception 
of  the  "person-project"  developed  in  the  preceding  chapter.1 
There  it  was  pointed  out  that  the  development  of  this  ideal  pro 
jection  in  the  processes  of  sympathetic  Einfiihlung — the  passage 
of  simple  organic  sympathy  with  its  presumption  of  existence, 
into  feelings  of  the  imagination  in  sembling,  the  abstraction  of 
the  feeling  from  individual  presuppositions  and  its  schematic 
character  in  general — makes  it  at  once  over-individual  and  per 
sonal,  ready  to  be  read  either  into  the  self  or  the  alter,  and  ac 
knowledged  in  judgment  as  identical  with  the  self  or  the  other. 

II.  THE  CHARACTER  OF  THE  IDEAL  PERSON  AS  DETERMINED  BY 
THE  PROCESSES  IN  WHICH  IT  is  CONSTRUCTED 

i.  Idealisation  Involved  in  Sympathetic  Einfiihlung. 

The  character  of  the  presupposed  personality  is  determined  by 
certain  factors  inherent  in  sympathetic  projection  itself.  The 
first  of  these  may  be  described  as  the  tendency  to  idealise.  By 
idealisation  is  here  understood  the  tendency,  inherent  in  pro 
jection  in  its  imaginative,  semblant  mode,  to  enlarge,  so  to  speak, 
the  feeling  and  feeling  disposition  as  projected  and  assumed  to 
exist  in  the  "  other."  The  projected  feeling,  abstracted  from 
its  individual  presuppositions,  is  assumed  to  be  deeper  and 
broader,  more  completely  identified  with  the  person. 

Illustrations  of  this  fact  are  numerous.  Most  apparent 
in  undeveloped  persons,  and  differing  in  degree  with  differences 
in  temperament,  it  is,  nevertheless,  present  to  some  extent 
in  all  personal  relations.  The  savage  or  barbarian,  lost  in 

1  Chap,  viil,  p.  248. 


264  Valuation  :   its  Nature  and  Laws 

admiration  of  the  strength  and  pride  of  his  chief,  ignores  the 
shadows,  the  negative  factors  of  which  he  is  conscious  in  himself, 
and  it  may  be  added,  the  negative  condition  worths,  the  pains 
which  come  to  himself  through  this  superiority.  The  child 
worships  the  "  good  and  beautiful  lady."  The  simple  man 
believes  the  scholar's  knowledge  boundless.  More  developed 
persons  are  subject  to  the  same  illusion.  The  illusions  of  love 
will  suffice  for  our  purpose,  since  they  are  well  adapted  to  show 
the  psychological  source  of  the  illusion.  The  lover  feels  sym 
pathetically  the  love  of  his  adored,  but  as  the  result  of  the  very 
conditions  of  sympathetic  Einfiihlung,  i.e.,  abstraction  from  indi 
vidual  presuppositions,  it  is  a  purified  and  enlarged  love  that  he 
beholds.  In  himself  he  is  conscious  of  its  dependence  upon  organic 
and  other  presuppositions — he  is  conscious  of  its  admixture  with 
condition  worths,  in  this  case  lust— and  though  he  may  believe 
the  disposition  constant,  he  is  conscious  of  the  variations  in 
feeling.  Although  intellectually  he  might  infer  similar  con 
ditions  in  the  other,  intuitively  he  is  unable  to  project  them, 
and,  therefore,  ignores  them.  The  enlarged  purified  love  he 
beholds  is  assumed  to  have  a  corresponding  disposition— a  corre 
late  in  the  object  of  his  devotion,  and  that  disposition  is  assumed 
to  have  depth  and  breadth  in  the  personality  not  realisable  in 

himself.1 

This  imaginative  projection,  with  its  accompanying  ab 
straction  of  the  feeling  from  individual  and  limiting  presuppo 
sitions,  and  its  consequent  enlargement,  is,  however,  but  the 
first  stage  of  idealisation.  The  projected  feeling  has  both  an 
individual  and  an  over-individual  reference,  and  when  that 
feeling  is  sympathetically  realised  as  an  attitude  of  the  alter, 
it  is  again  read  back  into  the  self.  Idealised  in  the  sense  we 
have  described,  by  imaginative  projection,  it  is  again  referred 
to  the  self  in  its  idealised  form.  Lipps  has  described  the 
process  in  terms  which  I  can  do  no  better  than  quote  in  full. 
"  As  I  look  about  me  the  man  appears,  now  in  this  point,  now 
in  that,  increased  beyond  the  measure  found  in  myself.  That 
means,  as  we  know,  that  expressions  in  others  awake  in  me 
the  idea  of  an  increase  of  an  element  in  my  own  nature.  So 
arises  in  me  a  new  idea  of  personality  which,  just  in  so 

1  Lipps,  Die  ethischcn  Gntndfragen,  p.  44-     It  is  an  empirical  fact— upon  which 
Lipps  rightly  lays  considerable  weight  in  his  argument  for  the  recognition  of  mtr 
personal  worth  in  the  alter,  as  opposed  to  its  derivation  from  egoism— that  the  normal 
accompaniment  of  sympathetic  projection  is  the  contrasting  of  the  assumed  attitude  or 
disposition  in  the  "  alter  "  with  our  own,  in  the  direction  of  idealisation  of  the     alter. 


Personal  Worths  265 

far  as  it  represents  an  extension  of  my  real  personality,  is  in 
comparison  with  the  latter  an  ideal  personality.  There  arises 
in  me,  finally,  in  the  course  of  this  process  a  representation  of 
the  'ideal  personality."  * 


2.  Division  of  the  Personality — Extrusion  of  the  Negative  Moment 
Further  Stages  in  Idealisation. 

In  the  process,  as  thus  far  sketched,  we  have  but  the 
beginning  of  the  ideal  construction  of  the  personality.  The 
primary  contrast  between  worths  of  condition  and  of  the  person 
is  the  germ  of  the  idealisation  described,  but  the  contrast, 
and  the  idealisation  which  it  conditions,  do  not  stop  here.  The 
contrast  is  thus  far  between  the  imagined  and  the  actual  feeling, 
between  the  feeling  as  projected  into  the  alter  or  read  back 
into  the  self,  and  as  individual  feeling  with  its  individual  pre 
suppositions.  The  identification  of  this  ideal  project  with 
the  other,  or  in  a  return  movement  with  the  self,  involves  a 
further  step  in  the  process  of  characterisation,  which  may  be 
described  as  contrast  of  the  personal  worth,  now  identified  with 
the  self  or  the  alter,  with  the  condition  worths  now  conceived 
as  sub-personal.  This  leads  to  a  division  of  the  personality. 

The  division  of  the  personality,  whether  of  the  self  or  the 
alter,  is  a  well-known  phase  of  ideal  construction.  The  dis 
tinction  between  the  higher  and  the  lower  self  is  presupposed 
in  all  ethical  judgments.  The  analysis  of  the  situation  shows 
that  it  arises  out  of  an  explicit  acknowledgment,  as  an  opposition, 
of  what  was  at  first  merely  a  felt  contrast.  The  identification 
of  the  disposition,  as  a  positive  quality,  with  the  person,  requires 
some  corresponding  acknowledgment  of  the  opposing  negative 
tendencies,  and  this  gives  rise  to  the  intensification  of  the  primary 
contrast  in  a  conceptual  division  of  the  self.  The  contrast 
between  the  flesh  and  the  spirit,  between  desire  and  will,  are 
well-known  characterisations  of  the  situation.  In  such  cases 
the  objects,  the  "  spirit  "  and  the  "  will,"  acquire  a  new  mean 
ing,  and  are,  through  this  contrast,  more  completely  identified 
with  the  ideal  personality. 

We  should  fail,  however,  to  understand  the  real  signifi 
cance  of  this  stage  if  we  did  not  realise  that  it  may  be  but  germinal 
to  a  later  stage,  in  which  division  of  the  personality  develops 
into  complete  extrusion  of  the  lower  tendencies,  of  the  negative 

1  Lipps,  Die  ethischen  Grundfragen,  p.  37. 


266  Valuation :   its  Nature  and  Laws 

moment.  Such  a  stage  is  reached,  for  instance,  in  the  notion 
of  temptation  and  in  the  personification  of  the  tempting  ten 
dencies  in  the  form  of  a  hostile  personality.  The  extreme  of 
this  externalisation  we  have  in  certain  religious  personifications 
where  the  highest  personal  worths  are  identified  with  the  good 
God,  while  the  evil  is  projected  into  opposing  personalities  and 
forces.  The  individual  who  identifies  his  will  with  the  will  of 
God  frequently  acquires,  through  the  completeness  of  the  con 
trast,  the  absolute  value  of  complete  sanctification,  and  to  him 
is  imputed  by  other  persons  the  ideal  values  of  sainthood.  Similar 
heightening  of  the  sense  of  value  follows  when  the  negative 
factor  is  externalised  in  an  opposing  social  group,  as  when  the 
personal  worth  of  the  martyr  is  increased  through  contrast  with 
the  surrounding  evil.  As  a  child  of  his  age,  some  of  its  evil  is 
probably  in  him,  but  the  individuation  of  his  personality  abo\it 
some  supreme  personal  worth  as  a  centre,  brings  with  it  ex 
trusion  of  the  negative  elements.  Such  complete  contrasts 
are  commonly  fully  realised  only  in  the  isolation  of  the  person 
ality,  in  aesthetic  construction  where  the  negative  moments  are 
for  the  moment  ignored,  but  there  are  quasi-aesthetic  moments 
present  in  all  ideal  construction,  and  they  are  accompanied  by 
belief  in  the  ideal.  Even  purely  aesthetic  idealisations  may, 
as  we  shall  see  later,  under  certain  circumstances,  pass  over 
into  belief,  and  affect  our  actual  ethical  feelings  and  judgments. 

3.  Intuitive  (^Esthetic)  Individuation  of  the  Personality — Acquire 
ment  of  Complementary  Value. 

Thus  far  in  our  study  of  the  ideal  construction  of  the 
personality,  its  schematic  character  and  over-individual  aspects 
have  been  emphasised.  With  the  ideal  imaginative  projection 
of  an  individual  feeling-attitude  and  its  contrast  with  condition 
worths,  it  acquires  a  personal  reference  and  meaning.  But 
when  the  ideal  construct  of  the  person  is  thus  formed,  when 
a  quality  of  that  person  is  valued  intrinsically  as  an  expression 
of  the  person,  further  processes  of  individuation  appear,  and 
the  quality  or  qualities  in  question  acquire  complementary 
value  through  their  relations  to  each  other  as  parts  of  a  har 
monious  totality.  The  contrast  or  division  of  the  personality, 
and  ultimately  the  aesthetic  or  quasi-aesthetic  isolation  of  the 
person,  afford  the  necessary  conditions  for  that  individuating 
reconstruction  and  rearrangement  of  the  elements  of  the  person- 


Personal  Worths  267 

ality  which  makes  repose  in  the  object  possible,  and  gives  rise  to 
complementary  immanental  value. 

This  individuation  of  the  personality  is  of  the  intuitive  type 
of  construction  discussed  in  a  preceding  chapter,1  where  certain 
forms  of  this  individuating  construction,  both  perceptual  and 
ideational,  were  described,  and  their  fundamental  laws  deter 
mined.  In  particular,  we  considered  the  emergence  of  new 
objects  of  appreciation,  such  as  sensuous  beauty,  manners, 
cleanliness,  through  the  rearrangement  of  sensational  and 
perceptual  activities ;  also  the  individuation  of  abstract  ideal 
constructions,  such  as  a  sum  of  money,  where  the  sum  as  a 
unity  or  totality,  acquires  an  intrinsic  value  not  constituted  by 
the  instrumental  values  of  its  separate  elements.  The  in 
dividuating  construction  of  the  personality  does  not  differ  in 
principle  from  these  types,  but  merely  in  the  material  elements. 

The  function  of  the  law  of  contrast  in  building  up  the 
ideal  personality  is  primary.  In  fact,  the  other  two  laws  are, 
as  we  have  seen,  in  a  sense  expressions  of  it.  The  contrast  of 
a  quality  of  a  personality,  with  other  qualities,  by  which  it 
acquires  complementary  value,  may  be  seen  at  two  points. 
Such  a  quality  may  be  contrasted  either  with  the  opposite 
qualities  of  surrounding  persons,  or  with  opposing  qualities 
in  the  same  personality  which  have  been  or  are  being  over 
come.  Thus  the  holiness  of  a  martyr  stands  out  as  completely 
identified  with  his  personality,  in  contrast  to  the  universal 
corruption  about  him.  Augustine  is  all  the  more  a  saint  for 
the  opposition  of  his  complete  devotion  to  ideal  ends  in  later  life 
to  the  lower  feelings  and  desires  of  his  earlier  career.  Whether 
the  contrast  is  between  an  inner  and  a  more  external  self,  or 
between  the  self  and  society — in  either  case  the  contrast  enhances 
the  sense  of  reality,  and  therefore  of  the  value  of  the  person 
ality.  The  fondness  of  those  who  have  undergone  experiences 
of  conversion  for  contrasting  their  present  with  their  past,  arises 
from  the  increase  in  the  sense  of  personal  worth,  resulting  from 
complete  identification  of  present  ideals  with  the  personality, 
the  sense  of  elevation,  which  results  from  the  contrast. 

The  principle  of  the  "  total  series,"  which  was  seen  to  be 
so  important  in  aesthetic  construction,  has  an  important  role 
to  play  in  the  real  activities  and  judgments  of  personal  worth 
experience.  It  may  almost  be  described,  perhaps,  as  the  dra 
matic  tendency  in  the  characterisation  of  the  self  and  the  alter. 

1  Chap,  vi,  pp.  173  ff. 


268  Valuation:   its  Nature  and  Laws 

Unity  and  continuity  form  the  goal,  consciously  or  unconsciously, 
of  all  characterisations  of  personality.  The  ordering  of  the 
acts,  of  the  expressions  of  a  life,  as  they  appear  in  temporal 
relations  of  succession,  in  such  a  manner  that  they  have  a  quali 
tative  order  or  meaning,  as  determined  by  relations  of  teleo- 
logical  dependence,  is  one  of  the  chief  sources  of  the  comple 
mentary  values  imputed  to  personalities.  This  quasi-sesthetic 
characterisation  of  the  personality  is  inseparable  from  any 
reconstruction  of  our  own  past  or  construction  of  our  future, 
for  the  reason  that  the  self  is  primarily  a  worth  construction. 
It  is  equally  inseparable  from  the  biographical  and  historical 
reconstruction  of  other  personalities. 

Finally  we  may  note  the  important  role  of  the  principle  of 
"  end  feeling  "  in  this  characterisation  of  the  personality.  In 
general  the  tendency,  already  noted,  to  reconstruct  the  temporal 
relation  teleologically  through  processes  of  selection  and  ex 
clusion,  leads  to  the  selection  of  the  end  term  of  the  series  as 
the  keynote  of  the  whole,  and  as  the  chief  determinant  of  the 
character  of  the  whole.  Even  in  the  characterisations  of  the 
self  and  the  alter  of  ordinary  life,  this  ordering  of  the  elements 
of  the  total  character  under  the  end-moments  is  much  in  evidence. 
The  expression,  in  various  forms,  of  the  thought  that  no  one  is 
to  be  reckoned  happy  until  he  dies,  shows  the  emphasis  put 
upon  the  last  moments.  In  like  manner  the  importance  of 
the  last  moments  before  death  in  determining  the  judgment 
upon  a  person  as  a  whole  is  shown  in  the  emphasis  which  re 
ligion  puts  upon  the  "  making  of  a  good  end."  But  it  is  in  the 
aesthetic  characterisations  of  literature,  where  illusion-disturbing 
judgments  are  inhibited,  and,  to  a  degree  also,  in  the  quasi- 
aesthetic  characterisations  of  biography  and  history,  that  the 
working  of  this  law  of  ideal  construction  is  most  apparent, 
and  the  importance  of  the  principle  in  the  determination  of 
the  worth  imputed  to  the  personality  as  a  whole  is  best  shown. 
A  moment  of  supreme  manifestation  of  strength  or  self-sacrifice, 
at  the  end  of  a  relatively  meaningless  life,  may  give  it  a  supreme 
meaning  and  hallow  all  the  other  moments ;  may,  indeed,  through 
very  contrast  with  the  weakness  or  evil  of  the  former  acts, 
heighten  the  value  imputed  to  the  personality.  Finality  as 
purpose,  is  logically  independent  of  temporal  finality,  but  not 
for  intuitive  construction.  In  one  sense  it  does  not  matter  when 
the  chief  note  of  a  man's  life,  as,  for  instance,  his  heroic  moment, 
occurs,  but  in  another  sense  it  does.  It  is  a  timeless  value,  but 


Personal  Worths  269 

for  our  unification  of  his  character  it  makes  all  the  difference  in 
the  world  whether  the  heroic  act  came  early  and  was  followed 
by  mediocrity  or  weakness,  or  whether  a  meaningless  life  receives 
meaning  from  a  final  beautiful  act.  As  in  the  unities  of  the 
"  temporal  arts,"  music  and  the  drama,  so  in  the  aesthetic 
characterisation  of  the  person  we  seek  to  make  the  two  kinds 
of  "  ends  "  coincide. 

In  the  preceding  principles  of  characterisation  of  persons  are 
disclosed  the  laws  according  to  which  the  process  of  idealisation, 
begun  in  sympathetic  Einfiihlung  and  consequent  contrast  of 
person  and  condition  worths,  is  carried  on  in  the  intuitive  in- 
dividuation  of  the  person.  In  both  cases,  there  is  acquire 
ment  of  value.  In  the  first  stages  of  the  process  of  ideali 
sation  the  acquired  value  is  transgredient,  i.e.,  the  projection 
of  the  attitude,  its  acquirement  of  common  meaning,  and  its 
contrast  with  condition  worths,  issue  in  a  new  demand  which 
is  felt  in  the  self  as  personal  obligation,  and  with  reference  to 
the  alter  as  a  demand  for  intrinsic  personal  values.  The  acquired 
value  of  the  individuating  construction,  on  the  other  hand,  is 
immanental,  and  arises  from  repose  in  the  object,  the  unitary 
personality.  Here  the  values  are  partially  or  wholly  aesthetic, 
and  the  feelings  of  value  find  expression  in  the  aesthetic  predi 
cates  of  perfection,  nobility,  beauty  of  character,  etc.  In  the 
acquirement  of  this  complementary  immanental  value,  many 
qualities  of  the  person  are  significant  largely  because  of  their 
role  as  necessary  elements  in  the  unique  totality,  and  have 
only  personal  value,  being  without  instrumental  value  for  social 
ends. 

4.  Conclusions. 

The  general  conclusions  to  be  drawn  from  this  study  of 
the  processes  of  idealisation  of  the  person  may  be  thus  stated. 
In  this  process  a  new  meaning  is  acquired,  a  new  level  of  valua 
tion  formed.  Through  reference  to  the  ideal  of  the  person, 
and  through  contrast  with  condition  worths,  the  disposition 
becomes  the  object  of  feelings  qualitatively  different  from  the 
feelings  of  simple  appreciation.  But  not  only  is  this  qualita 
tively  new  meaning  acquired.  The  feelings  of  value,  with  these 
acquired  presuppositions,  have  greater  transgredient  and  im 
manental  reference,  greater  depth  and  breadth  in  the  personality. 
They  represent,  therefore,  an  absolute  increase  in  the  degree  of 
value  or  affective-volitional  meaning.  In  general,  personal 


270  Valuation  :   its  Nature  and  Laws 

worths  have  preference  over  condition  worths.  The  demand 
to  realise  personal  worths,  as  represented  in  feelings  of  "  per 
sonal  "  obligation,  is  more  intense  than  in  the  case  of  the  quasi- 
ethical  obligations  attached  to  objects  of  condition  worth. 
Personal  qualities  have  in  them  a  greater  capacity  for  con 
tinuous  valuation  than  objects  of  condition  worth.  These 
conclusions  we  shall  see  further  substantiated  in  our  studies  of 
personal  obligation  and  imputation  of  personal  worth  in  the 
following  chapter. 


III.  THE  LAWS  OF  VALUATION  AS  APPLIED  TO  OBJECTS  OF 
PERSONAL  WORTH 

i.  The  Problem. 

The  ideal  schematic  person,  the  existence  of  which  is  im 
plicitly  assumed  or  presumed  in  all  feelings  and  judgments  of 
personal  worth,  has  now  been  sketched  in  its  broad  outlines. 
It  is  seen  to  be  the  conceptual  term  for  a  system  of  dispositions 
or  affective-volitional  tendencies  which  is  assumed  to  exist, 
now  in  the  self,  now  in  the  alter.  As  such  it  is  the  necessary 
background  or  presupposition  of  the  entire  group  of  values 
which  we  call  personal.  It  was  further  shown  how  this  sche 
matic  ideal  may  be  individuated  by  the  individuating  principles 
inherent  in  ideal  construction,  and  how  the  elements  may  acquire 
complementary  value  as  part  of  a  unique  totality.  It  is,  there 
fore,  with  such  expectations,  assumptions,  demands,  that  the 
individual  who  judges  comes  to  the  objects  of  judgment,  and 
his  imputation  of  merit  or  demerit  is  the  expression  of  the  feelings 
of  satisfaction  or  dissatisfaction  following  upon  judgments  of 
existence  or  non-existence  of  the  expected  dispositions. 

These  assumptions  or  demands  are,  on  their  part,  expressions 
of  dispositions  generated  in  the  processes  of  sympathetic  par 
ticipation,  and  have  been  determined  by  the  selective  processes 
of  idealisation  inherent  in  such  participation.  When  this  fact 
is  duly  recognised,  it  becomes  apparent  that  our  description 
has  thus  far  abstracted  wholly  from  one  important  aspect  of  the 
feelings  of  personal  worth,  and  of  their  corresponding  judg 
ments,  namely,  the  aspect  of  quantity  or  degree.  Since  the  as 
sumptions  underlying  judgments  of  personal  worth  represent 
the  funded  meaning  acquired  in  processes  of  sympathetic  par 
ticipation  and  idealisation,  the  individual  comes  to  the  object 


Personal  Worths  271 

of  judgment,  not  only  with  a  pre-disposition  to  assume  the  ex 
istence  of  certain  qualities,  but  also  to  expect  these  qualities 
in  certain  amounts.  Consequently  the  feelings  and  judgments 
of  personal  worth  will  be  determined  by  the  degree  to  which 
these  expectations  are  realised.  The  relative  value  of  any 
quantity,  we  have  seen,  is  a  function  of  the  relation  of  that 
quantity  to  the  amount  of  the  presupposed  demand.  That 
demand,  again,  is  determined  by  the  dispositions  to  feel  or  desire 
created  by  previous  acts  of  valuation.  In  this  case  the  demand 
is  the  reflex  of  the  ideal  personality,  that  is,  the  funded  meaning 
of  the  dispositions,  acquired  through  reference  to  the  construct 
of  the  person.  Our  task  is  now  to  define  this  demand  in  quanti 
tative  terms,  to  determine  the  laws  of  valuation  of  these  ideal 
objects  of  personal  worth. 

2.  The  Problem  in  the  Light  of  our  General  Study  of  the  Laws  of 

Valuation. 

The  solution  of  this  problem  is,  in  a  sense,  merely  the  ap 
plication  to  a  special  question,  to  a  specific  class  of  worth  objects, 
of  the  general  principles  developed  in  the  chapter  on  the  Laws  of 
Valuation.     It  would  seem  that  in  defining  the  objects  of  per 
sonal  worth  as  ideal  and  intrinsic,  we  have  already  determined 
the   laws   of   their  valuation,  for  it  has   been   shown,  at  con 
siderable  length,  that  such  objects  have  the  capacity  of  con 
tinuous  valuation,  that  degree  of  value  increases  with  increase 
in  the  amount  of  the  object.     Objects  of  condition  worth  and 
of  instrumental  value  we  found  to  be  subject  to  the  law  of 
Limiting  Value,  and  the  concrete  judgments  of  value  in  these 
spheres  were  seen  to  reflect  the  working  of  this  law,  i.e.,  the  thres 
holds  and  limits  are  determined  by  it.     On  the  other  hand,  ideal 
intrinsic  objects,  with  the  capacity  for  individuation  and  ac 
quirement  of  complementary  value,  are  not  subject  to  this  law. 
From  this  conclusion  it  was  further  inferred  that  the  presuppo 
sition  or  postulate  of  judgment  in  this  sphere  is  the  possibility 
of  continuous  valuation  and  of  the  existence  of  absolute  values. 
With  regard  to  the  special  class  of  personal  worths,  it  would 
follow,  if  these  general  principles  are  true,  that  the  demand  for 
such  objects  is  unlimited,  that  the  assumption  underlying  our 
judgments  of  personal  worth,  our  judgments  both  of  obliga 
tion  and  of  imputation  of  merit  and  demerit,  would  be  that 
indefinite  increase   of   dispositions  with   personal  worth  means 


272  Valuation:    its  Nature  and  Laws 

continued  increase  in  degree  of  value.  It  would  be  "  rational," 
to  use  the  illustration  of  the  earlier  chapter,1  to  desire  the  in 
crease  of  such  dispositions  indefinitely;  and  our  judgments  of 
obligation  and  imputation,  in  reflecting  that  desire  or  demand, 
would  likewise  be  rational,  would  be  but  the  reflection  of  actual 
facts  and  laws  of  worth  experience. 

This  conclusion,  based  upon  considerations  of  a  general 
character,  we  shall  find  justified  in  the  sequel,  and  an  analysis 
of  our  actual  judgments  in  this  sphere  will  confirm  the  hypothesis. 
It  will  be  seen  that  our  judgments  of  personal  obligation  and  im 
putation  presuppose  this  postulate.  Nevertheless,  a  special 
analysis  of  sympathetic  participation  and  idealisation  from  this 
point  of  view  will  not  only  give  a  more  concrete  basis  for  this 
general  conclusion,  but  will  develop  certain  facts  necessary  for 
the  adequate  interpretation  of  judgments  of  personal  worth. 

3.  Feelings  of  Personal  Worth  as  Modified  by  the  Factor  of  Quantity 

of  the  Object. 

Analysis  of  the  Factor  of  Quantity. 

The  feelings  of  personal  worth  have  their  origin  in  im 
mediate  sympathy  and  sympathetic  participation,  which  is  in 
the  first  place  organic.  Out  of  this  develop  imaginative  pro 
jection  and  ideal  construction  of  dispositions,  with  their  as 
sumptions  and  judgments  of  existence.  It  is  clear  that  in 
our  investigation  of  the  effect  of  quantity  on  the  degree  of 
feeling  we  must  keep  before  us  this  difference  in  objects  and 
presuppositions  of  the  feelings.  In  the  case  of  simple  organic 
sympathy  the  factor  of  quantity  appears  in  two  forms:  as  repeti 
tion  of  sympathetic  participation,  and  as  intensity  of  the  feeling 
in  which  the  subject  sympathetically  participates.  When  the  level 
of  ideal  construction,  i.e.,  assumption  and  judgment,  is  reached, 
the.  object  is  not  the  immediate  emotional  expression,  but  the 
disposition  presupposed,  and  the  factor  of  quantity  is  different. 
The  quantity  of  the  disposition  may  be  displayed  in  two  ways  : 
either  by  repetition  of  the  expression  in  acts,  or  by  strength  of 
disposition  displayed  in  a  single  act.  In  the  first  case,  the 
quantity  of  the  disposition  is  measured  in  its  extent,  the  degree 
to  which  it  is  habitual  in  the  personality ;  in  the  second  case,  it 
is  measured  in  its  depth,  the  degree  to  which  the  disposition  is 
fundamental  in  the  personality. 

1  Chap.  VI,  pp.  186  ff. 


Personal  Worths  273 

(a)    Organic  Sympathy. 

Organic  sympathy  we  found  to  be  genetically  the  lowest 
level  of  sympathetic  participation.  The  effect  of  repetition  upon 
these  participation  feelings  is  the  ordinary  one  of  dulling  of 
sensitivity,  leading  ultimately  to  the  arrest  of  participation. 
Analysis  of  experience  leaves  no  doubt  of  this  fact.  It  is  ap 
parent  alike  in  aesthetic  Einfiihlung  where  the  objects  are  im 
personal  and  in  ethical  participation  in  the  feeling  of  persons. 
In  the  case  of  "  inner  imitation,"  where  the  inducing  conditions 
are  perceptual  movement-forms  of  nature,  repetition  is  followed 
by  the  dulling  of  the  organic  resonance,  as  was  clearly  illustrated 
in  the  studies  of  the  preceding  chapter.1  Still  more  is  this  true 
in  the  case  where  the  inducing  conditions  are  the  expressions 
of  persons.  In  organic  sympathy  with  the  joys  and  sorrows  of 
others,  such  sympathy  is  distinctly  limited  to  short  periods  and 
to  favourable  conditions,  and  loses  its  intensity  with  repetition." 
In  both  cases,  however,  there  is  the  possibility  of  substituting  for 
organic  sympathy  an  intuitive  realisation  of  the  feeling  —  more 
technically  expressed,  of  feeling  of  imagination  or  emotional 
abstract  —  in  the  attitude  of  sembling. 

The  dulling  of  sensitivity  with  repetition  has  as  its  parallel 
a  corresponding  effect  of  satiety  —  i.e.,  when  the  intensity 
of  the  emotional  expression,  to  be  sympatheticaly  realised, 
is  above  a  certain  normal  amount.  The  emotional  demand 
made  upon  the  sympathetic  person  by  an  extreme  of  joy  or 
sorrow,  by  unlimited  enthusiasm,  devotion,  or  sacrifice,  as 
exhibited  in  the  alter,  may,  as  far  as  the  organic  resonance  is  con 
cerned,  have  an  effect  entirely  analogous  to  satiety  in  the  sphere 
of  other  sensation-feelings,  and  especially  so  when  the  subject's 
temperamental  equipment  sets  a  limit  to  these  emotional  ex 
periences.  Here  again,  however,  the  limits  of  organic  sym 
pathy  do  not  necessarily  mark  the  limits  of  all  forms  of  sym 
pathetic  participation.  The  substitutes  for  the  full  emotional 
resonance,  the  feelings  of  the  imagination  in  sembling,  may 
notably  extend  the  range  of  our  participation.  In  aesthetic 
participation,  as  in  the  drama,  we  find  that,  if  certain  conditions 
are  met,  an  indefinite  increase  of  feeling,  as  expressed  by  the 
actor,  may  be  participated  in  sympathetically  by  the  spectator 
in  a  manner  and  degree  impossible  in  the  case  of  those  feelings 
described  as  "  real."  The  aesthetic  isolation  and  illusion  bring 

1  Chap.  VIII,  p.  240.  *  Chap,  vi,  p.  182. 


T 


274  Valuation:   its  Nature  and  Laws 

about  an  unusual  extension  of  our  capacities  of  sympathetic 
participation,  making  possible  a  complete,  though  temporary, 
identification  of  the  feeling  of  the  self  with  that  of  the  other 
and  the  identification  of  the  sympathetic  participant  with  the 
personality  as  dramatically  presented. 

(b)  Judgment  and  Assumption  Feelings. 

Organic  sympathy  is,  then,  subject  to  the  law  of  Limiting 
Value  as  determined  by  dulling  of  sensitivity  and  satiety.  But 
organic  sympathy  does  not  mark  the  limit  of  sympathetic 
participation.  There  are  forms  of  participation  in  which  the 
feelings  have  as  their  presuppositions  assumptions  and  judg 
ments.  Our  further  problem  is  clearly  to  determine  the  effect 
upon  these  dispositions,  and  upon  their  corresponding  implicit 
assumptions,  of  actualisation  of  feeling  through  these  cognitive 
acts.  What  is  the  effect  of  the  factor  of  quantity  in  this  sphere  ? 

Here  the  problem  is  at  first  sight  more  complicated,  but 
it  is  immediately  simplified  when  we  recognise  that,  whatever 
this  effect  may  be,  it  is  not  one  of  dulling  of  sensitivity  and 
satiety.  These  laws  do  not  apply  to  judgment  and  assumption 
feelings  as  such,  for  they  are  not  sensation-feelings.1  With 
this  negative  conclusion,  it  becomes  at  once  clear  that  we  are 
not  concerned  with  the  mere  mechanical  effects  of  repetition 
and  over-stimulation,  but  rather  with  the  question  of  the  limits 
of  what  we  have  called  judgment  capacity.2  When  the  quantity 
of  an  object  is  apprehended  in  acts  of  cognition,  the  value  of 
the  quantity  is  either  instrumental  or  intrinsic.  When  the 
feeling  of  value  is  mediated  by  instrumental  judgments,  it  is 
subject  to  the  law  of  Limiting  Value  for  reasons  developed  in 
the  chapter  referred  to,  and  which  need  not  be  repeated  here. 
This  law  would  apply  to  the  value  of  a  disposition,  in  so  far  as 
that  value  is  instrumental.  On  the  other  hand,  when  the  value 
is  intrinsic,  and  presupposes  the  assumption  of  the  existence  of 
an  individuated  whole  of  which  it  is  a  part,  the  law  of  Comple 
mentary  Values  becomes  operative,  and  the  capacity  of  the  object 
for  continuous  valuation  depends  entirely  upon  the  degree  to 
which  individuation  and  isolation  are  possible.  Whether,  then, 
there  is  increase  in  degree  of  value,  depends  upon  the  degree  to 
which,  with  increase  in  quantity  of  an  object  of  personal  worth, 
the  object  can  be  isolated  and  intrinsically  valued.  Whether 
these  complementary  values  determine  the  demand  for  personal 

1  Chap,  vi,  p.  159.  a  Chap,  vi,  p.  171. 


Personal  Worths  275 

worths  presupposed  in  our  actual  judgments  of  imputation  and 
obligation,  depends  upon  the  degree  to  which  such  intrinsic 
valuations  modify  the  dispositions  presupposed  by  these  judg 
ments. 


4.    The  Effect  of  Idealisation  on  our  Actual  Judgments  and 
Judgmental  Dispositions. 

(a)  Idealisation  as  Imaginative  Construction. 

This  intrinsic  valuation  is  possible,  however,  only  on  the 
condition  of  aesthetic  or  quasi-aesthetic  isolation  of  the  person 
ality,  described  in  the  study  of  the  processes  of  idealisation. 
The  preponderating  role  of  imaginative  projection,  with  its 
contrast  and  individuation,  in  the  process  of  idealisation  has 
been  emphasised.  Do  these  feelings  of  the  imagination  or 
assumption  feelings  enter  into  the  formation  of  our  permanent 
beliefs,  or  implicit  assumptions  of  reality  ?  Evidently  the  vital 
question  in  the  present  discussion  concerns  the  effect  of  these  feel 
ings  of  imagination  upon  our  dispositions  to  actual  feeling,  and 
therefore  upon  our  worth  judgments.  This  is,  of  course,  but 
another  aspect  of  the  general  question  fully  discussed  in  an 
earlier  chapter.1  There  we  insisted  that  feelings  of  the  imagina 
tion,  as  well  as  judgment  feelings,  modify  our  feeling  dispositions, 
our  implicit  assumptions,  and  are,  therefore,  of  functional 
importance  in  the  processes  of  valuation.  Here  we  might  apply 
that  conclusion  without  further  analysis  of  the  facts,  but  our 
development  of  the  present  study  will  be  more  satisfactory  if  we 
make  an  independent  analysis  of  aesthetic  participation. 

(b)  The  Effect  of  the  Semblant  Mode  in  Simple  Appreciation. 

The  effect  of  the  semblant,  aesthetic  mode  in  simple  appre 
ciation  has  already  been  shown,  but  a  reconsideration  of  the  facts 
at  this  point  will  enable  us  to  understand  better  its  role  in  the 
characterisation  of  persons.  Cognitively  considered,  the  attitude 
was  found  to  be  one  of  conscious  self-illusion  in  that  in  most 
experiences  of  the  aesthetic  type,  more  particularly  artistic 
creation  and  appreciation,  the  elimination  of  illusion-disturbing 
moments  is  a  conscious  process,  and  the  judgments  which  would 
destroy  the  illusion  are  never  completely  inhibited.  Neverthe 
less,  even  here  the  distinction  between  reality  and  illusion 

1  Chap,  v,  pp.  137-9. 


276  Valuation:   its  Nature  and  Laws 

cannot  be  pressed  too  far.  The  aesthetic  mode  of  appreciation 
is  genetically  related  to  other  worth  attitudes,  in  that  it  is  a 
value  movement  toward  activity,  in  which  assumption  takes 
the  place  of  judgment,  and  in  which  desire  becomes  dispositional. 
But  precisely  because  of  this  genetic  relation,  the  immanental 
value  acquired  in  the  movement  modifies  our  actual  feelings  of 
value,  as  is  seen  in  the  phenomena  of  imputed  value.1  Even  in 
art  the  assumption  feelings  are  not  without  effect  upon  our 
actual  desires  and  feelings.  In  such  a  work  of  art  as  Tennyson's 
Lotos-Eaters — a  well-nigh  perfect  illusion  of  mood  where,  by  the 
elimination  of  all  illusion-disturbing  factors,  the  dominant  mood 
of  peace  and  forgetfulness  is  completely  realised — it  is  quite 
possible  that  its  effect  may  persist  in  non-aesthetic  attitudes. 
It  is  impossible,  it  is  true,  that  the  momentary  assumption  of 
the  existence  of  such  a  land  of  dreams  should  pass  over  into 
actual  belief  in  its  existence,  but  since  it  was  a  real,  though  dis 
positional  desire  that  was  temporarily  satisfied,  the  illusion  may 
have  its  effect  as  an  undertone  in  determining  actual  feelings 
and  judgments.  Such  dreams  not  only  create  a  belief  which 
leads  to  its  own  realisation,  as  wrien  Columbus  said,  "  It  was 
not  astronomy  or  geometry,  but  his  reading  Isaiah's  prophecy 
of  a  new  heaven  and  a  new  earth,  that  set  him  on  his  discovery," 
but  they  may  also  form  the  basis  of  a  critical  judgment  of  actual 
conduct  and  life,  which  may  lead  to  the  realisation  of  the  new 
heavens  and  the  new  earth  of  the  reformer. 


(c)  The  Effect  of  the  JEsthetic  in  the  Characterisation  of  Persons. 

When  we  pass  to  the  role  of  imagination  in  the  character 
isation  of  persons,  and  its  effect  upon  the  implicit  assumptions 
underlying  our  personal  judgments,  the  distinction  between 
imagination  and  belief  is  still  more  vague.  In  our  study  of 
ethical  and  aesthetic  projection,  it  was  found  that  the  difference 
lies,  not  in  the  projected  feeling,  but  in  the  secondary  partici 
pation-feeling  following  upon  the  judgment  or  assumption  of 
the  existence  or  non-existence  of  a  dispositional  correlate.  In 
the  case  of  Lear  there  cited,  the  projected  feeling,  as  feeling, 
may  be  more  fully  realised  than  in  actual  life.  It  is  the  im 
pulse  or  disposition  to  participate  that  is  at  a  minimum,  and 
therefore  the  corresponding  feelings  of  participation  may  be 
said  to  lack  reality.  They  are  assumption-feelings.  Neverthe- 

1  Chap,  vii,  pp.  229  ff. 


Personal  Worths  277 

less  these  assumption-feelings  may  pass  over  into  judgment- 
feelings  in  several  significant  ways.  The  vulgar  tendency  to 
take  the  passions  of  the  stage  as  real,  and  sometimes  to  act 
accordingly,  shows  this  relation  of  assumption  to  judgment  in 
a  crude  form.  The  ideal  limiting  case  of  the  aesthetic  attitude, 
desireless  intuition,  is  but  imperfectly  realised.  Will  is  present 
dispositionally,  ready  to  flare  up  upon  the  crudest  semblance  to 
reality.  Much  more  significant,  however,  for  the  larger  life 
of  worth  experiences  is  what  may  be  described  as  the  "  after- 
feeling  of  reality,"  the  conviction  or  judgment,  that  such 
situations,  such  passions,  etc.,  are  actually  real.  We  have  here, 
then,  the  curious  situation,  the  meaning  of  which  has  not  been 
fully  appreciated,  that  a  feeling,  the  character  of  the  presuppo 
sitions  of  which,  as  assumptions,  was  fully  realised  at  the  time, 
may  gradually  pass  over  into  a  real  feeling  with  judgment  as  its 
presupposition.  Phenomena  of  this  sort  are  not  far  to  seek. 
An  aesthetic  realisation  of  idealised  passions  and  emotions  may 
generate  expectations  which  colour  the  actual  judgments  of 
real  life.  Similarly,  a  man  may  come  to  believe  in  his  own  dreams, 
which  at  first  were  recognised  as  dreams.1 

(5)  Absolute  Personal  Values :    They  exist  as  Practical  Absolutes. 

The  self  is  an  ideal  construction  of  an  individuating 
character  involving  contrast,  serial  order,  and  totalisation 
through  the  idea  of  finality.  The  working  out  of  this  con 
struction  involves  an  isolation  and  extrusion  of  the  "  negative 
moments,"  possible  only  in  aesthetic  idealisation.  In  such 
idealisation  absolute  values  are  realised,  that  is,  situations  appear 
where  the  object  of  personal  worth  is  completely  identified  with 
the  personality ;  the  elimination  of  opposing  elements  is  absolute 
and  the  ideal  of  perfection  is  realised. 

1  The  nature  of  this  transition  from  aesthetic,  imaginative  construction  of  a  person 
ality,  with  its  mere  assumption  of  reality  and  its  conscious  "self-illusion,"  to  actual 
belief  is  well  illustrated,  in  an  extreme  form,  in  some  cases  of  illusions  of  mediumship. 
In  that  interesting  and  instructive  study  of  mediumship,  M.  Flournoy's  A  Journey  to 
Mars,  it  would  seem  to  be  established  that  the  creation  of  secondary  personalities  was 
preceded  by  periods  of  incubation,  in  which  an  ideal  fictitious  personality  was  being  con 
structed  from  materials  got  from  reading,  conversation,  etc.  Gradually  there  was  a 
systematisation  of  tendencies  and  attitudes  about  a  fundamental,  and  an  accompanying 
arrest  of  illusion-disturbing  tendencies.  Finally  the  segregation  became  so  complete 
that  the  system  received  a  new  name  and  became  a  new  personality.  The  transition 
from  assumption  to  judgment  with  its  accompanying  belief  was  gradual.  What  the 
trance  contributed  to  the  situation  was  simply  that  it  afforded  the  conditions  for  com 
plete  auto-suggestion,  arresting  all  tendencies  which  would  be  illusion-disturbing,  which 
would  again  transform  belief  into  assumption. 


278  Valuation:   its  Nature  and  Laws 

In  actual  ethical  characterisation  this  ideal  is  only  im 
perfectly  realised.  The  ever-recurrent  division  of  the  self  into 
the  lower  and  the  higher  (the  Kantian  contrast  between  the 
empirical  and  the  rational,  intelligible  will),  is  a  necessary 
condition  of  ideal  construction  and  of  ethical  judgment,  but  is 
normally  incomplete.  The  complete  identification  of  the  ideal 
object  of  personal  worth  with  the  personality  is  merely  an  ideal 
to  be  attained,  because  of  the  difficulty  of  complete  extrusion 
of  the  negative  moment.  There  are,  however,  unique  experi 
ences,  both  in  the  processes  of  self-realisation  and  in  the  charac 
terisation  of  the  alter,  where  the  illusion  of  complete  identification, 
of  perfection,  which  is  ordinarily  possible  only  in  purely  aesthetic 
experiences,  becomes  part  of  the  real  ethical  process,  where  the 
aesthetic  assumption  acquires  the  conviction  of  judgment. 
These  we  may  describe  as  practical  absolutes. 

The  situations  where  these  unique  experiences,  these  supreme 
moments,  appear  are  those  in  which  the  contrast  between  the 
individual  and  the  surrounding  social  values,  or  between  the 
ideal  personal  worths  and  the  lower  condition  worths,  is  so 
complete  that  the  negative  moments  are  wholly  externalised, 
and  the  personality  is  completely  unified  through  identification 
with  the  ideal  object ;  the  aesthetic  isolation  and  illusion  is 
complete.  This  means  that  psychologically,  for  immediate 
experience,  there  are  absolute  personal  worths — however  these 
may  be  judged  from  a  more  objective  impersonal  and  over- 
individual  point  of  view.  The  ideal  constructions  of  religious 
experience  and  the  identification  of  the  individual  with  these 
constructions  are  cases  in  point.  The  supreme  sacrifice  of 
Christ  becomes  an  object  of  belief,  and  the  identification  of 
the  individual  will  with  his  will  has,  at  least  in  exceptional 
cases,  produced  the  experience  or  illusion  of  complete  holiness. 
Now  the  interesting  feature  of  these  religious  experiences  is 
that,  while  like  the  aesthetic  they  depend  upon  isolation  (Christ 
is  one  with  the  Father  and  the  believer  is  "  hid  with  Christ  in 
God  "),  in  the  religious  experiences  there  is  such  complete  arrest 
of  all  "  illusion-disturbing  "  factors  and  all  opposing  elements, 
that  assumption  passes  over  into  belief.  They  afford  momentary 
realisations  of  supreme  values  which  negate  all  questions  of 
possibility  and  probability,  and  all  such  reference  beyond  the 
moment  as  characterises  ethical  and  moral  judgments.  But 
the  point  should  be  emphasised  that,  while  these  experiences  are 
conditioned  by  aesthetic  and  quasi-aesthetic  detachment  or  isola- 


Personal  Worths  279 

tion,  they  continue  to  have  effect  beyond  these  momentary  realisa 
tions.  The  aesthetic  and  religious  assumptions  generate  expecta 
tions  which  function  as  ideal  norms  in  the  ethical  judgments  of 
obligation  and  imputation.  The  question  at  this  point,  it  will  be 
observed,  is  not  whether  the  subject  of  these  experiences  has  or 
has  not  the  capacity  for  actual  worth  feeling  of  this  character, 
or  whether,  indeed,  individuals  in  actual  life  have  the  capacity 
for  displaying  such  ideal  dispositions,  but  merely  whether 
psychologically  the  assumption  generated  in  the  aesthetic  ex 
perience  modifies  actual  belief  and  judgment. 

It  now  remains  to  study  the  points  at  which  these  psych 
ological  absolutes  are  found,  and  to  show  the  psychological 
conditions  in  which  they  are  realised.  They  are  what  may  be 
called  Tragical  or  Heroic  Elevation,  and  Inner  Peace  and 
Harmony.  The  first  appears  as  the  limiting  case  of  trans- 
gredient  worth,  the  latter  as  the  limit  of  immanental  worth. 
Tragical  elevation  may  appear  at  two  points  :  either  where 
for  an  attitude  which  the  person  has  completely  identified  with 
the  self,  and  which  is  therefore  a  personal  worth,  all  condition 
worths — including  life,  which  is  the  presupposition  of  condition 
worths — are  sacrificed  ;  or,  secondly,  where  the  individual  sets 
himself  in  complete  opposition  to  external  worth  judgments  of 
society  and  goes  to  destruction  for  the  worth  which  he  identifies 
with  himself.  The  point  here  made  may  be  illustrated  by 
saying  that  absolute  personal  worth  has  been  realised  even  if 
the  object  for  which  the  sacrifice  has  been  made  is,  from  the 
impersonal  point  of  view  of  instrumental  judgment,  not  con 
sidered  worth  the  sacrifice.  Psychologically  viewed,  the  com 
plete  unity  of  the  personality  thus  attained  is  a  product  of  com 
plete  contrast  or  opposition  in  which  the  central  quality  of  the 
personality  is  so  emphasised  that  the  minor  elements  are  lost 
sight  of,  have  become  irrelevant,  as  have  also  the  secondary 
judgments  as  to  the  effects  of  the  act.  The  leader  who  cham 
pions  to  the  death  a  lost  cause  is  the  object  of  absolute  imputed 
worth,  irrespective  of  the  effect  of  the  sacrifice — even  if  the 
cause  was  evil  and  should  have  been  lost.  In  a  similar  manner 
the  imperfections  and  weaknesses  of  the  man  are  ignored  or 
felt  to  be  wiped  out  by  the  final  act.  It  is  the  "  end  feeling  " 
that  gives  the  tone  to  the  whole  life.  As  long  as  the  sacrifice 
and  opposition  is  not  complete,  our  judgments  upon  the  man 
as  a  personality  are  complicated  by  secondary  judgments  as 
to  the  existence  of  negative  elements  and  their  social  effects. 


280  Valuation :    its  Nature  and  Laws 

But  let  the  sacrifice  or  opposition  become  complete,  and 
these  secondary  judgments  lapse.  The  person  is,  in  very 
truth,  elevated  above  them  ;  they  become  externalised  and 
irrelevant. 

In  a  similar  fashion,  and  corresponding  to  the  ideal  of  the 
"  heroic,"  the  ideal  of  the  "  beautiful  soul,"  or  inner  harmony, 
represents  an  ideal  construction  of  the  personality  in  which  all 
disturbing  moments  are  eliminated,  and  in  which  there  is  the 
repose  of  satisfied  conation.  Here,  again,  we  have  what  may  be 
described  as  an  absolute  moment  in  valuation,  but  in  this  case 
the  passing  of  the  relative  into  absolute  value  arises,  not  from  a 
complete  identification  of  the  disposition  with  the  personality 
through  one  supreme  moment  of  effort  and  sacrifice  transcend 
ing  all  relative  estimation,  but  rather  from  the  complete  identi 
fication  which  comes  with  repetition  and  habit. 

The  realisation  of  either  of  these  moments  is  possible  obviously 
only  through  the  quasi-aesthetic  isolation  of  the  personality,  the 
conditions  of  which  have  already  been  described.  Whether 
realised  as  a  feeling  of  absolute  personal  value  in  the  sublime 
moments  of  obligation,  or  as  a  sense  of  perfection  of  another 
revealed  in  glimpses  of  absolute  sacrifice  or  perfect  harmony 
of  character,  such  experiences  rest  upon  assumptions  which 
are  made  possible  only  by  abstracting  from  the  causal  and 
instrumental  point  of  view.  Such  assumptions  or  postulates, 
with  regard  to  the  possibility  or  actuality  of  absolute  personal 
worths,  may,  accordingly,  fail  of  justification  from  other  points 
of  view,  which  include  more  general  theoretical  considerations. 
But  the  importance  of  these  moments  from  the  point  of  view 
of  the  present  discussion  is  to  be  found  in  the  fact  that  they 
are  practical  absolutes,1  so  to  speak,  points  where  conation,  and 
with  it  all  relative  valuation,  comes  to  rest.  Though  realised 
only  when  the  isolation  of  the  individual  is  assumed  in  aesthetic 
experiences,  they  create  expectations  which  determine  the  norms 
and  standards  of  actual  ethical  judgments  of  personal  worth, 

1  For  a  similar  use  of  the  term  "  absolute  "  compare  Simmel's  Philosophic  des  Geldes, 
p    213   where  he  distinguishes  between  absolute  and  relative  ends  :  "Absolut— m  dem 
hier  fraglichen,  practischen  Sinne— ist  der  Wert  der  Dinge  an  denen  em  Willensprozess 
definitiv  Halt  macht."     The  use  of  the  term  "practical  absolutes'   to  describe  these 
moments  of  tragical  elevation  and  inner  peace  should  be  emphasised.     They  are  practi 
cal  in  the  sense  that  the  objects,  the  belief  in  the  existence  of  which  gives  rise  t 
satisfaction  of  conation  beyond  which  relative  increase  is  impossible,  exist  Jar  the  pro 
cesses  of  valuation,  the  practical  activities  of  feeling  and  will  for  which  they  are  ends, 
but  not  necessarily  apart  from  these  processes.     They  are  moments  in  which  mdividi 
processes  of  feeling  and  conation  come  to  complete  fruition,  but  Irom  a  more  objective, 
impersonal  point  of  view,  this  belief  in  absolute  objects  might  appear  illusory. 


Personal  Worths  281 

as  the  succeeding  studies  of  these  judgments  will  show  We 
may,  then,  conclude  this  discussion  by  affirming  that  objects 
of  personal  worth  may  acquire  absolute  value,  in  so  far  as  they 
are  intrinsically  valued  as  qualities  of  an  individual,  and  that, 
moreover,  the  implicit  assumption  which  underlies  all  judgments 
of  personal  worth  is  precisely  this  belief. 


CHAPTER    X 

\ 

PERSONAL  WORTHS   (Continued) 

I.  INTERPRETATION  OF  THE  CONCRETE  JUDGMENTS  OF  PERSONAL 
WORTH  IN  TERMS  OF  THE  PRECEDING  THEORY  OF  THEIR 
ORIGIN  AND  NATURE 

i.  The  Problem. 

IT  has  been  maintained  that  the  feelings  and  objects  of 
personal  worth  constitute  a  well-defined  region  of  worth  ex 
perience,  and  the  studies  of  the  preceding  chapter  have  gone 
far  to  justify  this  view.  The  valuation  of  persons  as  persons 
constitutes  a  relatively  independent  type,  one  which  presupposes 
a  differentiation  of  object  and  attitude,  and  which  is  charac 
terised  by  specific  presuppositions  and  postulates.  If  this 
view  is  justified,  we  should  expect  to  find  these  conclusions  sub 
stantiated  by  a  detailed  analysis  of  the  actual  judgments  in 
this  sphere — the  judgments  of  personal  obligation,  and  of  im 
putation  of  personal  merit  and  demerit,  already  distinguished. 
We  should  expect  these  judgments  to  disclose  certain 
empirical  uniformities,  both  in  the  qualitative  predicates 
employed  and  in  the  quantitative  aspects  of  the  judgment,  i.e., 
in  the  way  in  which  the  degree  of  value  varies  with  the  amount 
of  the  object  of  value.  Such  uniformities  a  careful  analysis 
of  the  phenomena  discloses,  and,  when  properly  interpreted, 
they  are  seen  to  reflect  the  laws  of  valuation  of  personal 
worths  already  developed.  To  such  analysis  and  interpreta 
tion  we  must  now  turn,  seeking  in  the  facts  of  judgment  a 
proof  or  disproof  of  our  theory.  The  problem  of  this  chapter  is, 
accordingly,  the  formulation  and  interpretation  of  the  empirical 
laws  of  characterisation  of  persons  and  of  estimation  of  personal 
worth. 

The  first  condition  of  such  a  study  is  obviously  the  isolation 
of  the  phenomena  in  question,  the  judgments  of  personal  worth, 

282 


Personal  Worths  283 

of  personal  obligation  and  merit  or  demerit,  from  other  types 
of  judgment.  In  order  to  interpret  adequately  these  judgments 
it  will  be  necessary  to  differentiate :  (i)  the  objects  of  judgment, 
and  (2)  the  terms  and  predicates  in  which  the  characterisation 
and  estimation  take  place,  from  the  objects  and  predicates 
in  other  types  of  worth  judgment.  It  is  further  necessary  to 
differentiate  the  norms  or  standards,  the  expectations  or  implicit 
assumptions  with  which  the  judging  subject  comes  to  the 
characterisation  of  the  person,  and  to  the  estimation  of  personal 
worth,  from  similar  norms  in  other  types  of  estimation.  If 
such  differentiation  of  attitude,  such  isolation  of  phenomena 
is  possible,  we  may  hope  to  account  for  the  worth  judgments 
in  question  by  tracing  them  back  to  the  processes  of  sympathetic 
participation  and  ideal  construction  in  which  the  objects,  predi 
cates,  and  presuppositions  arose. 

2.  The  Objects  of  Personal  Worth. 

The  distinction  between  objects  of  "  condition "  and 
"  personal  "  worth  is  implicit  in  our  experience  prior  to  any 
reflective  distinction  between  egoism  and  altruism.  It  has 
already  been  shown  that  there  is  a  well-defined  sphere  of  in 
trinsic  appreciation  of  dispositions  and  qualities  of  the  person, 
quite  apart  from  the  estimation  of  the  utility  values  of  those 
dispositions  or  qualities,  as  instrumental  to  condition  worths 
of  the  subject.  The  objects  of  personal  worth  are  ideal  objects 
which  have  acquired  a  common  over-individual  meaning  through 
the  processes  of  sympathetic  Einfiihlung  and  ideal  construction ; 
and  therefore  have  acquired  a  further  complementary  value 
through  reference  to  the  individuated  whole,  the  person.  It 
is  as  a  quality  or  expression  of  this  whole  that  it  has  its  meaning 
and  value.  We  have  already  traced  the  processes  in  which 
the  objects  of  personal  worth  are  constructed  and  contrasted 
with  condition  worths,  and  need  not  repeat  this  here.  It  is 
sufficient  to  emphasise  the  fact  that  this  contrast  of  objects 
of  personal  with  objects  of  condition  worth  is  present  in  all 
concrete  worth  judgments,  and  will  presently  appear  in  our 
study  of  the  predicates  employed  in  the  characterisation  of 
persons  and  in  the  quantitative  estimation  of  personal  worth. 
The  objects  of  personal  worth  judgments  are,  then,  qualities 
or  dispositions  of  the  person  ;  and  value,  in  the  form  of  merit  or 
demerit,  is  imputed  to  the  person  on  the  basis  of  possession  or 


284  Valuation  :   its  Nature  and  Laws 

non-possession  of  these  qualities.  But  there  is  a  further  differ 
entiation  in  concrete  worth  experience  which  must  be  taken  into 
account  in  our  isolation  of  personal  worth  judgments,  one  to 
which  more  detailed  study  must  be  given.  These  ideal  objects, 
qualities,  and  dispositions  may  be  judged  from  two  distinct  points 
of  view,  the  personal  and  the  o^r-personal  or,  in  its  extreme 
form,  impersonal.  As  an  object  of  'personal  value  the  quality  or 
disposition  is  judged  intrinsically,  as  part  of  the  ideal  whole,  the 
person.  In  personal  judgment,  the  judging  subject  abstracts 
from  all  reference  of  the  disposition  to  social,  over-individual 
ends,  from  the  instrumental  value  which  the  disposition  in 
question  has  for  the  ends  of  social  participation.  The  personal 
judgment,  as  such,  presupposes  the  isolation  or  detachment  of 
the  person  from  social  references,  and  the  characterisation  of 
him,  and  estimation  of  his  value,  in  the  light  of  expectations 
generated  in  immediate  sympathetic  participation  and  ideal 
characterisation.  In  the  over-personal  attitude,  on  the  other 
hand,  the  subject  abstracts  from  just  these  personal  references 
and  meanings  of  the  quality  or  disposition,  and  ultimately,  by 
processes  to  be  described  later,  reaches  a  relatively  impartial  or 
impersonal  point  of  view  in  which  his  judgments  are  determined 
by  the  demands  or  expectations  of  a  wider  social  consciousness. 

This  difference  comes  out  strikingly  in  certain  charac 
teristic  facts  which  will  be  immediately  recognised.  We  may 
describe  them  in  general  terms  as  the  shifting  of  the  centre  of 
interest  or  attention,  the  affirmation  of  one  system  of  assump 
tions,  and  the  negation  or  inhibition  of  another.  In  all  those 
cases  where  the  specifically  personal  attitude  is  uppermost, 
and  where  the  intrinsic  ethical  or  aesthetic  predicates  of  respect 
and  admiration  are  called  out,  our  characterisation  of  persons 
almost  constantly  puts  the  qualities  which  we  may  describe 
as  "  lovable,"  the  purely  personal  quasi-ethical  qualities,  before 
the  more  moral  attributes  with  their  larger  social  reference. 
So  also  the  more  aesthetic  qualities,  such  as  harmony  or  strength 
of  character — often  irrespective  of  the  acts  in  which  they  are 
shown — may  take  precedence  of  the  more  moral,  directly  social 
virtues.  Limiting  cases  appear  where  personal  devotion  and 
admiration,  not  to  say  worship,  exists  with  almost  total  sup 
pression  of  moral  judgment.  The  fact  seems  to  be  that  there  are 
some  personal  qualities  the  reference  of  which  to  wider  social 
values  is  the  most  remote  and  indirect,  others  which  have  a 
significance  both  for  personal  and  social  judgment,  and 


Personal  Worths  285 

still  others,  perhaps,  almost  wholly  impersonal  and  social  in 
their  reference.  The  important  point  is  that  the  demands 
or  assumptions,  in  personal  participation,  differ  in  significant 
ways  from  those  which  represent  the  subject  in  his  capacity 
of  participant  in  the  larger  demands  of  society.  Such 
isolation  of  the  person  is,  nevertheless,  always  relative  ;  it  ap 
proaches  to  completeness  only  in  the  activities  of  aesthetic 
characterisation,  where  the  conditions  of  detachment  and  isola 
tion  are  most  favourable,  but  it  is  present  as  a  determinant  in 
all  concrete  judgments  of  personal  worth. 

For  this  distinction  between  the  personal  and  impersonal 
attitudes  in  judgment  the  terms  ethical  and  moral  have  been  used. 
This  distinction  corresponds  fairly  well  with  popular  usage,  the 
terms  moral  and  immoral  being  employed  with  reference  to  those 
standards  which  are  universal  and  impersonal — what  Kant  de 
scribed  as  the  region  of  perfect  obligation ;  while  ethical  and  un 
ethical  are  employed  to  designate  that  larger  and  more  indeter 
minate  region  of  differentiated  personal  ideals,  but  of  imperfect 
social  demand.  We  shall  presently  see  in  detail  how  the  personal 
and  impersonal  judgment  of  the  same  disposition  differ  both 
qualitatively  and  quantitatively  as  a  result  of  this  differentiation 
of  attitude,  but  before  considering  this  question  we  must  make 
a  preliminary  study  of  the  terms  or  predicates  employed  in 
judgments  of  personal  worth. 

3.  The  Terms  of  Estimation  of  Personal  Worth. 

The  terms  in  which  a  person  is  characterised  are,  as  we 
have  already  seen,  ethical  and  aesthetic.  The  ethical  predic 
ates,  good  and  bad,  are  imputed  to  a  character  on  the  ground 
of  the  possession  or  non-possession  of  qualities  demanded  of 
him  as  a  person  in  personal  relations.  The  aesthetic  predicates, 
nobility,  vulgarity,  beauty,  ugliness  of  character,  represent 
complementary  values  arising  from  the  harmonious  or  in 
harmonious  relations  of  qualities  within  the  personality. 

An  analysis  of  the  ethical  predicates  good  and  bad,  as  used 
in  the  characterisation  of  persons,  and  in  estimation  of  their 
personal  worth,  shows  clearly  the  differentiation  of  object  and 
attitude,  the  relative  isolation  of  the  personality  already  de 
scribed.  In  the  first  place,  when  once  the  qualities  and  dis 
positions  which  are  instrumental  to,  and  the  condition  of, 
personal  participation  or  intercourse  have  been  differentiated 
and  fixed  by  the  selective  processes  of  Einfiihlung,  and  are 


286  Valuation:   its  Nature  and  Laws 

intrinsically  valued,  their  value  is  estimated  wholly  in  terms 
of  their  depth  and  breadth  in  the  personality,  the  degree  of 
their  identification  with  the  person.  The  various  virtues  are  all 
estimated  in  these  terms.  They  extend  all  the  way  from  those 
quasi-ethical  qualities  which  are  significant  merely  in  the  char 
acterisation  and  appreciation  of  the  individual,  the  lovable  and 
admirable  qualities,  to  the  more  fundamental  virtues,  such  as 
courage,  integrity,  persistence,  etc.,  having  wider  social  instru 
mental  value.  In  so  far  as  the  attitude  is  personal,  character 
isation  of  the  person  as  good  means  simply  supremacy  of  per 
sonal  over  condition  worths,  and  as  bad  the  supremacy  of  con 
dition  over  personal  values.  The  distinction  between  the  two 
levels,  and  between  forward  and  backward  value  movements, 
having  once  been  made,  the  choice  of  ease  and  comfort,  of  bodily 
good  of  any  sort,  instead  of  such  qualities  as  have  a  more  personal 
reference,  is  always  the  object  of  negative  judgment.  With 
the  exception  of  certain  limiting  cases  to  be  considered  later, 
this  law  is  practically  universal. 

The  terms  of  estimation  of  personal  worth  are,  therefore, 
intra-personal.     They  reflect  the  relative  isolation  of  the  per 
sonality  from  social  judgments,  and  abstraction  from  the  instru 
mental  social  value  of  the  quality.     This  comes  out  more  clearly 
when  we  contrast  the  terms  in  which  the  personal  worth  of  a 
quality  is  measured  with  those  employed  in  the  measurement  of 
its  social  impersonal  value.     In  order  to  measure  the  degree  of 
personal  worth,  the  extent  to  which  the  valued  quality  is  identi 
fied  with  the  personality,  its  depth  and  breadth  in  the  person, 
it  is  necessary  to  have  some  means  of  comparison,  and  this  is 
found  in  the  contrast  between  personal  and  condition  worths. 
All  estimation  of  relative  value  involves  two  factors,  a  positive 
and  a  negative  ;    the  degree  of  value  of  an  object  is  measured 
indirectly  by  the  extent  to  which  other  objects  are  sacrificed 
for  it.     In  the  case  of  imputation  of  personal  worth,  the  worth 
of  a  person  is  determined  by  his  readiness  to  sacrifice  condition 
to   personal   worth.     The   terms   of   estimation   are,   therefore, 
wholly  intra-personal,  within  the  ego,  and  reflect  the  division 
of  the  self  involved  in  the  ideal  construction  of  the  person. 

This  is,  however,  in  striking  contrast  to  the  terms  of  estima 
tion  employed  when  the  disposition  or  quality  is  measured 
from  the  more  specifically  moral  point  of  view.  Here  the 
moral  value  of  a  disposition  is  measured  in  terms  of  sacrifice 
of  the  individual  for  the  over -individual,  of  egoism  for  altruism. 


Personal  Worths  287 

In  this  attitude  of  judgment,  both  condition  and  personal  worths 
are  lumped  together  as  individual,  and  set  in  opposition  to  over- 
individual  social  values.  From  this  point  of  view,  estimation  of 
the  value  of  a  disposition  abstracts  entirely  from  the  intrinsic 
value  of  the  disposition  as  a  quality  of  a  person,  and  considers 
it  wholly  in  the  light  of  the  disposition  to  participate  in  over- 
individual  ends  displayed  by  the  quality  in  question,  the  de 
gree  of  which  disposition  is  measured  by  willingness  to  sacrifice 
both  condition  and  personal  worths.  It  is  true  that  in  the  main 
personal  worths  may  on  reflection  be  seen,  in  the  light  both  of 
their  genesis  and  meaning,  to  retain  a  potential  reference  to 
social  ends.  It  is  also  true  that  in  general  social  values  have 
complementary  personal  values  which  can  be  reflectively  worked 
out.  But  this  is  not  absolutely  true,  as  will  be  shown  in  the 
sequel.  Personal  and  social  values  are  at  some  points  indifferent 
in  the  sense  that  the  concepts  individual-ethical  and  socially 
valuable  are  only  partially  and  occasionally  identical.  There 
are  certain  qualities  and  actions  which  come  under  the  concept 
of  the  individual-ethical  which  from  the  standpoint  of  social 
morality  are  indifferent,  as  well  as  certain  social  demands  which 
may  be  in  abeyance  in  personal  relations.  The  full  extent  and 
bearing  of  this  fact  will  be  discussed  in  another  connection.1 
The  point  of  importance  here  is  that  these  remote  relations, 
if  they  exist  at  all,  are  irrelevant  for  the  specific  judgment,  in 
the  specific  situation. 

An  illustration  will  bring  the  situation  into  clearer  relief, 
and  emphasise  still  more  strongly  the  relative  isolation  of  the 
personality  involved  in  imputation  of  personal  worth.  Any 
individual  of  more  than  ordinary  character,  displaying  to  an 
unusual  degree  the  most  fundamental  personal  worths  of  strength, 
daring,  and  persistence  in  the  pursuit  of  his  ends,  forms  his  own 
standard,  and  to  a  certain  extent  finds  it  accepted  by  others. 
We  may  estimate  the  acts  of  some  unscrupulous  master  of  men, 
or,  in  fact,  of  an  ordinary  robber,  low  indeed  when  we  view  them 
in  the  light  of  our  normal  expectation  of  what  his  attitude  should 
be  toward  social  ends.  But  when,  suppressing  these  judgments 
temporarily,  we  consider  merely  his  courage  and  perseverance, 
his  readiness  to  sacrifice  condition  worths  for  what  are  to  him 
personal  worths,  our  estimate  is  largely  modified.  For  the  time 
being,  at  least,  these  qualities  dominate  us,  and  we  seem  to  get  a 
glimpse  into  ultimate  realities  of  will  deeper  than  the  superficial 

1  Chapter  xiv. 


288  Valuation  :   its  Nature  and  Laws 

distinctions  of  egoism  and  altruism.  This  employment  of  a 
double  standard,  as  involuntary  as  it  is  disconcerting  at  times, 
is  indicative  of  a  real  duality  in  our  conception  of  the  good 
which  must  be  worked  out  completely  before  it  can  be  overcome. 
In  this  connection  a  final  question  appears.  There  are 
two  ways  in  which  the  supremacy  of  personal  worths,  their 
identification  with  the  personality,  may  be  realised.  The 
same  relative  supremacy  might  be  attained,  either  by  actual 
increase  of  the  disposition  which  has  personal  worth,  or,  in 
directly,  by  the  decrease  or  weakening  of  the  disposition  which 
has  merely  condition  worth.  Is  the  imputed  value  the  same 
in  both  cases  ?  Experience  has,  I  think,  an  entirely  unequivo 
cal  answer  to  this  question.  In  so  far  as  our  judgment  is  one 
of  purely  personal  worth,  increase  of  personal  worth  may  be 
acquired  only  in  the  first  way.  The  mere  weakening  of  the  con 
dition  worths  decreases  the  personal  worth  in  that,  however 
idealised  and  spiritualised  these  personal  qualities  may  become, 
they  have  their  roots  in  those  more  elemental  qualities  of 
strength  and  spontaneity  of  instinct  and  will  to  which,  as  we 
have  seen,  a  quasi-ethical,  sub-personal,  and  sub-social  obligation 
attaches.  The  weakening  of  these  impulsions  involves  the 
weakening  of  the  personal  values  with  which  they  are  set  in 
contrast.  If,  on  the  other  hand,  our  judgment  is  purely  social 
and  impersonal,  the  way  in  which  the  desired  relation  of  personal 
to  condition  worths  is  attained  is  irrelevant,  for  we  are  concerned 
only  with  its  extrinsic  effects,  i.e.,  with  the  instrumental  value  of 
the  disposition  in  question  for  social  ends.  The  point  to  be  em 
phasised  is  that  our  specific  judgments  are  always  relative  and 
partial.  It  is  immaterial  for  practical  purposes,  for  our  "  snap  " 
social  judgments,  whether  correct  personal  habits  and  qualities 
arise  from  mere  weakness  of  passions  or  from  strength  of  will.  The 
effect  is,  in  either  case,  the  same.  But  for  our  judgment  upon 
the  person  it  makes  all  the  difference  in  the  world. 

4.  The  Difference  in  Relative  Value  of  the  Same  Objects  (Dispos 
itions  or  Qualities]  according  as  the  Value  is  Personal  or 
Over-personal. 

The  difference  in  objects  and  in  terms  of  estimation  of  the 
two  spheres  of  personal  and  impersonal  judgment  has  shown 
clearly  the  tendency  to  isolation  of  the  person  in  judgments  of 
personal  worth.  The  difference  becomes  still  more  marked 


Personal  Worths  289 

when  the  quantitative  aspect  of  these  judgments  is  considered. 
There  are,  as  has  been  shown,  some  qualities  or  dispositions  which 
have  both  a  personal  and  impersonal  reference,  and  toward  which 
our  attitude  is  mixed.  Even  here,  however,  it  is  not  difficult  to 
separate  the  personal  from  the  impersonal  attitude,  for  the  reason 
that  the  quantity  of  the  disposition  displayed  brings  out 
strikingly  different  judgments,  according  as  the  attitude  of  judg 
ment  is  personal  or  impersonal. 

This  is,  first  of  all,  evident  in  what  may  be  described 
as  the  difference  in  sensitiveness  of  the  personal  and  impersonal 
or  social  thresholds,  the  difference  in  the  standard  of  personal 
and  moral  imputation.  The  same  absolute  amount  of  dis 
position  displayed  by  an  act  may  have  a  very  different  relative 
significance  in  the  two  cases.  Such  a  disposition  as  truthfulness, 
which  has  both  a  personal  and  a  social  reference,  affords  a  good 
illustration.  A  painful  scruple  or  a  slight  divergence  from 
strict  truthfulness  may  have  a  value  for  personal  participation, 
and,  therefore,  for  personal  imputation,  while  it  is  negligible 
for  impersonal  social  participation.  On  the  other  hand,  it  is 
equally  true  that  in  the  case  of  some  dispositions  the  threshold 
of  moral  worth  judgment  may  be  passed  without  any  variation 
in  the  personal  worth  judgment.  This  is  most  apparent  in 
those  cases  where  we  recognise  that  the  moral  norm  has  some 
thing  of  the  conventional  in  it,  where  its  reference  to  important 
ends  is,  though  real,  somewhat  remote  and  indirect.  My  friend 
may  have  slightly  transgressed,  and  as  representing  the  social 
attitude,  I  call  him  to  account ;  but  the  very  expression  of  my 
face  and  the  tone  of  my  voice  shows  that  my  personal  atti 
tude  has  not  altered.  This  is  also  true  in  more  important 
situations.  Candour  compels  one  to  recognise  that  even  varia 
tions  from  more  fundamental  norms,  as,  for  instance,  those 
which  regulate  the  relations  of  the  sexes,  may,  when  they  are 
the  expression  of  a  frank,  generous,  and  spontaneous  passion, 
leave  the  personal  attitude  essentially  unchanged. 

In  the  second  place  there  are  numerous  nuances  of  per 
sonal  appreciation  which,  while  significant  for  personal  par 
ticipation  and  for  personal  judgments,  do  not  call  out  strictly 
moral  approval  or  disapproval.  This  fact  we  have  already 
observed  in  its  qualitative  aspect.  There  are  extensive  regions 
of  personal  qualities  which  are  only  remotely  significant  from 
the  abstract  moral  point  of  view ;  those  qualities,  for  instance, 
which  were  described  as  "  lovable."  But  this  fact  becomes 


290  Valuation  :   its  Nature  and  Laws 

still  more  apparent  when  we  view  the  situation  from  its  quanti 
tative  side.  In  the  case  of  those  qualities  which  call  out  both 
the  personal  and  impersonal  reaction,  there  are  wide  variations 
in  quantity  of  disposition  which  jail  to  influence  the  moral  judg 
ment.  In  imputing  worth  on  the  basis  of  a  disposition  dis 
played,  if  the  imputation  is  personal,  the  total  disposition  out 
of  which  the  action  springs  is  taken  into  account,  as  an  expression 
of  the  personality,  and  the  degree  of  value  imputed  tends  to 
vary  directly  with  the  amount  of  disposition  displayed.  In 
specifically  moral  judgment,  on  the  other  hand,  we  judge  rather 
in  the  light  of  the  minimum  of  disposition  without  which  the 
act  could  not  take  place,  i.e.,  its  instrumental  value  for  society. 
All  else,  excess  of  disposition  beyond  the  constant  of  social 
expectation,  whatever  of  uniquely  individual  emotion  and 
sentiment  is  displayed  by  the  act,  tends  to  be  irrelevant, 
and,  if  included  in  the  judgment  at  all,  receives  an  imputed 
value  slight  in  proportion  to  the  amount  of  disposition  displayed. 
A  consideration  of  the  preceding  phenomena  confirms  the 
view  that  the  judgment  of  personal  worth  is  a  relatively  distinct 
type  of  judgment,  involving  distinct  attitudes  and  presuppo 
sitions,  and  one  which  may  be  isolated  from  other  types  for  more 
detailed  study.1  This  well-defined  difference  in  the  significance 
of  qualities  or  dispositions,  in  respect  both  to  quality  and  quan 
tity,  according  as  the  point  of  view  from  which  they  are  judged 
is  the  personal  or  impersonal,  indicates  that  what  we  have 
described  as  imputation  of  personal  worth  follows  its  own  laws, 
as  determined  by  its  own  distinctive  presuppositions.  If  we  can 
define  these  presuppositions  we  shall  have  the  basis  for  the  ex 
planation  of  this  type  of  judgment  and  its  laws. 

1  Meinong,  Psychologisch-ethische   Untersuchungen  zur  Wtrththeorie,   p.    295,  has, 
to  an  extent,  developed  the  same  distinction  without,  however,  recognising  the  full 
significance  of  the  personal  attitude  in  worth  judgment.     He  describes  the  difference 
as  one  between  emotional  and  intellectual  imputation.     In  intellectual  imputation  the 
subject   of  the   worth   judgment  is  not   the   person    in   immediate  personal   relations 
with  the  object   of  the  judgment,  but  the   impersonal  subject  representing  the  total 
social  consciousness.     In  emotional  imputation  the  subject  is  an  individual  in  immediate 
sympathetic  relations  with  another  person.     In  like  manner  in  intellectual  imputation 
the  act,  upon  the  basis  of  which  value  is  imputed,  is  judged  wholly  in  the  light  of  its 
instrumental  value  for  social  ends.     The  emotional  aspect  of  the  act  and  the  place  11 
the  total  personality  of  the  disposition  which  it  presupposes,  are  irrelevant,  while  11 
emotional  imputation  these  determine  judgment,  and  the  instrumental  values  tend  t 
become  irrelevant. 


Personal  Worths  291 


II.  THE  PRESUPPOSITIONS  OF  RELATIVE  ESTIMATION  OR 
MEASUREMENT  OF  PERSONAL  WORTH 

i.  The  Thresholds  and  Norms  of  Personal  Obligation  and 
Imputation:  their  Origin  in  the  Processes  of  Sympathetic 
Participation. 

The  presuppositions  which  underlie  judgments  of  personal 
worth  are  clearly  different  from  those  which  determine  the 
so-called  moral  and  impersonal  judgments.  The  preceding 
attempt  to  isolate  the  phenomena  of  personal  imputation  justifies 
this  conclusion.  The  question  now  arises  whether  it  is  possible 
to  define  these  presuppositions  more  explicitly,  and  thus  to 
derive  the  concrete  phenomena  of  personal  imputation  from 
them. 

The  presupposition  of  feelings  and  judgments  of  personal 
worth  is,  expressed  in  the  most  general  terms,  what  we  have 
described  as  the  ideal  personality.1  More  closely  examined, 
this  ideal  was  seen  to  consist  in  certain  implicit  assumptions 
as  to  the  existence  or  non-existence  of  qualities  or  dispositions 
in  the  person,  certain  expectations  generated  in  sympathetic 
participation  and  ideal  construction.  When  these  expectations 
are  met,  when  the  object  is  judged  to  exist,  the  feeling  is  one  of 
satisfaction  and  the  judgment  is  positive  ;  when  the  object  does 
not  exist,  the  feeling  is  one  of  dissatisfaction  and  the  judgment 
negative. 

The  demand  for  the  existence  of  certain  qualities  or  dis 
positions,  their  possession  by  the  self  or  the  alter,  is  the  pre 
supposition  of  the  feelings  of  personal  worth  described.  The 
qualities  in  question  are  the  necessary  condition  of  personal 
participation,  and  of  those  feelings  of  respect  and  admiration 
for  the  person  as  such.  But,  as  we  have  also  seen,  these 
assumptions  or  expectations  have  a  quantitative  aspect,  as 
determined  by  the  laws  of  valuation  applied  to  feelings  of  par 
ticipation.  The  degree  of  feeling,  and  therefore  the  degree 
of  worth  imputed  to  the  person,  is,  consequently,  a  func 
tion  of  the  relation  of  the  quantity  of  disposition  displayed, 
of  the  supply,  to  the  demand  presupposed  As  in  the  sphere 
of  condition  worths,  the  intrinsic  and  instrumental  values  of 
economics,  the  degree  of  value  of  the  object  is  a  function  of  the 

1  Chap,  ix,  p.  262. 


292  Valuation  :   its  Nature  and  Laws 

two  factors,  demand  and  supply  ;  so  also  here  the  value  imputed 
to  the  person  is  a  function  of  the  relation  of  existential  judg 
ments  as  to  the  amount  of  disposition  displayed  by  the  person 
to  the  implicit  assumption  or  expectation  presupposed.1 

In  our  more  general  study  of  the  Laws  of  Valuation  -  it 
was  pointed  out  that  aU  judgments  of  value,  on  whatever  level 
presuppose  certain  norms  and  limits  which  define  in  conceptual 
terms  the  capacity  of  the  demand  or  feeling-disposition  pre 
supposed.  These  we  defined  as  the  Thresholds  of  Value,  and 
while  they  were  described  in  detail  only  in  the  case  of  "  condition  " 
worths,  it  was  shown  that  the  same  conception  might  be  applied 
to  the  higher  levels  of  valuation.  As  in  the  case  of  economic 
worth  judgment,  in  order  to  measure  the  relative  value  of  different 
quantities  of  a  good,  it  is  necessary  to  define  the  demand  quantita 
tively,  that  is,  to  determine  the  conceptual  points  within  which 
the  demand  moves,  its  norms  and  its  limits,  and  the  law  of 
increase  and  decrease  of  the  demand  with  change  in  quantity  of 
the  good,  so  also  here,  in  the  sphere  of  personal  worth,  the  same 
requirements  must  be  met. 

We  have  already  applied  these  concepts  in  our  preliminary 
demarcation  of  the  region  of  quasi-ethical  and  ethical  personal 
worths  from  the  impersonal  moral  worths.  These  points  of 
difference,  expressed  in  the  most  general  terms,  consisted  in 
the  two  facts  :  (i)  that  an  act  expressive  of  a  certain  quantity 
of  disposition  might  call  out  a  personal  worth  feeling  in  emotional 
imputation  when  it  would  not  rise  above  the  threshold  of  im 
personal  moral  judgment  ;  and  (2)  that  increase  of  quantity 
of  disposition,  displayed  in  such  act  or  acts,  continues  to  call 
out  personal  reaction  long  after,  from  the  impersonal  moral 
point  of  view,  the  increment  has  become  worthless  or  has 
passed  over  into  negative  worth.  The  limits  of  personal 


1  ,D°U^  may  naturally  ar[se  as  to  the  applicability  of  the  concepts  of  demand  and 
supply  in  this  connection  in  that,  while  the  judgment  of  value  seems  to  be  determined 

:  demand  or  expectation  presupposed,  the  ethical  demand  seems  to  be  unaffected 

by  the  supply.     This  very  general  and  uncriticised  assumption  is  just  the  point  at  issue. 

Ihe  ethical  demand,  as  an  abstract  norm  and  as  theoretically  formulated,  does  have  the 

appearance  of  such  independence.     The  ideal  of  personality  as  expressed  in  the  demand 

be  a  person  and  respect  others  as  persons"  can  be  said  to  be  thus  unconditional  for 

e  very  good  reason  that  it   is  practically  meaningless.     It   is  only  when   the  ideal 

becomes  specific,  when  the  demand  is  for  a  specific  quality  or  disposition  of  the  person 

ality,   that  it   becomes   the  basis  of  actual   concrete  judgments    of  value.     But    such 

demands  _  in  the  individual  are,   as  we  have  already  seen   in  our  study  of  the  ideal 

construction    of  the   personality,    determined,   both    as    to    what   specific   qual.ties  are 

smanded  and  as    to  the  amounts  expected,  by  the  empirical  conditions  of  personal 

participation. 

2  Chap,  vi,  p.  146. 


Personal  Worths  293 

worth  are,  therefore,  much  wider  than  in  the  case  of  moral  or 
impersonal  worth  judgment;  the  lower  threshold  is  lower,  the 
upper  threshold  is  higher.  These  facts  indicate  that  the  region 
of  ethical  and  quasi-ethical  personal  worths,  which  presuppose 
relative  isolation  of  the  personality,  is  a  wider  region  than  the 
moral.  We  must  now  seek  to  determine  the  thresholds  of  personal 
worth  feeling  more  definitely,  since  they  constitute  the  critical 
points  in  all  imputation  of  personal  worth. 

(a)  The  Normal  Threshold — The  Norm  of  Characterisation. 

All  imputation  of  personal  worth,  whether  positive  or  nega 
tive,  whether  of  merit  or  demerit,  takes  its  start  from,  pre 
supposes,  what  may  be  described  as  the  normal  threshold.  The 
normal  threshold  represents  that  amount  of  disposition  which 
corresponds  to  the  normal  expectation,  the  habit  or  implicit 
assumption,  generated  in  personal  intercourse,  in  sympathetic 
participation  and  its  accompanying  ideal  constructions.  As  a 
result  of  repeated  processes  of  "  reading  back  and  forth  "  of 
feeling  and  feeling-dispositions,  the  subject  comes  to  expect, 
of  himself  and  the  "  other  "  alike,  acts  expressive  of  a  certain 
constancy  of  disposition.  Expressed  in  the  more  definite  terms 
of  estimation  of  personal  worth,  he  expects  a  certain  degree  of 
supremacy  of  personal  over  condition  worths,  he  expects  certain 
personal  worths  to  have  acquired  a  certain  depth  and  breadth  in 
the  personality.  Since  this  expectation  constitutes  the  normal 
presupposition  of  the  characterisation  of  persons  which  under 
lies  personal  intercourse,  we  may  describe  it  as  the  characterisation 
norm.1 

It  is  clear  that  this  constant,  this  norm  of  characterisation, 
corresponds  to  habit,  and  that,  when  the  amount  of  dis 
position  displayed  by  a  person  merely  fulfils  the  demands  of 
the  "  correct,"  meets  the  implicit  assumption  with  which  we 
approach  him,  it  does  not  call  out  any  explicit  judgment,  positive 
or  negative.  To  the  person  who  displays  a  disposition  corre 
sponding  to  our  normal  expectation  we  impute  neither  merit  nor 

The  significance  of  this  concept  becomes  clearer  if  we  contrast  it  with  another 
conception  of  which  we  shall  make  use  later,  namely,  the  participation-norm.  When 
we  come  to  the  study  of  the  impersonal,  specifically  moral,  judgment  upon  acts 
(as  distinguished  from  the  ethical  and  quasi-ethical  personal),  we  shall  find  that  there 
also  a  certain  normal  expectation  can  be  clearly  distinguished  and  defined.  But  it  is 
different  from  the  normal  expectation  in  judgments  of  personal  worth,  and  is  determined 
by  different  laws.  It  represents  that  amount  of  a  socially  desirable  disposition  which  is 
normally  expected  of  an  individual,  and  is  determined  by  its  instrumental  value  for 
social  ends. 


294  Valuation  :   its  Nature  and  Laws 

demerit.  It  is  only  when  the  disposition  in  question  varies 
appreciably  in  the  way  either  of  excess  or  defect,  that  any 
explicit  reaction  and  judgment  takes  place.  But  from  this 
point  on  there  is  imputation  of  degrees  of  merit  or  demerit  pro 
portional  to  the  amount  of  variation  of  the  disposition  above 
or  below  the  normal.  We  must  now  seek  to  determine  the 
limits  within  which  this  variation  moves. 


(b)  The  Upper  and  Lower  Limits  of  Personal  Worth — The 
Characterisation-Minimum. 

In  the  case  of  any  given  quality  or  disposition  which  has 
become  inseparable  from  the  ideal  of  the  -person,  indispens 
able  for  personal  relations  and  the  values  of  characterisation, 
there  grows  up,  as  we  have  seen,  a  normal  expectation  or  norm 
of  characterisation.  From  this  norm  there  may  be  consider 
able  variation.  With  excess  of  disposition  there  is  imputation 
of  positive  worth  or  merit ;  with  defect,  imputation  of  negative 
worth  or  demerit.  But  the  character  of  this  series  of  judgments 
is  determined  by  certain  limiting  presuppositions  or  assumptions, 
beyond  which  the  attitude  and  type  of  judgment  undergoes  a 
change.  These  limits  or  thresholds  depend  upon  what  we 
may  describe  as  the  characterisation-minimum. 

By  this  minimum  of  characterisation  is  meant  the  smallest 
amount  of  a  disposition  or  quality  necessary  for  personal  sym 
pathy  and  personal  relations,  and  for  the  feelings  of  respect  and 
admiration  which  are  characteristic  of  personal  attitudes.  This 
term,  constructed  on  the  analogy  of  the  existence-minimum  in  the 
sphere  of  condition  worths,  denotes  a  conceptual  point  marking 
the  division  between  condition  worths  and  acquired  personal 
worth.  The  minimum  of  existence  marks  the  limit  of  relative 
condition  worth  where  it  passes  over  into  absolute  worth,  and 
is  the  term  used  to  describe  the  smallest  quantity  of  an  object 
necessary  for  existence,  and  for  which  no  substitute  can  be  found. 
The  minimum  of  characterisation  is  the  smallest  quantity  of  the 
disposition  of  a  person  necessary  for  personal  participation, 
and  therefore  for  valuation  as  a  person,  and  without  capacity 
of  substitution.  When  the  amount  of  the  disposition  sinks  below 
this  limit,  the  person  becomes  worthless  from  the  personal  point 
of  view,  and  this  minimum  constitutes  the  lower  threshold  of 
personal  imputation.  But,  on  the  other  hand,  since  it  represents 
the  minimum  of  an  indispensable  personal  good,  it  may  acquire 


Personal  Worths  295 

absolute  worth,  and  complete  sacrifice  of  condition  worths  for 
this  indispensable  minimum  may  call  out  imputation  of  absolute 
value.  This  point  represents  the  upper  limit  of  relative  estima 
tion  of  personal  worth. 

The  minimum  of  characterisation  appears,  therefore,  in  two 
concrete  situations.      In  the  first  place,  as  the  amount  of  the 
disposition  falls  below  the  normal  expectation,  the  imputation 
of   demerit  becomes   more  and  more   emphatic,  until  a  point 
is  reached  where  the  worth  judgment  experiences  a  qualitative 
transformation.      The    personal   attitude   passes  over  into  im 
personal  judgment  or  into  mere  altruism,  which,  as  we  have 
seen,  is  a  feeling  arising  upon  the  recognition  of  the  condition 
worths   or  ww-worths,   the    pleasures   or  pains,   of   the   person 
judged.      The   minimum    necessary   for  personal    participation 
has  been  reached.      Such  a  situation  we  sometimes  describe  by 
such  phrases,  as  an  act  or  person  is  "beneath  criticism"   or 
"  perfectly  worthless."     In  the  second  place,  such  a  minimum 
appears  as  a  functional  moment  in  another  concrete  situation. 
When  an  individual,  for  the  sake  of  a  minimum  of  personal 
worth   (say  honour),   sacrifices   all  his  condition  worths,   even 
life   itself,  our    judgment   passes    over   from    the    negative   to 
the   positive    side,   and    indeed    from    relative    into    absolute 
valuation  ;    the  moment  of  the  heroic  or  tragical  elevation  is 
reached.     This  situation  is  entirely  in  harmony  with  what  we 
have  found  to  be  the  character  of  personal  worths.     In  thus  com 
pletely  identifying  himself  with  the  minimum  of  personal  worth 
necessary  for  his  own  and  others'  ideal  construction  of  his  self, 
through  sacrifice  of  condition  worths,  the  subject  has  displayed 
another  personal  worth,  courage  or  devotion,  on  the  basis  of 
the  recognition  of  which  positive  worth  is  imputed.     There  has 
been  what  may  be  described  as  a  certain  substitution  of  personal 
worths. 

2.  The  Substitution  of  one  Personal  Worth  for  Another — Its 

Limits. 

This  leads  us  to  a  consideration  of  the  entire  question  of 
substitution  of  personal  worths.  The  minimum  of  characterisa 
tion  exists  only  in  the  case  of  qualities  or  dispositions  for  which 
there  are  no  substitutes.  We  have  also  seen  that  such  a  mini 
mum  of  characterisation  is  presupposed,  or  is  present  as  an 
implicit  assumption,  in  all  imputation  of  personal  worth.  Are  all 


296  Valuation  :   its  Nature  and  Laws 

personal   worths   without   this   capacity   of   substitution  ?     To 
answer   this   question   properly,  we   must   make   a   distinction 
between  dispositions  which  are  without  substitutes  in  particular 
situations  of  personal  participation  and  dispositions  which  are 
without    capacity    of     substitution    absolutely.      In    economic 
thinking,  the  crust  of  bread  is  the  minimum  of  existence  only 
under  certain  conditions  of  time  and  place  which  make  substitu 
tion  impossible.  So  also  in  personal  or  ethical  imputation— a  given 
minimum  of  a  disposition  may  be  without  substitute  in  certain 
relations  of  sympathetic  participation,  and  therefore  in  the  valua 
tion  of  the  self  or  the  alter.     Thus  if  we  take  the  more  superficial 
personal  worths,  beauty,  fame,  intellectual  power,  tact,  we  find 
hat,  while  the  more  intimate  personal  relations  are  impossible 
without   them— and  those  who   do  not  exhibit   the  necessary 
minimum    are    judged    impersonally    rather    than    personally, 
nevertheless,  they  are  really  goods  with  the  capacity  of  sub 
stitution,  for  a   display  of   the   deeper  personal  worths,  such 
as  honour,  devotion,  courage,  strength,  etc.,  may  restore  the 
relation  of  personal  imputation. 

Even  more  fundamental  values  than  these,  such  as  chastity 
and   honour,    usually   without   capacity   of   substitution,   may, 
under   exceptional   circumstances,   be   sacrificed   for  still  more 
Jtimate  personal  worths,  still  more  primitive  qualities.     Thus 
we  find  that,  while  the  personal  worth  of  chastity  is,  in  women, 
Tmally    without    capacity    of    substitution,    that    of    honour 
normally  without  substitute  in  men,  at  the  same  time  there  are 
cases  where  the  loss  of  both  may  be  compensated  for  by  the 
iisplay  of  still  more  ultimate  personal  worths.     To  the  woman 
who  sacrifices  her  chastity  for  her  starving  child,  or  for  the  love 
of  a  man,  or  for  love  of  country,  if  the  love  be  great  enough, 
and  the  lost  chastity  a  real  sacrifice,  we  impute  a  personal  worth 
which  may  reach  the  absolute  moment.     The  same  is  true  of 
the  sacrifice  of  honour  for  love.     What  has  taken  place  here  is 
that  a  still  more  ultimate  personal  worth  has  been  substituted 
for  a  lower.     And  the  failure  thus  to  sacrifice,  as  in  the  Statue 
and  the  Bust,  calls  out  negative  worth  judgments.     It  would 
appear,  then,  that  the  only  qualities  which  are  absolutely  without 
ibstitutes  are  the  most  fundamental  personal  worths,  strength 
and  harmony  of  character.     In  so  far  as  there  is  isolation  of  the 
personality,  our  sense  of  personal  worths  is  emphatic  on  these 
points. 

Nevertheless,  it  must  be  recognised  that  the  situations  here 


Personal  Worths  297 

described  are  limiting  cases.  The  isolation  of  personality,  with  the 
consequent  detachment  from  social  demand,  is,  for  most  of  us, 
perhaps  fortunately  for  society,  confined  to  the  purely  aesthetic 
moments.  In  most  of  our  judgments  upon  actual  personalities, 
these  personal  judgments  are  so  complicated  with  social  moral 
judgments  that  the  laws  of  personal  imputation  do  not  stand  out 
clearly.  The  isolation  of  personality  which  is  presupposed  in  all 
personal  judgments  is  only  imperfectly  realised.  But  the  fact 
that  it  can  be  realised  aesthetically,  and  that,  when  thus  realised, 
the  judgments  of  imputation  follow  the  laws  here  described,  is 
significant  for  the  whole  theory  of  value. 

3.  The  Role  of  the  Characterisation-Minimum  in  Personal 

Obligation. 

To  this  study  of  the  significance  of  the  minimum  of  char 
acterisation  in  imputation  of  personal  worth  to  the  alter,  should 
be  added  a  word  as  to  its  reflex  meaning  for  the  self,  in  his 
feelings  of  personal  worth  as  reflected  in  the  sense  of  personal 
obligation.  Here,  too,  the  minimum  of  characterisation  is  pre 
supposed  in  all  sense  of  relative  obligation — not  as  the  limit 
of  personal  participation,  but  of  ideal  characterisation  of  the 
self. 

Two  illustrations  will  be  sufficient  to  show  how  it  functions. 
We  are  told  most  enlightening  accounts  of  transformations  in 
the  sense  of  personal  obligation,  which  take  place  in  the  case 
of  those  who  by  circumstances  are  compelled  to  live  for  long 
periods  in  uncivilised  regions,  apart  from  personal  participa 
tion  in  the  life  of  those  among  whom  their  sense  of  obligation  has 
been  acquired.  Personal  worths  are  socially  derived,  and  fre 
quently  we  find  them  dropping  off  for  lack  of  continuation  of  the 
processes  of  sympathetic  participation  in  which  they  were  ac 
quired.  We  find  a  regressive  substitution  of  more  elemental  for 
more  developed  attitudes.  Among  others,  cleanliness  and  manners 
are  often  lost.  But  we  are  also  told  of  pathetic  cases  where,  for 
the  sake  of  cleanliness,  or  other  habits  which  distinguish  a 
gentleman,  most  important  condition  worths  will  be  sacrificed, 
and  in  extreme  cases  cleanliness  may  become  an  indispens 
able  minimum,  for  which  life  itself  is  risked,  if  not  actually 
sacrificed.  The  situation  obviously  is  this.  Through  contrast 
this  personal  worth,  normally,  perhaps  not  absolutely  without 
substitutes,  is,  nevertheless,  now  so  completely  identified  with 


298  Valuation:    its  Natitre  and  Laws 

the  personality,  that  it  becomes  a  symbol  of  all  other  personal 
worths,  and  for  it  even  life  is  sacrificed.  Here  the  most  emphatic 
obligation  is  found  at  the  minimum  of  characterisation.1 

There  seems  to  be  no  question  of  the  existence  of  some 
personal  worths— in  the  region  of  personal  obligation  as  well 
as  of  imputation,  for  which  there  are  no  substitutes.  They 
constitute  the  irreducible  minimum  of  certain  dispositions 
necessary  for  personal  participation  in  the  experience  of  the 
alter  and  of  personal  attitude  toward  him,  as  well  as  the  irre 
ducible  minimum  necessary  to  that  ideal  characterisation  of  the 
self  which  is  the  condition  of  the  continuation  of  valuation  on 
the  higher  level  of  personal  worths.  As  such  they  have  absolute 
value.  It  does  not  follow,  however,  that  these  same  dispositions, 
in  their  aspect  of  instrumental  value  for  social  over-individual 
ends,  are  equally  without  capacity  for  substitution.  The  over- 
individual,  instrumental  value  of  dispositions  is  determined  by 
other  laws. 


III.  ANALYSIS  OF  THE  LAWS  GOVERNING  FEELINGS  OF 
PERSONAL  WORTH  AS  ILLUSTRATED  IN  IMPUTATION  OF 
MERIT  AND  DEMERIT  AND  IN  PERSONAL  OBLIGATION 

i.  Imputation  of  Personal  Worth  to  the  Alter. 

With  these  norms  and  limits  as  conceptual  instruments  of 
analysis,  we  may  now  turn  to  a  quantitative  study  of  feelings 
of  personal  worth  as  expressed  in  the  judgments  of  imputation 
and  obligation.  The  worth  of  the  disposition  is  measured 
directly  by  the  changes  in  the  emphasis  of  the  judgments  of 
imputation,  indirectly  by  the  strength  of  the  feeling  of  obligation 
as  expressed  in  the  judgment  "  I  ought."  The  imputed  worth 
expresses  the  degree  to  which  the  expectations  generated  in 
sympathetic  projection  are  satisfied;  negative  worth  judgments 
reflect  the  degree  of  variance  from  the  expectation.  Similarly, 

1  In  Conrad's  novel,  Lord  Jim,  Loid  Jim,  the  chief  character  of  the  book- 
having,  by  processes  of  imaginative  construction,  identified  most  completely  his 
personal  worth  with  the  attitude  of  bravery,  and  failing,  through  weakness  of  the  flesh, 
to  perform  a  brave  act  which  he  demands  of  himself,  i.e.,  the  rescue  of  shipwrecked 
people — feels  his  sense  of  personal  worth  completely  lost,  and  devotes  his  entire  life  to 
its  recovery,  a  recovery  which  is  finally  realised  only  in  the  tragical  elevation  of  self- 
sacrifice.  One  of  the  interesting  features  of  this  study  is  that  this  sense  of  personal 
obligation  is  represented  as  persisting  independently  of  any  social  demand,  in  a  savage 
environment  where  his  past  is  wholly  unknown.  Here  we  have  the  significance  of  the 
characterisation-minimum  at  its  clearest,  an  illusion  perhaps  from  the  point  of  view  of 
social  obligation  and  imputation,  but  for  our  empirical  study  of  the  worth  consciousness 
a  very  significant  illusion. 


Personal  Worths  299 

the  degree  of  obligation  measures  the  extent  to  which  the  ideal 
object,  the  disposition,  has  been  identified  with  the  personality, 
and  therefore  has  worth.  In  tracing  out  the  modifications  in 
these  feelings,  and  their  corresponding  judgments— between  the 
limits  of  the  normal  threshold,  at  which  the  imputed  value  is 
null,  and  the  characterisation  minimum  where  the  imputed 
value  is  absolute,  we  shall  have  the  means  of  determining  the 
validity  of  our  analysis  of  the  origin  and  nature  of  personal 

worths. 

We  may  begin  this  discussion  by  taking  as  our  object 
of  study  feelings  or  dispositions  irrelevant  for  moral  judgment, 
that  is,  for  impersonal  participation,  but  decidedly  significant 
for  personal  participation;  thence  proceeding  to  the  consider 
ation  of  dispositions  significant  for  both  personal  and  impersonal 
participation,  but  which,  after  a  certain  minimum  of  the  disposi 
tion  is  reached,  become  irrelevant,  if  not  superfluous,  for  the 
demands  of  impersonal  participation. 

As  illustrative  of  the   first  we  may  take  some  nuance  of 
conjugal  or  filial  sentiment  too  fine  to  rise  above  the  threshold 
of   the   moral  sense,  one   which   is  essentially  the   product  of 
sympathetic   Einfiihlung  between  two  persons.     Such  are  the 
finer  and  more    spiritual    faiths   and    loyalties  which  may  be 
demanded   as   the  basis    or   unspoken  presupposition   of   inti 
mate  personal  relations,  but  which  can  scarcely  be  demanded 
in  the  more  impersonal  but  still  moral  relations  which  are  the 
concern  of   the  impartial  spectator.     The  somewhat  romantic 
demand  for  absolute    trust,  for  the   faith  which  believeth  all 
things  despite  appearances,  is  the  necessary  condition  of  the 
higher   forms    of    love    and  friendship,   of   conjugal   and  filial 
relations,   but  lies   beyond   the  sphere   of    the  strictly   moral. 
A  certain  minimum  of  the  crasser  conjugal  and  filial  attitudes 
our  moral  sense  does  indeed  demand,  but  with  others,  it  is  strictly 
speaking,  not  concerned.     The  individual  not  concerned  with 
them  is  aware  of  them  indeed  only  through  aesthetic  presentation. 
When  once  such  an  attitude  of  trust  or  loyalty  has  been  identified 
with  a  personality  and  a  certain  expectation  formed,  personal 
worth  feeling  may  run  through  a  whole  gamut  of  positive  and 
negative   changes  without   the   threshold  of   impersonal  moral 
judgment  having  once  been  crossed,  just  as  there  may  be  con 
siderable  modifications  in  our  moral  judgment,  as,  for  instance, 
judgments  of  disapproval  upon  disposition  and  act  which  are  not 
significant  for  the  law. 


300  Valuation  :   its  Nature  and  Laws 

Given  such  a  disposition  as  a  personal  worth,  the  judgments  of 
imputed  value,  between  the  two  critical  points  of  expectation,  the 
norm  and  the  minimum  of  characterisation,  arrange  themselves 
as  follows.  Decrease  in  disposition  below  the  normal  expectation 
is  followed  by  corresponding  increase  in  the  demerit  imputed, 
until  the  minimum  of  characterisation  is  reached,  a  point  which, 
of  course,  varies  with  individuals  and  circumstances.  But  when 
this  point  has  been  passed,  the  indispensable  minimum  of 
personal  participation  having  been  reached,  the  distinctly  personal 
relation  has  been  severed,  and  either  mere  condition  worths 
become  significant  or  an  impersonal  moral  attitude  appears. 

The  phenomena  at  this  point  are  very  interesting.  When 
the  minimum  of  characterisation  has  been  reached,  one  of  two 
fundamental  changes  in  attitude  on  the  part  of  the  subject 
of  judgment  takes  place.  Either  he  may  substitute  an 
attitude  of  pity  on  the  basis  of  his  appreciation  of  the  "  con 
dition  "  worths  or  mi- worths  of  the  alter,  or  he  may  substitute 
the  intellectual,  impersonal  attitude  of  moral  judgment  for 
the  personal,  and  emotional.  We  have  already,  in  our  discussion 
of  the  minimum  of  characterisation,  considered  the  former  possi 
bility.  It  is  a  value  movement  of  the  backward  type  in  which 
organic  sympathy  takes  the  place  of  the  intuitive  projection 
upon  which  personal  worths  are  founded,  or  in  which  the  con 
ceptual  construct  of  the  pleasure  or  pain  of  the  other  is  made 
the  end  of  altruistic  acts,  instrumental  to  increase  of  pleasure 
or  decrease  of  pain.  The  other  possibility  is  the  substitution 
of  the  impersonal  attitude,  in  which  the  isolation  of  the  person 
ality  ceases  and  the  impartial  moral  judgment  takes  its  place. 
In  connection  with  the  personal  worths  considered,  conjugal 
and  filial  attitudes,  both  types  of  substitution  are  in  evidence. 
Love  passes  over  into  pity  or  into  scrupulous  justice.  When 
the  minimum  of  personal  participation  has  been  reached,  valu 
ation  does  not  cease,  but  object  and  attitude  change.  It  is 
significant  that  the  minimum  may  be  passed  and  the  personal 
relation  severed  when,  from  the  standpoint  of  the  impartial 
spectator,  the  distinctively  moral  predicate  "bad"  cannot  be 
applied.  Similarly,  as  we  shall  see  later,  the  personal  worth 
relation  may  be  maintained  long  after,  from  the  moral  point  of 
view,  the  person  is  the  object  of  judgments  of  disapproval. 
There  is  a  sense  in  which  feelings  of  personal  worth  are  beyond 
the  moral  region  of  good  and  bad,  above  and  below. 

If  we  turn  now  to  those  dispositions  which  are  significant 


Personal  Worths  301 

for  both  personal  and  impersonal  imputation,  but  which  in 
certain  quantitative  aspects  are  no  longer  relevant  for  moral 
judgment,  we  find  the  same  laws,  with  slight  modifications, 
at  work.  The  sense  of  honour,  as  a  disposition  calling  out 
both  personal  and  moral  judgment,  is  a  good  illustration,  for 
it  is  the  class  name  for  a  group  of  dispositions  significant  from 
different  standpoints  of  judgment,  varying  all  the  way  from 
the  purely  personal,  through  social  groups  of  varying  extension, 
to  the  impersonal  impartial  standpoint.  It  needs  no  discus 
sion  to  point  out  that  the  expectation  in  these  different  cases 
varies  greatly.  As  in  the  similar  case  discussed  above,  an 
individual  may  lose  all  personal  honour  for  his  friend  or 
social  group  long  before  he  has  done  anything  which  will  call 
out  strictly  moral  judgment.  It  is  also  equally  true  that  an 
excess  of  sacrifice  for  the  ideal  of  honour  may,  from  the  personal 
standpoint,  call  out  an  absolute  personal  value,  while  from  the 
social  moral  point  of  view  it  is  irrelevant  or  indeed  quixotic, 
if  not  the  object  of  distinct  disapproval.  In  fact,  the  indispens 
able  minimum  which  is  demanded  of  an  individual  in  a  purely 
personal  relation,  or  even  in  a  limited  group,  may  exceed  the 
normal  demand  in  impartial  moral  judgment.  Thus  we  find  that 
the  sacrifice  for  honour  is  susceptible  of  indefinite  increase, 
and  that  with  this  increase  there  is  a  corresponding  increase 
of  personal  worth,  in  so  far  as  we  have  the  condition  of  isolation 
of  the  personality  fulfilled.  On  the  other  hand,  in  so  far  as  the 
social  moral  point  of  view,  with  its  intellectual  instrumental 
judgments,  intervenes,  excess  of  disposition  beyond  a  certain 
point  tends  to  call  out  negative  worth  judgments. 

It  will  be  worth  while  to  consider  one  more  disposition 
which  may  be  judged  both  as  a  personal  and  social  worth,  namely, 
altruism,  especially  since  it  is  at  this  point  that  the  most  inter 
esting  problems  of  the  relation  of  personal  to  social  values 
appear.  By  altruism  is  understood  the  disposition  to  sym 
pathetic  participation  in  the  worth  feeling  of  others.  This 
disposition  may  obviously  be  valued  both  intrinsically  as  a 
personal  worth  and  instrumental^  with  reference  to  its  value 
for  social  ends.  Now  for  certain  reasons,  which  will  be  de 
veloped  in  the  following  chapter,  the  instrumental  value  of  the 
altruistic  disposition,  the  disposition  to  sacrifice  condition  and 
personal  worths  for  over-individual  values,  is  not  susceptible 
of  indefinite  increase.  We  shall  find  that  a  certain  normal 
display  of  altruism  is  demanded,  but  that  increase  in  excess 


302  Valuation  :   its  Nature  and  Laws 

of  that  norm  is  governed  by  a  law  analogous  to  the  law  of 
Diminishing  Utility  in  the  sphere  of  economic  goods.  A  point 
is  finally  reached  where  sacrifice  is  no  longer  approved,  and  may 
indeed  call  out  negative  judgments  of  disapproval.  But  it  is 
significant  that  if  the  altruistic  individual  be  isolated  from 
social  moral  judgment,  he  may  acquire  personal  worth  indefinitely 
through  such  sacrifice  of  condition  and  personal  worths.  In  so 
far  as  the  disposition  is  valued  intrinsically,  as  an  attribute 
of  the  personality,  it  is  not  subject  to  this  law,  but  with  the  in 
crease  of  the  disposition  there  is  a  corresponding  increase  of 
imputed  worth.  The  limiting  case  is  the  tragical  elevation 
which  comes  with  the  sacrifice  of  all  condition  and  personal 
worths  for  over-individual  good,  the  sacrifice  of  life  itself.  Such 
sacrifice,  if  its  value  were  determined  by  impersonal  intellectual 
judgment,  would  not  be  found  to  have  the  absolute  value  which 
we  attribute  to  it  intrinsically  as  a  personal  worth.  And  when 
we  thus  value  it  intrinsically,  we  do  so,  it  should  be  observed, 
as  expressive  of  the  capacity  of  the  individual  to  identify  himself 
with  the  over-individual  object,  abstracting  from  the  instru 
mental  value  of  the  act.  As  a  personal  worth,  altruism  may 
acquire  absolute  value,  as  a  social  worth  it  is  not  susceptible 
of  indefinite  valuation. 

This  difference  in  the  standards  of  value  applied  to  the 
same  disposition,  according  as  it  is  referred  to  the  personality 
or  to  over-individual  objects,  will  come  out  more  clearly  in 
the  succeeding  chapters,  where  the  norms  of  over-individual 
values  are  developed,  and  the  laws  of  preference  between  personal 
and  over-individual  worths  are  studied.  Here  it  is  sufficient 
merely  to  note  that  in  this  intrinsic  valuation  the  question  of 
the  effect  of  increase  of  disposition  is  ignored  as  irrelevant. 
Secondary  instrumental  judgments  are  suppressed  in  the  im 
mediacy  of  the  personal  relation,  or  at  most  are  represented 
by  a  vague  assumption  of  indefinite  reapplicability,  or  instru 
mental  value,  of  the  disposition  in  question.  In  the  terminology 
of  ethical  discussion,  the  motive  alone  is  considered,  and  not  the 
effect.  Whether  this  attitude,  this  assumption,  does  not  in 
volve  an  illusion  is,  of  course,  a  question.  The  whole  problem 
of  worth  illusions  will  present  itself  for  consideration  at  the  con 
clusion  of  the  chapter. 


Personal  Worths  303 


2.  Personal  Obligation  as  Reflecting  the  Laws  of  Personal  Worth. 

The  individual's  feelings  of  the  worth  of  his  own  qualities, 
as  distinguished  from  his  sense  of  the  worth  of  the  qualities  of 
the  alter,  are  more  obscure  than  the  latter.  In  general,  however, 
it  may  be  said  that  his  sense  of  the  importance  of  his  personal 
qualities  is  reflected  in  his  feelings  of  personal  obligation.  The 
degree  of  the  feeling  of  obligation  represents  the  extent  to 
which  the  quality  has  been  identified  with  the  personality, 
and,  therefore,  the  degree  of  its  transgredient  reference.  The 
depth  of  the  feeling  of  satisfaction  and  inner  peace,  when  the 
obligation  is  met,  reflects  in  a  similar  manner  this  acquired 
personal  meaning. 

The  feeling  of  obligation,  as  analysed  in  a  preceding  chapter,1 
was  found  to  have  three  forms,  to  show  three  shades  of  meaning 
which  are  appreciatively  distinguishable.  Condition  worths 
may  acquire  an  instinctive  sub-personal  and  sub-social  obligation, 
through  arrest  of  conative  tendency,  as  illustrated  in  the  obli 
gations  which  appear  as  the  objects  of  condition  worth  approach 
the  existence  minimum. ~  Psychical  objects,  qualities  or  dis 
positions  of  the  personality  which  emerge  in  the  processes  of 
sympathetic  participation  and  ideal  construction,  arouse  feelings 
of  obligation  which  may  be  personal  or  social  and  over-individual, 
according  as  the  object  is  referred  to  the  self,  as  intrinsic  quality 
of  the  person,  or  to  some  over-individual  social  demand.  The 
feeling  of  obligation  toward  these  psychical  objects  may  be 
personal  or  over-individual,  according  as  the  demand  is  felt  as 
personal  or  social.  There  are  some  attitudes  or  dispositions  to 
which  merely  personal  obligation  attaches,  while  there  are  others 
which  have  both  personal  and  over-individual  or  social  obligation. 
It  is  with  the  feeling  of  personal  obligation  that  we  are  now  con 
cerned. 

These  purely  personal  obligations  are  not  difficult  to  dis 
tinguish  either  from  the  sub-personal,  instinctive,  or  from  the 
impersonal  moral.  If,  as  was  indicated  in  a  previous  chapter, 
it  may  be  said  that  strength,  sex,  etc.,  have  certain  obligations 
which  are  not  the  reflex  of  ideal  projections,  either  personal  or 
social,  it  is  also  true  that  there  are  some  qualities  of  the  person 
ality  which  have  obligations  which  are  irrelevant  for  social 
participation,  more  specifically  for  social  demand.  Their  ful- 

1  Chap.  VII,  pp.  209-11.  a  Chap,  vn,  pp.  213-16. 


304  Valuation  :    its  Nature  and  Laws 

filment  is  not  significant  for  social  demand  ;  they  do  not  rise 
above  the  threshold  of  impersonal  moral  judgment.  Nor  does 
failure  to  meet  these  obligations  call  forth  any  self-condemnation 
in  so  far  as  our  attitude  is  impersonal  and  moral.  The  dis 
tinction  sometimes  made  between  perfect  and  imperfect  obli 
gation  describes  the  situation.  Perfect  obligation  applies  to 
those  fundamental  dispositions  demanded  as  a  condition  of 
social  participation,  demanded  universally,  and  in  a  certain 
amount  which  constitutes  the  minimum  for  participation. 
Those  more  individual  and  personal  obligations,  on  the  other 
hand,  which  attach  to  dispositions  not  demanded  generally,  and 
to  dispositions  in  excess  of  the  minimum  demanded,  are  described 
as  imperfect. 

The  phrase  "  Noblesse  oblige "  includes  many  of  these 
personal,  and  from  the  impersonal  point  of  view  imperfect,  obli 
gations.  Thus,  to  take  the  obligations  of  nobility  in  its  original 
class  sense,  such  obligations,  as  the  result  of  intimate  sympathetic 
participation,  will  be  more  intimately  personal  than  the  more 
extensive  social  obligation.  Deeper  still  are  the  more  personal 
obligations  of  the  thinker  to  truth,  of  the  artist  to  beauty,  and 
of  the  saint  to  holiness.  They  frequently  transcend  completely 
any  cognisable  social  reference  ;  and  the  obligation  remains,  even 
when  the  attitude,  the  disposition  to  which  it  is  attached  is, 
either  by  reason  of  its  uniqueness  or  its  excess,  not  only  irrele 
vant,  but  even  inimical  to  wider  social  participation. 

The  chief  characteristic  of  personal  obligation  is,  then, 
that  its  objects  do  not  necessarily  correspond  to  the  objects 
of  impersonal  social  obligation.  Much  more  noticeable  is 
this  discrepancy  when  we  consider  personal  obligation  in  its 
quantitative  aspect.  The  intensest  personal  obligation  is  always 
about  the  minimum  of  characterisation,  but  this  minimum  may 
represent  an  amount  of  the  disposition  in  question  far  in  excess 
of  the  normal  or  correct  as  demanded  by  social  judgment. 
And,  whereas  the  impersonal  obligation  falls  off  in  intensity 
as  the  amount  of  disposition  demanded  by  a  given  act  exceeds 
the  normal,  the  personal  obligation  still  remains. 

This  divergence  arises  from  differences  in  the  processes 
of  sympathetic  participation  in  which  these  two  worth  attitudes 
have  their  origin,  and  thus  in  differences  in  the  demands  felt 
as  obligations.  In  the  case  of  all  those  attitudes  or  dispo 
sitions  which  have  both  over-individual  and  personal  obligation, 
there  is  a  certain  normal  disposition  which  we  may  describe 


Personal  Worths  305 

as  the  "correct,"  how  determined  we  need  not  in  the  present 
connection  consider,  about  which  the  intensest  over-individual 
obligation  is  found.  Any  variation  from  the  correct,  in  the 
way  of  lack,  is  felt  as  obligation  to  make  good  the  deficiency, 
and  this  feeling  of  obligation  finds  its  expression  in  the  obli 
gation  to  sacrifice  personal  and  condition  worths.  But  when 
the  sacrifice  reaches  a  certain  point  normally  demanded,  or 
when  the  over-individual  good  to  be  attained  by  the  sacrifice 
is  so  minute  as  not  to  be  significant  for  social  participation,  the 
over-individual  demand  or  impersonal  obligation  lapses.  Not 
so,  however,  if  the  attitude  or  disposition  is  identified  with  the 
personality,  as  a  personal  worth.  The  personal  demand  or  obliga 
tion  continues  beyond  this  point,  and  if  the  disposition  is  a 
personal  worth  for  which  there  is  no  substitute,  it  may  reach  the 
absolute  point  at  the  minimum  of  characterisation.  The  illustra 
tion  of  cleanliness,  already  considered  in  connection  with  the 
discussion  of  this  minimum,  is  a  case  in  point.  To  this,  as  an 
ideal  object  of  desire,  attaches  both  social  and  personal  worth ; 
it  is  the  object  of  both  over-individual  and  personal  demand. 
The  intensest  over-individual  obligation  is  found  at  the  minimum 
demanded  for  social  participation,  and  the  sacrifice  demanded 
will  be  greatest  at  this  point.  But  when  the  sacrifice  exceeds  the 
maximum  demanded,  over-individual  obligation  begins  to  fall 
off.  It  does  not  follow,  however,  that  personal  obligation  has 
ceased.  On  the  contrary,  when,  as  in  the  case  cited,1  cleanliness 
has  been  identified  with  the  personality  as  a  good  for  which 
there  is  no  substitute,  it  constitutes  an  indispensable  minimum, 
and  the  demand  for  sacrifice  may  be  absolute.  It  is  noteworthy 
that  quantitatively  this  minimum  for  characterisation  may  exceed 
that  which  constitutes  the  minimum  for  participation.  In  the  case 
before  us  the  degree  of  cleanliness  which  constituted  the  charac 
terisation-minimum  was  far  in  excess  of  the  social  demand. 
The  obligations  to  honour  and  altruism  are  of  the  same  general 
character.  Personal  obligation  to  an  ideal  of  honour  may 
continue  long  after  all  feeling  of  impersonal  over-individual 
obligation  has  lapsed;  and  in  this  particular  case,  the  per 
sonal  obligation  may  be  in  contradiction  to  the  social,  as  when 
the  obligation  to  avenge  one's  honour  conflicts  with  the  moral 
obligation  to  abstain  from  personal  vengeance. 

In  the  case  of  altruism  we  have  the  most  striking  illustration 
of  the  relation  under  discussion.     A  certain  degree  of  altruism 

1  See  above,  pp.  297  ff. 


306  Valuation  :   its  Natiire  and  Laws 

constitutes  the  maximum  of  social  demand,  the  maximum 
required  for  social  participation.  Beyond  that  point  impersonal 
obligation  falls  off,  but  not  necessarily  personal  obligation. 
The  intensest  personal  obligation  is  felt  at  the  characterisa 
tion-minimum,  and  the  amount  of  the  disposition  which 
constitutes  this  minimum  for  a  given  individual  may  be 
far  in  excess  of  the  maximum  of  social  demand.  The  charac 
terisation  minimum  constitutes  the  indispensable  minimum 
for  the  ideal  construction  of  the  self,  and  a  disposition  which 
ceases  to  have  value  for  over-individual  ends  may  acquire  in 
trinsic  complementary  value  through  reference  to  the  total 
personality. 

3.  The  Role  of  ^Esthetic  Complementary  Value  in  Judgments 

of  Personal  Worth. 

The  preceding  paragraphs  seem  to  indicate  that  objects 
of  personal  worth,  as  measured  by  the  feelings  of  merit  and 
obligation,  may  acquire  absolute  value.  Through  the  processes 
of  sympathetic  participation  and  isolation  of  the  personality, 
dispositions  may  be  so  identified  with  the  personality  as  to  be 
without  capacity  of  substitution,  and  to  demand  indefinite 
sacrifice  of  condition  worths.  With  the  increase  of  this  demand 
for  sacrifice,  there  is  a  corresponding  increase  of  the  individual's 
sense  of  worth,  increase  of  the  transgredient  reference  of  the 
feeling;  and  with  increase  of  the  disposition  to  sacrifice,  there 
is  a  corresponding  increase  of  the  merit  imputed  to  the  per 
sonality.  This  law  appears  to  hold,  however,  only  when  the 
individual  is  isolated,  and  when  judgments  as  to  the  partici 
pation  or  instrumental  value  of  the  dispositions  in  question 
for  social  good  are  suppressed.  The  limiting  case,  both  of 
personal  imputation  and  personal  obligation,  is  reached  in  the 
feeling  of  tragical  elevation  which  comes  with  complete  identi 
fication  of  the  attitude  with  the  personality  through  complete 
sacrifice  of  other  worths,  even  of  life  itself. 

But  an  analysis  of  actual  worth  judgments  seems  to  indi 
cate  that  personal  worths  are  thus  "  steigerungsfahig,"  sus 
ceptible  of  indefinite  valuation,  and  may  even  acquire  absolute 
worth,  when  the  identification  with  the  personality  is  not  meas 
ured  in  terms  of  sacrifice,  i.e.,  is  without  the  elements  of  opposition 
and  contrast.  If  the  personality  may  be  isolated  and  tragically 
elevated  by  the  extreme  of  sacrifice,  it  seems  equally  possible 
for  it  to  be  aesthetically  isolated  by  the  display  of  an  inner  unity 


Personal  Worths  307 

and  harmony  of  character  which  shows  no  traces  of  effort  and 
struggle.  For  it  is,  in  the  first  place,  a  fact  long  observed  that 
many  of  the  finer  attitudes  of  personal  worth  are  of  such  a 
character  that  they  lose  their  unique  quality  when  their  display 
involves  sacrifice  and  effort.  Love,  devotion,  gratitude,  honour, 
are  feelings  which  cannot  be  forced  without  losing  something 
of  worth.  We  care  neither  to  demand  them  of  others  nor  to 
have  others  demand  them  of  us.  The  moment  of  spontaneity 
and  harmonious  expression  which  constitutes  what  has  been 
called  the  "  beautiful  soul  "  seems  to  be  a  fundamental  factor 
in  determining  personal  worth. 

Now,  while  there  can  be  no  doubt  that  worth  is  imputed 
to  this  element  in  personality,  and  that  the  degree  of  worth  in 
creases  with  the  increase  of  the  disposition  above  the  normal, 
it  would  be  a  mistake  to  interpret  such  imputed  worth  as  merit. 
The  condition  of  the  imputation  of  merit  is  the  sacrifice  of 
condition  to  personal  worths,  the  contrast  and  opposition  which 
arises  from  the  ideal  construction  of  personal  worths.  In  the 
case  before  us,  however,  it  is  precisely  because  this  distinction 
has  lapsed,  and  the  objects  of  personal  worth  are  desired  with 
the  same  spontaneity  as  the  objects  of  condition  worth,  that 
the  new  quality  enters  into  our  feeling  of  the  worth  of  the  person 
ality.  It  has  been  held  that  these  values  presuppose  the 
consciousness  of  effort  and  sacrifice  in  the  same  way  that  the 
feeling  of  the  worth  of  a  work  of  art,  a  picture,  or  a  piece 
of  music,  may  include  in  it  a  realisation  of  the  effort  involved 
in  mastering  the  difficulties  of  execution.1  That  such  an  element 
does  enter  into  many  aesthetic  valuations  of  personality  is  beyond 
doubt,  but  it  has  then  ceased  to  be,  as  in  the  case  of  imputa 
tion  of  merit,  a  measure  of  value.  The  situation  is  rather  this. 
The  feeling  in  question  is  of  the  aesthetic  immanental  worth 
which  comes  from  the  assumption  that  the  disposition  desired 
is  fully  realised  in  the  personality.  This  is  the  positive  source 
of  the  worth  feeling.  But  such  repose  in  the  personality  pre 
supposes  its  isolation,  and  this  isolation  involves  the  ex 
trusion  of  all  negative  moments ;  otherwise  the  assumption 
of  such  complete  realisation  cannot  be  maintained.  The  cog 
nition  of  past  efforts  and  sacrifices  adds  nothing  to  the  intrinsic 
worth  of  the  object,  but  merely  strengthens  the  assumption  and 
isolates  the  object.  This,  I  am  inclined  to  think,  is  the  hidden 
emotional  logic  of  the  religious  consciousness  in  its  worship  of 

1  Simmel,  Einleitung  in  die  Aforalwissenschaft,  Vol.  I,  p.  227. 


308  Valuation :   its  Nature  and  Laws 

such  ideal  persons  as  Buddha  and  Jesus,  a  factor  in  the  ideal  con 
struction  of  which  there  is  always  the  story  of  supreme  tempta 
tions.  The  temptations  add  nothing  to  the  intrinsic  value,  but 
they  serve  to  isolate  the  person  and  to  strengthen  the  assump 
tion  upon  which  the  worth  feeling  rests. 

It  should  be  noted  that  the  isolation  of  the  personality, 
and  these  corresponding  aesthetic  valuations  of  persons  in  art 
and  in  life,  are  unique  and  fleeting  moments  in  experience,  for 
the  reason  that  the  assumptions  upon  which  they  rest  can  be 
sustained  for  but  short  periods  and  under  special  conditions. 
But,  as  has  already  been  pointed  out,  there  are  some  moments 
of  worth  experience  the  value  of  which  does  not  lie  in  repetition, 
but  which,  when  once  experienced,  create  expectations  which 
determine  succeeding  worth  judgments. 

Of  equal  importance  for  a  comprehensive  view  of  the  values 
of  characterisation  is  the  more  subjective  correlate  of  these 
judgments  of  personal  worth,  namely,  the  individual's  sense  of 
inner  peace,  or  absolute  worth,  which  comes  with  the  exclusion 
of  all  negative  elements,  and  with  them  of  effort  and  conflict. 
This  is  undoubtedly  a  real  factor  in  experiences  of  personal 
worth,  although,  like  the  experiences  of  the  preceding  paragraph, 
it  is  of  the  unique  and  non-habitual  type.  Unlike  the  feeling 
of  tragical  elevation,  which  comes  with  the  isolation  of  complete 
sacrifice,  and  which  supervenes  upon  effort,  these  experiences, 
of  which  the  mystical  religious  is  the  extreme  expression,  isolate 
the  self  by  identifying  it  with  a  Supreme  Being.  The  complete 
identification  of  the  individual  will  with  the  will  of  God,  in  which 
the  self  is  "  lost  in  wonder,  love,  and  praise,"  is  the  characteristic 
feature.  The  disposition  which  has  personal  worth  is  increased 
to  the  absolute  point  by  identification  of  the  individual  with  a 
being  in  which  the  absolute  is  assumed  to  be  already  realised. 
The  essentially  aesthetic  character  of  this  experience,  as  dis 
tinguished  from  the  ethical  character  of  the  feelings  of  obligation, 
appears  in  the  fact  that  that  which,  ethically,  is  always  a  goal, 
is,  in  the  religious  experience,  assumed  as  already  realised. 

There  can  be  no  manner  of  doubt  that  such  "  practical  " 
absolutes  exist,  and  that  they  are  believed  to  correspond  to 
actual  reality.  There  is  also  no  doubt  that  this  belief  finds 
expression  in  our  actual  judgments.  Whether  repeated  or  not, 
they  stand  as  representative,  as  the  measure  of  the  capacity  of 
the  individual  for  worth  experience.  As  such  they  determine 
his  obligations  and  judgments. 


Personal  Worths  309 


4.  The  Question  of  the  Validity  of  these  Judgments  of  Personal 
Worth  and  of  the  Implicit  Assumptions  or  Postulates  upon 
which  they  rest. 

Our  analysis  of  the  actual  judgments  of  personal  worth 
has  justified  the  main  contentions  of  this  chapter — that  personal 
worth  may  be  acquired  indefinitely;  that  there  are  absolute 
personal  worths ;  and  that  this  assumption  is  presupposed  as 
the  postulate  of  all  characterisation  of  persons  and  all  esti 
mation  of  personal  worth.  Psychologically,  this  belief,  this 
implicit  assumption,  is  explicable.  The  preceding  chapter 
was  devoted  entirely  to  the  study  of  the  genesis  of  the  assump 
tions  presupposed  in  feelings  of  personal  worth,  and  their  effect 
in  determining  our  actual  judgments — both  qualitatively  and 
quantitatively.  The  recognition  of  this  postulate  as  underlying 
all  judgments  of  personal  worth  was  the  culminating  point  of 
the  discussion. 

But  when  we  turn  from  the  purely  phenomenological  aspect 
of  the  problem  to  the  question  of  the  validity  of  these  judg 
ments,  and  therefore  of  the  assumptions  which  they  presuppose, 
a  somewhat  different  problem  arises.  May  not  these  assumptions, 
these  ideals  be  unfounded  in  fact,  and,  therefore,  the  judgments 
of  personal  worth  be  illusions  ?  When  absolute  worth  is  im 
puted  to  a  personality  in  the  moments  of  tragical  elevation  and 
inner  harmony,  it  is  on  the  basis  of  the  assumption  of  complete 
identification  of  the  person  with  the  ideal  object.  When  the 
subject  feels  absolute  obligation  toward  a  personal  worth,  or 
the  feeling  of  complete  inner  peace  which  comes  with  the  satis 
faction  of  that  obligation,  it  is  upon  the  assumption  of  the 
possibility  of  realising  the  object  or  the  belief  that  the  object 
is  already  realised.  If  these  assumptions  should  be  unfounded 
from  the  point  of  view  of  an  external  system  of  "  matter  of  fact," 
would  not  the  feelings  in  this  case  be  feelings  of  the  imagination, 
real  enough  as  psychological  phenomena,  but  unreal  in  the  sense 
that  their  cognitive  presuppositions  are  unfounded  ?  These 
assumptions  themselves  have  their  origin  psychologically  in 
imaginative  constructions  which  presuppose  the  aesthetic  isola 
tion  of  the  personality.  May  not  reflection  upon  the  "  causes 
of  things ' '  show  this  isolation  to  be  unreal,  and  the  assump 
tions  created  in  such  experiences  without  any  foundation  in 
fact  ?  If  such  should  turn  out  to  be  the  case,  we  might  very 


310  Valuation:   its  Nature  and  Laws 

properly  describe  them  as  "  pathetic  fallacies,"  on  the  analogy 
of  the  use  of  the  term  in  literature  and  art,  where  we  read  into 
inanimate  objects  of  nature  sentiments  and  ideas  which  are 
not  there  in  reality.  In  this  case  our  fallacy  would  be  that  we 
read  into  persons,  and  into  the  system  of  things  in  general, 
imaginative  constructions  which  do  not  exist,  and  assume  them 
to  be  real  instead  of  the  aesthetic  illusions  that  they  perhaps  are. 
This  problem,  it  will  be  readily  seen,  involves  the  whole 
question  of  the  "  rationality  "  of  these  assumptions  and  beliefs, 
a  question  already  discussed  in  a  preliminary  way  in  a  preceding 
chapter,  where  the  conditions  of  the  solution  of  the  problem 
were  stated.  The  question  as  to  what  constitutes  reality  and  as 
to  what  is  illusion  in  our  worth  experience,  now  appears  in  a  more 
concrete  and  pressing  form.  The  attempt  at  a  solution  we  shall 
reserve  for  the  axiological  discussions  of  the  concluding  chapter, 
where  the  general  problem  of  the  relation  of  judgments  of  value 
to  judgments  of  fact  and  truth  will  be  considered.  Meanwhile, 
we  must  turn  to  the  investigation  of  another  class  of  values,  the 
social  and  over-individual,  and  of  their  relations  to  the  personal 
and  individual.  It  is  frequently  the  facts  and  laws  of  these 
more  objective  valuations  which  seem  to  throw  doubt  upon  the 
assumptions  of  the  personal  sphere. 


CHAPTER   XI 

I.  OBJECTS  OF  IMPERSONAL,  OVER-INDIVIDUAL  VALUE — THEIR 
NATURE  AND  ORIGIN 

i.  Definition. 

A  THIRD  class  of  objects  of  value,  and  with  it  a  third  type  of 
valuation,  was  distinguished  in  our  introductory  chapter,  a 
class  of  objects  described  as  over-individual  and  a  type  of  valu 
ation  characterised  as  impersonal.  Objects  of  over-individual 
value  are  those,  the  value  of  which  is  founded  on  processes  of 
desire  and  feeling  which  are  social  and  over-individual ;  and  their 
valuation  by  the  individual  is  impersonal,  in  so  far  as  his  feelings 
and  judgments  presuppose  participation  in  the  impersonal  over- 
individual  desire  and  feeling.  Through  process  of  participation 
in  the  worth  consciousness  of  others,  the  feelings  and  desires  of 
the  individual  acquire  a  new  meaning,  an  over-individual  refer 
ence  to  ends  beyond  the  self. 

Two  classes  of  over-individual  objects  were  distinguished, 
the  moral  and  the  economic.  That  which  they  have  in  common 
is  the  fact  that  both  are  founded  in  a  certain  over-individual 
demand,  a  demand  which  may  be  variable  in  extent  and  in 
tensity,  but  which  is,  nevertheless,  always  felt  as  more  than 
personal.  Their  difference  lies  in  the  character  of  the  objects. 
The  economic  object  is  primarily  a  physical  object  of  condition 
worth  which  has,  through  processes  of  ideal  reconstruction 
consequent  upon  the  individual's  participation  in  the  worth 
judgments  of  others,  acquired  the  over-individual  reference 
that  makes  it  an  object  of  exchange  value.  The  moral 
object  is  an  act,  or  disposition  represented  by  the  act,  which 
is  no  longer  valued  merely  for  the  condition  worths  which 
result,  or  for  its  intrinsic  worth  for  the  personality,  but  for  its 
participation  value,  its  reference  to  this  over-individual  demand. 
An  economic  object  is  an  object  of  condition  worth  with  the 
acquired  capacity  of  exchange  ;  a  moral  object  is  a  disposition 


312  Valuation:   its  Mature  and  Laws 

of  a  person  which  has,  in  addition  to  its  immediate  subjective 
and  personal  meaning,  the  acquired  capacity  of  being  instru 
mental  to  certain  over-individual  ends,  participation  in  which 
is  socially  desirable.  Its  value  is,  therefore,  a  social  partici 
pation  value.  Both  values  are  objective  and  social,  objective 
exchange  and  objective  participation  value.  It  is  evident  that 
the  similarities  of  these  two  types  of  value  extend  so  far 'that 
in  both  cases  the  individual  who  passes  such  judgments  of  value, 
economic  or  moral,  becomes,  for  the  time  being,  an  "  impartial 
spectator,"  represents  an  attitude  in  valuation  attained  by 
abstraction  from  the  individual  and  personal  references  and 
meanings  of  the  object,  and  which  therefore,  to  the  extent  that 
this  detachment  is  realised,  is  impersonal. 

In  our  discussion  of  the  equivocations  in  judgments  of  value,1 
we  found  that  it  was  possible  to  attribute  to  the  same  object  posi 
tive  and  negative  worth,  according  to  differences  in  attitude  or  in 
the  presuppositions  of  the  feeling.  Thus  I  may  attribute  objective 
value  to  objects,  e.g.,  diamonds,  although  subjectively,  as  ob 
jects  of  condition  worth,  they  are  distasteful.  To  reach  this 
objective  impersonal  attitude,  a  certain  abstraction  from  sub 
jective  appreciations  is  necessary.  Through  participation  in 
the  worth  processes  of  others,  the  object  is  removed  from  im 
mediate  appreciation  and  becomes  mediately,  as  an  exchange 
value,  instrumental  to  more  remote  appreciations.  A  certain 
ideal  reconstruction  of  the  object,  described  in  a  previous 
chapter,2  has  taken  place.  I  now  conceive  the  value  of  the  object 
to  lie  in  its  capacity  of  exchange,  and  this  acquired  meaning 
is  founded  on  a  series  of  judgments,  which  in  constructing  the 
over-individual  demand,  have  at  the  same  time  abstracted  from 
immediate  appreciation. 

An  entirely  similar  situation  exists  in  the  case  of  the  morally 
qualified  judgment  upon  acts  and  dispositions,  the  psychical 
objects  of  moral  judgment.  Here  also  my  impersonal,  ob 
jective  judgment  upon  an  act  may  differ  widely  from  my  per 
sonal  judgment.  In  the  moral  judgment  that  which  determines 
the  value  of  the  act  is  not,  as  in  the  case  of  personal  worth,  its 
significance  as  indicative  of  a  feeling-attitude  or  disposition 
intuitively  realised  by  the  subject,  but  its  meaning  as  an  in 
strument  for  furthering  the  over-individual  ends  of  society. 
The  moral  participation  value  of  a  relation  of  the  sexes  "  with 
out  benefit  of  clergy  "  is  low  for  the  reason  that,  not  conforming 

1  Chap.  II,  pp.  22  f.  2  Chap.  VI,  pp.  169  f. 


Objects  of  Impersonal  Over-Individual  Value       313 

to  over-individual  demand,  it  set  limits  to  further  participation 
in  other  ends.  But  the  personal  worths  and  worths  of  appreci 
ation  which  grow  up  in  this  relation  may  possibly  be  of  the 
highest  order,  perhaps  even  because  of  the  social  isolation  and 
the  contrast  which  comes  with  it.  No  one  who  has  studied 
human  experience  closely,  from  an  unbiased  point  of  view,  will 
feel  disposed  to  deny  the  possibility  of  such  a  situation,  or 
to  deny  that  circumstances  are  conceivable  in  which  the  per 
sonal  obligation  to  such  relation  might  be  very  emphatic,  and 
in  which  the  disposition  displayed,  in  so  far  as  we  ignore  the  im 
personal  point  of  view,  might  lead  us,  not  only  to  condone,  but 
actually  to  impute  worth  to,  the  personality  involved.  But  from 
the  standpoint  of  the  distinctively  moral  or  social  j  udgment  such 
emotional  accompaniments  are  irrelevant. 

2.  The  Impersonal  Standpoint  in  Moral  Judgment — The  Morally 
Qualified  Act  and  the  Morally  Qualified  Judgment. 

Economic  and  moral  values  are  alike  objective  and  social, 
and  the  judgment  of  the  individual  in  such  cases  is  impersonal. 
It  is  the  impersonal  moral  judgment  upon  dispositions  which 
concerns  us  here,  since  a  detailed  study  of  economic  values  is 
not  within  the  province  of  this  investigation.  The  origin  and 
nature  of  economic  values  has  been  treated  in  a  general  way  in 
the  chapter  on  the  Laws  of  Valuation.1  With  this  present 
comparison  of  the  two  classes  we  may,  therefore,  turn  directly 
to  the  study  of  the  moral  values  of  dispositions. 

Returning,  then,  to  the  illustration  already  given  of  the 
distinction  between  ethical  and  quasi-ethical  or  personal  judg 
ments  and  moral  or  impersonal,  it  is  quite  evident  in  such  a  case 
—and  this  is  merely  an  extreme  instance  of  very  frequent 
contradictions  between  personal  or  private,  public  or  moral 
judgment — that  while  the  object  of  the  judgment,  the  act,  is 
superficially  the  same  in  both  cases,  the  meaning  of  the  act  is 
radically  different  in  the  two  situations.  The  presuppositions 
of  the  act  itself,  as  well  as  of  the  attitude  of  judgment  upon  the 
act,  represent  different  acquired  meanings.  One  meaning  of  the 
act  is  irrelevant  from  one  point  of  view,  the  other  equally  so  from 
another  point  of  view.  For  the  personal  attitude  and  judgment, 
the  act  has  meaning  as  a  spontaneous  expression  of  a  person 
ality,  and  is  measured  wholly  in  terms  of  sacrifice  of  lower 

1  Chap.  IV,  pp.  142  ff.  and  167  ff. 


3 T  4  Valuation  :   its  Nature  and  Laws 

condition  worths.  For  the  moral  point  of  view,  it  has  meaning 
as  readiness  or  failure  to  participate  in  an  over-individual  de 
mand,  and  is  measured  in  terms  of  readiness  to  sacrifice  indivi 
dual  and  personal  desires  and  demands. 

This  appears  in  both  the  morally  qualified  act  and  the  morally 
qualified  judgment.  The  morally  qualified  act  is  determined  by 
the  motive  presupposed.  An  act  is  thus  qualified  when  it 
expresses  a  disposition  to  impersonal  participation,  the  response 
to  an  impersonal,  over-individual  demand.  The  act  is  moral 
precisely  in  the  degree  to  which  subjective  and  personal  motives 
are  abstracted  from ;  and,  whereas  in  personal  imputation  the 
degree  of  imputed  personal  value  is  measured  in  terms  of  sacrifice 
of  condition  to  personal  worths,  in  impersonal  imputation — 
as  we  shall  see  later,  in  more  detail — the  moral  value  of  the  dis 
position  is  estimated  in  terms  of  sacrifice  of  egoism  for  altruism, 
egoism  including  from  this  point  of  view  both  condition  and 
personal  worths,  and  altruism  being  a  term  used  to  designate  all 
dispositions  to  participate  in  over-individual  ends.  Similarly,  the 
morally  qualified  judgment  is  the  judgment  determined  exclusively 
by  the  consciousness  of  the  over-individual  demand,  and  pre 
supposes  abstraction  from  individual  and  personal  demands.  It 
is  the  judgment  of  the  "  impartial  spectator,"  who  abstracts  from 
the  subjective  participation  value  of  an  act,  and  from  its  acquired 
complementary  personal  values,  and  reflects  in  his  judgment 
merely  the  objective  participation  value  of  the  disposition.1 

3.  Relativity  of  the  Distinction  between  the  Personal  and  the 
Impersonal  Attitude. 

But  while  the  morally  qualified  act  and  judgment  upon  that 
act  represent  ideally  the  impersonal  attitude  upon  the  part  of 
the  actor  and  spectator  alike,  nevertheless,  neither  particular 

1  That  the  foregoing  distinction  between  ethical  and  moral  values — and  the  consequent 
definitions  of  the  morally  qualified  act  and  judgment — will  commend  themselves  im 
mediately  to  the  reader  is  perhaps  too  much  to  expect.  The  confusion  underlying  the 
identification  of  the  two  has  been  too  prevalent,  and  of  too  long  standing  in  ethical 
writings,  to  be  easily  cleared  away.  For  this  very  reason,  the  limitation  of  the  term 
moral  in  our  way  is  perhaps  to  be  deprecated  :  a  term  entirely  unequivocal  in  meaning 
would  have  been  preferable,  if  indeed  such  were  to  be  found.  But,  on  tne  other  hand, 
there  are  in  favour  of  this  restricted  usage  two  reasons  which  to  the  writer  seem  decisive. 
In  the  first  place,  it  is  but  making  more  rigorous  for  accurate  analysis  a  real  distinction 
which,  as  we  have  already  seen,  is  present  in  all  valuation  of  acts,  and  is  even  recog 
nised  by  the  distinction  between  the  personal  and  the  moral.  In  the  second  place,  it  is 
justified  by  its  fruiifulness  in  analysing  and  interpreting  concrete  moral  judgments,  as 
they  are  presented  to  us  in  the  following  chapter.  The  results  of  that  analysis  must 
decide  whether  the  distinction  is  valid  or  not. 


Objects  of  Impersonal  Over- Individual  Value       3 1 5 

acts  nor  judgments  upon  these  acts  ever  show  the  moral  attitude 
in  its  purity.  Actual  situations  are  always  more  complex.  In 
the  sphere  of  concrete  judgments  the  distinction  between  the 
personal  and  impersonal  is  only  relative. 

In  our  study  of  personal  worths,  we  have  already  seen  that, 
while  there  were  some  aspects  of  the  person,  some  dispositions, 
which  are  wholly  irrelevant  for  impersonal,  moral  judgment, 
many  of  them  have  significance  for  both  types  of  judgment, 
and  become  irrelevant  for  moral  judgment  only  in  certain  quan 
tities.  Our  moral  judgment  takes  cognisance  of  them  only 
when  they  reach  a  certain  minimum,  and  ceases  to  find  them 
relevant  when  they  reach  a  certain  maximum.  Thus  it  was 
found  that,  in  the  valuation  of  the  same  object,  at  certain  points 
the  personal  passes  over  into  the  impersonal  and  the  impersonal 
into  the  personal  attitude.1  Again,  the  impersonal  attitude 
itself,  as  the  specifically  moral,  is  merely  the  limiting  case  of  a 
series  of  degrees  between  the  purely  personal  and  the  impersonal. 
Between  the  complete  isolation  of  the  personality,  which  is  the 
limiting  case  of  personal  imputation,  and  the  complete  ab 
straction  from  the  personal  moment,  which  is  the  ideal  of  moral 
judgment,  there  are  those  judgments  in  which  the  over-individual 
worth,  which  gives  the  disposition  its  participation  value,  is  a 
group,  class,  or  perhaps  national  worth.  The  ideal  object 
"  honour,"  already  considered  in  our  study  of  personal  worths, 
is  a  case  in  point.  Between  the  standard  of  honour  which  holds 
between  friends  and  that  which  is  abstractly  "  human,"  many 
specifications  of  the  standard  appear,  as  determined  by  class  and 
racial  consciousness.  The  objects  of  quasi-ethical,  ethical,  and 
moral  worth  judgment  tend  to  overlap,  and  the  subjects  of  these 
judgments  are,  therefore,  but  relative  differentiations  of  attitude 
acquired  through  differences  in  the  psychical  processes  pre 
supposed.  Moreover,  the  impersonal  attitude  of  the  "  impartial 
spectator  "  is  but  an  ideal  limit  at  which  the  personal  reference 
is  still  more  or  less  present. 

The  difference  between  personal  and  impersonal  judgment 
is  further  characterised  by  the  distinction  between  emotional 
and  intellectual  imputation.2  Here,  again,  the  relativity  of  the 
distinction  is  apparent.  Every  worth  judgment,  as  worth 
judgment,  is  the  expression  of  some  feeling.  Accordingly, 
when  the  impersonal,  moral  judgment  is  said  to  be  intellectual, 
the  expression  of  practical  reason  as  opposed  to  emotion,  it  is 

1  Chap,  x,  pp.  283  f.,  300  f.  -  Chap,  x,  p.  290  note. 


316  Valuation:    its  Nature  and  Laws 

meant  merely  that  certain  emotional  aspects  of  the  personal 
relation  which  are  determinants  in  personal  imputation  have 
become  irrelevant— not  that  the  attitude  itself  has  become 
purely  cognitive  and  has  ceased  to  be  emotional.  More  specif 
ically,  the  apprehension  of  the  over-individual  meaning  of  the 
individual's  act  and  disposition,  and  the  entrance  of  that  cog 
nitive  act  as  a  presupposition  of  the  feeling  of  value  of  the 
object,  inhibits  those  emotions  which  are  significant  merely  for 
immediate  and  personal  participation. 

Two  facts  stand  out  clearly— first,  that  the  objects  of  moral 
judgment   are  but  a  narrower,   differentiated  class   within  the 
larger  group   of  ethical  and  quasi-ethical  worths,  and  partially 
identical  with  them ;  and,  secondly,  that  the  impersonal  moral 
attitude  of  the  subject  of  the  judgment  is  attained  by  abstraction 
from  personal  and  emotional  elements,  whether  as  determinants 
of    the    act    or    of   the  judgment  upon  the   act.      From   these 
facts  we   may   draw  certain   important  conclusions   in    regard 
to  the  origin  and  nature  of  the  moral  standpoint  and  object. 
In  the  first  place,  it  is  clear  that  the  presuppositions   of  both 
types   of  judgment,   the   implicit   assumptions    and    demands, 
of  the  personal  and  impersonal  attitude  alike,  are  acquired  in 
processes  of  sympathetic   participation,  and   that   the   objects 
are  products  of  ideal  constructions  founded  on  these  processes. 
As  has  been  shown  in  our  sketch  of  sympathetic  participation, 
the  feeling-attitude  of  the  alter  has  first  an  immediate  appreci 
ative  value  for  the  subject,  what  we  may  in  this  connection 
describe  as  the   subjective  participation  value  of  the  attitude. 
Out  of  this,  on  the  one  hand,  develops  the  characterisation  of 
the   person,   and    the    independent,   intrinsic  valuation   of   the 
attitude  as  an  expression  of  the  person.     This  is  the  personal 
value  of  the  disposition.     On  the  other  hand,  from  this  "  sub 
jective  "  participation  value,  develops  the  objective  participation 
value,    the   over-individual   value   reflected   in   the   impersonal 
moral  judgment,  the  instrumental  value  which  the  disposition 
has  as  contributory  to  social  over-individual  ends.     But  in  the 
second  place,  it  is  equally  apparent  that  the  objective  over- 
individual   values,   consciousness    of   which   is   presupposed   in 
moral  judgment,  are  determined  by  social  demands  in  which 
the  individual  merely  participates  ;  they  are  the  product  of  social 
interaction,  or  sympathetic  participation  in  its  social  aspect. 

Our  task  is  now  to  define  these  objective  social  values  more 
fully,  to  determine  their  origin,  nature,  and  laws,  and  to  show 


Objects  of  Impersonal  Over- Individual  Value      3 1 7 

how  they  are  presupposed  in  moral  judgment.  The  present 
chapter  will  be  devoted  to  this  study  of  objective  partici 
pation  value,  while  the  following  will  attempt  to  show  that 
the  moral  judgment,  in  its  two  aspects  of  obligation  and  im 
putation  of  praise  and  blame,  reflects  these  values  and  their 
laws, 

II.  SUBJECTIVE  AND  OBJECTIVE  PARTICIPATION  VALUE  OF 

DISPOSITIONS 

i.  Their  Relations. 

The  over-individual  "  participation  value  "  of  a  disposition 
or  quality  is,  it  has  been  said,  a  function  of  the  relation  of  that 
object  to  social  over-individual  demand,  and  the  moral  judg 
ment  of  the  "  impartial  spectator  "  has  as  its  presupposition 
a  consciousness  of  this  relation.  With  this  statement  of  the 
situation,  it  becomes  evident  that  the  problem  of  the  ensuing 
study  is  twofold.  In  the  first  place,  we  have  to  consider  the 
nature  and  origin  of  this  over-individual  demand  and  of  the  ideal, 
over-individual  objects  of  the  demand  ;  and,  secondly,  the  manner 
in  which  the  individual  participates  in  this  demand,  and  how  the 
meaning  acquired  in  this  process  determines  his  feelings  and 
judgments  of  impersonal  value— how  the  consciousness  of 
over-individual  ends  and  demands  becomes  a  presupposition 
of  his  feelings.  The  problem  is,  accordingly,  one  involving 
both  individual  and  social  psychology. 

By  this  statement  it  is  meant  that  the  over-individual  de 
mand,  in  which  the  objective  participation  value  of  a  disposition 
is  founded,  must  be  studied  in  two  aspects:  as  a  social  fact, 
the  product  of  social  interaction ;  and  as  a  cognised  presuppo 
sition  of  the  individual's  feelings  and  judgments.  As  an  ob 
jective  social  fact,  it  is  the  product  of  social  interaction,  of 
sympathetic  participation  in  its  objective  social  aspect.  But 
as  a  product  of  social  interaction  it  is  the  resultant  of  modifications 
of  the  subjective  feelings  of  value  of  individuals,  as  determined 
by  these  processes  of  sympathetic  participation.  There  is, 
therefore,  a  definite  relation  between  the  subjective  and  ob 
jective  participation  values  of  dispositions.  This  relation  we 
must  now  seek  to  determine. 


318  Valuation:   its  Nature  and  Laws 


2.  Collective  Desire  and,  Feeling — Social  Synergies 
Demand  and  Supply. 

External  to  the  individual,  and  often  apart  from  any  con 
sciousness  in  him  of  their  existence,  there  are  certain  collective 
or  aggregate  desires  or  demands  which  may  be  described  as 
the  social  will,  and  which  may  be  localised  in  larger  or 
smaller  social  groups.  These  collective  desires  or  trends  owe 
their  existence  to  the  simple  fact  that,  in  consequence  of  uni 
formity  in  the  organic  and  biological  conditions  of  psychical  life, 
men  have  similar  instincts  and  desires,  and,  consequently, 
similar  passions  and  emotions.  It  is  possible  that  such  desires 
and  feelings  may  be  shared  without  any  accompa^-ing  con 
sciousness  of  their  "  common  meaning."  A  demand  to  which  I 
am  myself  a  contributing  factor  might  conceivably  exist, 
indeed  often  does  exist,  without  my  being  conscious  of  its  ex 
istence.  As  such,  my  desire  is  but  a  part  of  a  collective  or 
aggregate  demand.  But  it  is  possible  that  I  may  also  be  con 
scious  of  the  demand  of  which  my  desire  is  a  part.  Not  only 
has  the  object  of  desire  then  acquired  a  common  meaning,  but 
I  have  become  conscious  of  that  common  meaning,  and  in  so 
far  as  this  consciousness  enters  as  presupposition  of  my  feeling, 
my  desire  is  itself  modified,  and  with  it  my  feelings  and  judg 
ments  of  value.1 

Collective  desire  and  feeling,  when  it  has  acquired  this 
"  common  meaning,"  when  the  object  of  desire  and  feeling  is 
consciously  held  in  common,  we  may  describe  as  Social  Synergy  ;2 
and  the  objective,  over-individual  values  may  be  described  as  the 
resultants  of  social  synergies.  The  introduction  of  this  term  has 
for  its  purpose  the  clearest  possible  distinction  between  social 
forces  as  conscious  and  as  sub-conscious.  It  is  with  the  former 
that  we  are  here  concerned,  and  our  problem  is  to  analyse  these 
social  processes  of  conscious  inter-action  by  which  the  objective 
over-individual  values  are  determined.  These  processes  are 

1  Cf.  in  this  connection  Baldwin's  study  of  aggregate,  con-aggregate,  and  public 
meaning,  Thought  and  Things,  Vol.  I,  Chap,  vii,  sects.  5-10,  and  of  their  higher  stage, 
syndoxic  and  synnomic  meaning.  Ibid.,  Vol.  II,  chap.  in. 

The  term  Synergy  is  here  used  on  the  analogy  of  its  use  in  Psychology  (cf.  the 
definition  of  the  term  in  Baldwin's  Dictionary  of  Philosophy  and  Psychology),  where  by 
motor  synergy  is  understood  the  working  together  of  motor  tendencies  to  form  a  totality. 
The  term  Co-operation,  while  more  in  use,  has  this  disadvantage,  that  it  covers  only  a 
limited  field  of  the  phenomena  with  which  we  are  concerned.  Co-operation  implies 
participation  in  the  intellectual  processes  of  devising  means  to  ends,  while  many  phases 
of  participation  are  purely  sympathetic  and  emotional. 


Objects  of  Impersonal  Over-Individiial  Value       319 

the  activities  of  sympathetic  participation  and  ideal  construction 
already  studied  from  the  point  of  view  of  abstract  psychological 
analysis  of  the  individual.  Our  task  is  now  to  study  them  as 
factors  in  the  creation  of  those  social  synergies,  those  factors 
of  "  demand  and  supply  "  which  determine  the  over-individual 
participation  value  of  dispositions. 

The  two  factors  of  social  synergy,  demand  and  supply,  the 
demand  for  socially  desirable  acts,  and  the  readiness  of  indi 
viduals  to  supply  them,  are,  in  so  far  as  they  presuppose  this 
common  meaning,  the  products  of  social  thought  and  sympathy, 
that  is,  of  the  sympathetic  participation  of  the  individuals  in 
common  ends.  Through  sympathetic  participation  in  collective 
desire  and  feeling,  the  individual  becomes  aware  of  the  over-indi 
vidual  damand,  and  this  awareness  modifies  his  disposition  to  act 
or  to  expect  actions  from  others.  The  social  supply  is  similarly 
conditioned.  It  is  the  product  of  the  acts  of  individuals  con 
tributing  to  the  social  end,  and  the  dispositions  of  the  indi 
viduals  to  participate  and  to  contribute  to  the  social  supply 
is  determined  by  consciousness  of  the  demand  acquired  through 
sympathetic  participation.  This  consciousness,  although  be 
ginning  with  emotional  contagion  and  simple  appreciation  of 
common  organic  function,  passes  through  common  judgment 
and  belief,  to  common  sentiments  and  ideals. 

The  objective  participation  value  of  a  disposition  is,  then, 
a  function  of  the  two  factors  of  social  synergy,  supply  and 
demand ;  and  these  factors  are  determined  by  the  conscious 
ness  of  common  meaning  on  the  part  of  the  individuals  par 
ticipating.  In  order,  therefore,  to  determine  the  nature  of  these 
factors  and  their  laws,  it  is  clearly  necessary  to  discover  how  this 
consciousness  of  common  over-individual  meaning  modifies  the 
individual's  disposition  to  participate  and  to  demand  partici 
pation  on  the  part  of  others.  From  these  facts  and  laws  of 
subjective  participation  value  it  may,  then,  be  possible  to  deduce 
the  laws  of  social  synergy  and  of  the  objective  participation 
value  of  dispositions. 

In  the  chapter  on  the  Laws  of  Valuation  we  found,  it  will 
be  remembered,  that  economic  method  consists  in  the  analysis 
of  the  laws  of  subjective  value,  on  the  assumption  that  the 
objective  or  exchange  value  can  be  developed  from  the  laws 
of  subjective  value,  a  procedure  which  justifies  itself.  The 
situation  here  is  analogous,  for,  as  in  the  case  of  an  economic 
good,  its  exchange  value  is  a  function  of  the  laws  of  desire 


320  Valuation  :  its  Nature  and  Laws 

and  feeling,  of  subjective  worth  in  the  individuals  desiring 
the  good,  so  the  participation  value  of  a  given  disposition  is  a 
function  of  the  laws  governing  the  tendency  or  disposition  to 
participate  of  the  individuals  involved.  Consequently,  having 
studied  the  phenomena  of  sympathetic  projection  and  the 
consequent  modifications  of  feelings  of  value  and  dispositions 
in  the  individual,  we  may  pass  to  the  objective,  social  point  of 


view. 


III.  THE  LAWS  OF  SUBJECTIVE  PARTICIPATION  VALUE 

i.  The  Individual's  Feelings  of  Participation  Value  as  Determined 
by  Social  Sympathy — Extensive  Sympathetic  Projection. 

The  individual  becomes  aware  of  collective  desire  and  feeling 
through  extensive  sympathetic  projection.  In  this  process 
his  feelings  of  value  acquire  an  over-individual  reference  and 
meaning.  Analysis  of  extensive  projection,  and  its  inducing 
conditions,1  leads,  moreover,  to  the  following  general  conclusion  : 
with  the  extension  of  the  range  of  sympathetic  participation, 
the  feeling  of  participation  becomes  more  and  more  impersonal, 
and  the  object  of  participation  value  becomes  more  and  more 
abstract.  Let  us  recall  briefly  the  grounds  for  this  conclusion. 

In  the  first  place,  the  feeling  becomes  over-personal  in  char 
acter.  Its  transgredient,  over-individual  reference  is  beyond  the 
personality.  Even  on  the  level  of  simple  organic  sympathy, 
immediate  participation  in  group  emotion  or  passion  "  takes  the 
individual  out  of  himself."  The  forces  of  group  suggestion 
inhibit  the  more  individual  and  personal  feelings  and  reactions, 
until  finally  the  abstraction  of  the  feeling  from  its  purely  indi 
vidual  presuppositions  reaches  such  a  point  that  it  is  no  longer 
referred  back  to  the  self  and  identified  with  the  self  in  an  act 
of  judgment.  The  presupposition  of  the  feelings  of  value  in 
such  organic  sympathy  is  the  vague  presumption  of  an  over- 
individual  trend  or  demand  not  definitely  localised  in  a  person 
ality,  either  in  the  self  or  the  alter.  While  still  implying  the  self 
remotely,  it  is  now  free  from  explicit  reference  to  private  and 
personal  meanings.  In  the  second  place,  the  conditions  of  ex 
tensive  projection,  the  character  of  the  inducing  conditions,  lead 
to  a  certain  selection  among  the  feeling-attitudes  of  the  individual. 
Only  the  most  fundamental  and  general  attitudes  are  susceptible 

1  Chap,  via,  pp.  253  ff. 


Objects  of  Impersonal  Over- Individual  Value      321 

of  extensive  projection,  and  of  acquiring  this  impersonal,  over- 
individual  reference.  The  condition  of  projection  of  an  attitude 
of  an  individual  into  social  groups  is  abstraction  from  the 
individual  and  personal  presuppositions  and  references  of  the 
feeling.1  From  the  consideration  of  these  facts  it  is  apparent  that 
the  over-individual  demand,  as  immediate  or  merely  felt,  is  already 
over-personal  in  character.  The  demand  for  a  given  attitude  or 
disposition,  though  vaguely  felt,  is  not  localised  in  any  person, 
either  the  self  or  the  alter,  and  may  thus  ultimately  become  the 
presupposition  of  a  relatively  impersonal  and  impartial  attitude. 
With  the  emergence  of  this  impersonal  qualification  of  the 
transgredient  reference  of  feeling,  begins  the  ideal  construction 
of  over-individual  ends  to  which  the  feeling  is  referred.  As  in 
the  case  of  feelings  of  personal  worth,  the  acquired  meaning  of 
the  feeling  is  referred  to  the  ideal  construct  of  the  personality, 
so  here  the  acquired  impersonal  reference  leads  to  conceptions 
of  an  over-individual  will  and  of  impersonal  ends.  But  it  is 
at  this  point  that  the  chief  difference  between  the  two  types  of 
feelings  and  of  their  presuppositions  appears.  The  ideal  con 
struction  of  the  "  person  "  is,  as  we  have  already  seen,  indi 
viduating  and  intuitive  in  character,  while  the  over-individual 
objects  are  abstract  and  conceptual.  Beginning  with  the 
initial  contrast  between  condition  and  personal  worths,  and  the 
idealisation  of  the  person  described,  the  detachment  of  the  indi 
vidual  from  social  relations  further  advances  until  he  is  intrinsic 
ally  valued  as  an  end  in  himself.  The  disposition  of  the  person 
in  question  is  intrinsically  valued  as  an  expression  of  the  person, 
and  acquires  complementary  value  through  its  relation  to  the 
person  conceived  as  an  individuated  whole.  In  the  case  of  the 
ideal  construction  based  upon  extensive  sympathetic  partici 
pation,  the  situation  is  otherwise.  With  the  growing  imperson 
ality  of  the  feeling,  the  ideal  construct  to  which  it  is  referred 
becomes  more  and  more  abstract.  With  every  increase  in  the 
extension  of  the  concept  of  the  over-individual  end  to  which 
the  feeling  refers,  it  is  progressively  more  and  more  difficult 
to  refer  it  back  to  the  person  as  an  intrinsic  personal  end, 
in  other  words,  to  visualise  it  concretely.  Thus  the  indi 
vidual  is  felt  to  be  merely  the  locus  or  the  bearer  of  these  social 
over-individual  ends  or  ideals,  and  his  dispositions  are  conceived 
as  merely  instrumental  or  contributory  to  their  realisation.  The 
consequences  of  this  are  significant.  As  soon  as  immediate 

1  Chap,  vin,  p.  257. 


322  Valuation:   its  Nature  and  Laws 

sympathetic  participation  develops  into  ideal  construction 
and  judgment,  the  feelings  of  value  have  as  their  presuppositions 
instrumental  judgments,  and  the  demand  represents  not  in 
trinsic  but  instrumental  values. 

2.  Feelings  of  Participation  Value  as  Modified  (Quantitatively)  by 
Extensive  Projection — The  Laws  of  Social  Sympathy. 

The  modification  through  extensive  projection,  of  the 
presuppositions  of  the  individual's  feelings,  and  their  conse 
quent  change  in  reference  have  now  been  shown.  A  certain 
relation  between  the  extension  of  the  common  meaning,  the 
impersonality  of  the  individual's  feeling,  and  the  degree  of 
abstractness  of  the  object  of  the  feeling  has  appeared.  Is  it 
possible  to  establish  any  relation  between  this  extension  of  the 
common  meaning  and  its  intension,  between  the  degree  of 
expansion  of  sympathy  and  its  intensity  ?  If  such  a  relation 
appears,  it  may  afford  the  basis  for  the  formulation  of  the  laws 
of  subjective  participation  value,  and  ultimately  of  objective 
social  values. 

The  primary  condition  of  the  sense  of  over-individual  value 
is,  as  we  have  seen,  sympathetic  participation  in  its  aspect 
of  extensive  projection.  Beginning  with  its  simplest  form, 
where  the  conditions  are  similarities  of  organic  conditions, 
and  developing  through  the  stages  of  imaginative  and  ideal 
construction,  the  extension  of  the  range  of  sympathetic  pro 
jection  involves  abstraction  from  personal  presuppositions  and 
references,  and  therefore  a  qualitative  modification  in  the  direc 
tion  of  impersonal  reference.  This  qualitative  change  is  also 
accompanied  by  characteristic  changes  in  the  quantitative  aspect 
of  the  feeling.  In  order  to  understand  these  changes  we  must 
make  a  careful  analysis  of  the  quantitative  factors  involved. 

The  factors  here  under  consideration  are,  on  the  one  hand, 
the  degree  of  sympathy  felt,  or  its  intensity?  and,  on  the  other 
hand,  the  extension  of  the  range  of  the  sympathetic  feeling, 
through  the  inclusion  of  more  and  more  individuals  in  the  group 
in  which  the  individual  participates.  What  is  the  effect  of 
extension  of  sympathy  upon  its  intensity  ?  And  further- 
since  the  desire  to  participate  in  over-individual  ends  is  a  function 
of  the  intensity  with  which  the  over-individual  demand  is  felt, 

1  The  term  intensity  is  here  used  in  the  general  sense  of  de.^ree,  not  "sensational 
intensity  "  (Cf.  chap,  in,  pp.  73  f.). 


Objects  of  Impersonal  Over-Individual  Value      323 

how  is  this  desire  affected  through  the  satisfaction  of  the  demand 
by  the  acts  and  dispositions  of  others  ? 

If  we  begin  with  that  form  of  extensive  sympathy  described 
as  participation  in  group  desire  or  emotion,  immediate  partici 
pation  in  collective  attitudes,  we  find  that  the  effect  of  such 
participation  upon  the  feelings  of  the  individuals  participating 
is  clearly  marked.     We  have  already  seen  that  in  such  cases  of 
"contagious    emotion,"    expansion    of   the    feeling   is    possible 
only  in   the  case  of  relatively  primitive  and  undifferentiated 
emotional  attitudes,  and  is  limited  to  relatively  short  periods. 
Within  relatively  small  groups,  and  for  short  periods,  the  emotion 
or  passion  may  attain  a  high  intensity  and  complete  expansion, 
so  that  the  group  feels  intensely  and  as  one  man.     We  may 
almost  speak  of  a  mob  soul.     As  reflected  in  the  individual 
thus  participating,   the  effect  is  seen  in  an  intensification  of 
his    feeling  of    over-individual    reference.     Within    limits,  the 
degree  of  sympathy,  and  with  it  the  intensity  of  the  over-indi 
vidual  demand,  increases  with  the  extension  of  the  range  of 
sympathy.     This   temporary   increase   in   the   feeling   of  value 
is  manifested  in  increase  of  disposition  to  participate,  to  respond 
to  the  social  demand,  and  in  increase  of  demand  for  similar 
response  on  the  part  of  others.     But  it  is  a  well-known  fact  that 
such  participation,  based  upon  organic  sympathy,  is  followed 
by  phenomena  analogous  to  the  experiences  of  dulling  of  sensi 
tivity  and  satiety,  which  we  have  seen  to  accompany  aU  excess 
of  organic  sympathy.     Extensive  sympathy  of  this  fortuitous 
kind,  based  upon  external  and  often  superficial  similarities  of 
attitude,  and  made  possible  only  through  the  arrest  of  more 
individual  and  personal  attitudes   and  habits,   is  followed  by 
reaction.     The  temporary  presumption  of  existence  of  objects 
and  ends,  corresponding  to  the  group  emotion  or  passion,  fails 
to  develop  into  judgment  and  judgmental  habit,  or  implicit  as 
sumption. 

It  is  obvious  that  while  the  experiences  of  immediate 
participation  in  group  emotions  and  passions  may  afford  the 
basis  of  consciousness  of  over-individual  ends  and  demands, 
this  is,  in  the  very  nature  of  the  case,  limited  to  rare  occasions 
and  to  relatively  small  groups.  Any  extension  of  social  sym 
pathy  beyond  these  limits  must  rest  upon  imaginative  and 

LI^  Well-kn?wn  effect  °Tf  emotional  revivals  in  producing  dulling  of  sensitivity  and 
wo  dyfor  thesfeff^V11  ^T  ^^  °f  the  West  **  ^  have  a  significant 
"burnt  out  <"  Y  SP  °f  thC  regi°n  Visited  ^  Such  mob  emotions  as 


324  Valuation :   its  Nature  and  Laws 

conceptual  constructions.  Our  ability  to  realise  sympathetically 
over-individual  trends  and  demands  depends  upon  conditions  of 
a  more  intellectual  character.  Imaginative  projection,  with  its 
assumption  of  the  existence  of  corresponding  dispositions, 
depends  upon  similarity  of  attitude.  How  does  extension  of 
the  range  of  sympathetic  projection  modify  the  conditions 
of  such  sympathetic  participation,  and,  therefore,  the  degree  of 
sympathy  ? 

It  has  already  been  shown  that  the  condition  of  imaginative 
projection  is  abstraction  of  the  feeling  from  its  purely  individual 
reference,  that  it  is  only  in  this  way  that  the  "  person  project  " 
acquires  its  common  over-individual  meaning.  It  has  also  been 
shown,  however,  that  if  the  processes  of  abstraction  from  private 
and  personal  meanings  continue,  the  reference  of  feeling  becomes 
impersonal ;  the  feeling-abstract  is  no  longer  referred  back  to  the 
person,  but  is  projected  and  localised  beyond  the  person.  Now  the 
important  point  in  this  connection  is  that,  owing  to  the  nature  of 
the  inducing  conditions  of  imaginative  projection,  we  find  with 
increase  in  the  range  of  sympathy,  decrease  in  similarity  of  in 
ducing  conditions,  and,  consequently,  growing  abstraction  and 
impersonality  of  the  feeling.  It  may  be  shown,  I  think,  that, 
with  this  decrease  in  resemblance  and  consequent  growth  in 
abstractness  and  impersonality  of  the  feeling,  the  feeling  of 
sympathy,  and  with  it  the  feeling  of  value  of  the  attitude  in 
question,  will  decrease  in  intensity. 

Several  significant  facts  indicate  the  truth  of  this  inference. 
In  the  first  place,  as  has  already  appeared,  extensive  projection 
exercises  a  selective  influence  among  our  feeling  attitudes. 
Owing  to  the  variations  in  individuals,  the  nature  and  con 
ditions  of  which  have  already  been  considered,  there  is  a  weed 
ing  out  of  all  feeling-attitudes  which  are  unique  and  personal, 
these  being  retained  for  intimate  and  personal  relations  alone. 
Certain  others  will  be  stamped  as  class  and  group  attitudes,  and 
only  the  most  fundamental  attitudes  will  survive  in  more 
extended  sympathetic  projection.  But  a  similar  selection  is 
made  with  reference  to  the  quantitative  aspect  of  feeling- 
attitudes.  Even  in  the  case  of  those  attitudes  which,  so  to 
speak,  survive  extensive  as  distinguished  from  intensive  pro 
jection,  there  is  a  characteristic  modification  of  the  intensity 
of  the  projected  feeling.  Gradually  the  extremes  of  feeling, 
the  unique  and  individual  variations,  are  inhibited  in  favour 
of  a  certain  normal  intensity  which  in  actual  experience  has  been 


Objects  of  Impersonal  Over- Individual  Value       325 

ound  to  meet  with  response.  Accordingly,  we  may  go  a  step 
further  and  say  that,  with  increase  in  the  range  of  sympathy 
there  is  not  only  decrease  in  similarity  of  inducing  conditions, 
and  consequently  growth  in  abstractness  and  impersonality  of 
the  feeling,  but  also,  as  the  result  of  these,  decrease  in  intensity.1 
Immediate  sympathetic  participation,  whether  it  be  conditioned 
by  similarity  of  organic  conditions  or  by  similarities  of  ideal 
presuppositions  of  the  feelings,  is  thus  limited  by  the  psycho 
logical  nature  and  conditions  of  sympathy.  This  fact  evidently 
has  a  meaning  both  for  individual  and  social  psychology.  It 
states  both  the  law  according  to  which  the  feeling  of  the  demand 
to  participate  is  modified  in  the  individual's  consciousness, 
and  also  the  law  of  this  demand  in  its  more  social  aspect. 

Viewing  sympathetic  participation  in  this  objective,  socio 
logical  way,  we  find  the  truth  of  this  law  manifest  in  illustra- 
trations  which  are  numerous  and  instructive.  Among  primitive 
men,  social  sympathy,  and  the  concomitant  sense  of  collective 
values,  is  intense  and  immediate.  The  limitations  of  the 
group,  the  relative  homogeneity  of  their  affective  experience, 
and  of  its  corresponding  expressions,  are  conditions  which 
favour  rapid  expansion  of  feeling  and  its  dominance  over 
the  consciousness  of  the  entire  group.  Within  more  highly 
organised  societies,  where  the  organisation  consists  in  differentia 
tion  of  smaller  groups  within  the  larger  whole,  we  find  the  most 
intense  emotional  participation  in  connection  with  those  af 
fective  attitudes  which  correspond  to  limited  groups.  And  as 
a  further  consequence,  we  find  the  actual  feelings  of  obligation 
and  judgments  of  praise  and  blame  more  emphatic  with  reference 
to  group  attitudes.  We  need  call  attention  only  to  those 
specialised  obligations  and  virtues  of  the  politician  and  the 
labourer  within  our  own  civilisation,  and  to  the  equally  special 
ised  obligations  of  military  classes  in  societies  of  the  aristocratic 
type. 

1  It  is  facts  of  this  character  which  Giddings  (Inductive  Sociology,  p.  217)  has  sought 
to  gather  together  in  his  formulation  of  the  so-called  "  Law  of  Sympathy."  "  Using 
the  word  sympathy,"  he  says,  "  for  all  the  feelings  which  are  included  in  the  conscious 
ness  of  kind,  the  law  of  sympathy  is,  the  degree  of  sympathy  decreases  as  the  generality 
of  resemblance  increases."  Further,  "  it  loses  intensity  as  it  expands  to  the  more  remote 
resemblances,  and  becomes  intense  as  it  contracts  to  the  narrower  degrees." 


326  Valuation :   its  Nature  and  Laws 

3.  The  General  Law  of  Subjective  Participation  Value— The  Law 
of  Limiting  Value. 

The  conclusion  thus  far  reached  is  that,  while,  up  to  a  point, 
with  extension  of  sympathy  its  intensity  increases,  with  con 
tinued  extension  the  degree  of  sympathy  decreases.  With  the 
growing  impersonality  of  the  reference  the  intensity  is  lowered. 
This  general  law  finds  its  explanation,  apparently,  in  the 
limits  set  to  social  sympathy  by  the  nature  of  its  psychological 
conditions.  Immediate  sympathy,  with  its  intuitive  projection 
of  the  feeling  attitude,  becomes  more  and  more  impossible  with 
extension  of  the  social  group.  Our  feelings  and  judgments  of 
value,  in  so  far  as  they  presuppose  this  immediate  sympathetic 
participation,  with  its  underlying  assumptions,  manifest  the 
working  of  this  law.  Nevertheless,  all  our  feelings  and  judg 
ments  of  participation  value  are  not  necessarily  conditioned 
by  this  immediate  intuitive  sympathy.  On  the  basis  of  imme 
diate  sympathetic  participation,  conceptual  constructions  of 
dispositions  and  over-individual  ends,  to  which  these  dispositions 
are  instrumentally  referred,  are  built  up,  and  when  once  these 
concepts  are  formed,  the  dispositions  are  valued  instrumentally. 
What  is  the  effect  of  increase  of  quantity  of  such  dispositions 
upon  the  individual's  feeling  of  their  value  ? 

The  quantity  of  the  supply  of  such   dispositions  may  be 
increased  or  decreased  either  by  the  addition  or  subtraction  of 
new  individuals  to  the  number  of  participants,  or  by  increase  or 
decrease  in  the  amount  of  the  given  disposition  in  members 
of  the  social  group.     The  question  is  :    What  is  the  effect  of 
these  quantitative  factors  upon  the  individual's  feelings  of  value 
—and    consequently    upon    the    affective-volitional    disposition 
to  participate   which    these    feelings   presuppose?      How   does 
the  increase  of  the  quantity  of  the  over-individual  good  affect 
the  individual's  desire  to  add  to  it  by  participation,  and  his 
feeling  of  value  when  others  add  to  it  by  acts  of  participation  ? 
I  think  there  can  be  no  question  that  these  feelings  of  partici 
pation  value,  and  consequently  the  dispositions  to  participate, 
are  subject  to  the  general  law  of  Limiting  Value  developed  in 
the   chapter   on    Laws   of   Valuation.     In   general,   with   each 
additional  judgment  of  the  extension  of  the  area  of  partici 
pation,  or  of  increase  in  the  area  of  the  common  good  through 
these  increments,  there  is  a  relative  falling  off  in  the  increment 
to  the  degree  of  worth,  until  finally  a  point  is  reached  where 


Objects  of  Impersonal  Over- Individual  Value      327 

additions  to  the  total  good  are  without  appreciable  effect  upon 
worth  feeling. 

There  are  several  facts  which  indicate  in  a  significant  manner 
the  existence  of  this  law.  In  the  first  place,  the  individual's 
desire  to  participate — and  ultimately,  we  shall  see,  his  feeling  of 
obligation  to  participate — reflects  this  situation.  When  once 
the  demand  for  the  over-individual  object  has  been  created, 
the  individual's  desire  to  add  to  the  sum  of  social  good  seems  to 
depend  upon  the  degree  to  which  the  good  is  already  realised. 
The  greatest  energy  of  participation  appears  normally  where 
the  lack  of  the  social  good  is  most  apparent,  and  falls  off  as  the 
disposition  to  participate  becomes  more  and  more  general. 
Where  voting  is  a  general  duty  and  opportunity,  and  where 
the  act  of  voting  constitutes  a  contribution  to  over-individual 
impersonal  good,  we  feel  both  the  desire  and  obligation  to  vote 
when  the  participation  is  least ;  and  as  it  becomes  more  general 
our  feeling  of  its  worth  and  our  desire  to  participate  diminishes. 
In  like  manner,  any  contributions  to  social  worths,  such  as  acts 
of  politeness,  are  felt  as  of  most  worth  when  in  contrast  to  a 
more  or  less  general  lack  of  civility.  They  do  not  rise  above 
the  threshold  of  worth  judgment  when  such  participation  is 
universal.  On  the  other  hand,  up  to  the  point  at  which  these 
deficiencies  are  emphatically  felt,  and  we  identify  ourselves  with 
the  rising  minority  demand,  there  is  a  tendency  to  fall  in  with 
those  who  abstain  from  the  acts  in  question,  for  the  reason  that 
it  is  of  no  use  to  do  otherwise.  The  actual  participation  value 
of  our  act  is  insignificant. 

In  general,  then,  the  subjective  participation  value  of  a 
disposition,  whether  it  be  intrinsic  or  instrumental,  whether 
it  be  immediate  or  mediate,  seems  to  be  subject  to  the  law  of 
Limiting  Value.  Of  course  such  an  abstract,  general  statement 
of  the  facts  must  be  modified  in  important  particulars,  as  we 
shall  see  in  more  detail  presently,  when  to  the  impersonal  par 
ticipation  value  of  the  act  personal  and  group  worths  are  added 
as  complementary  values.  When  the  object  with  over-individual 
reference  has  also  a  personal  or  class  reference,  acquired  in  im 
mediate  emotional  participation,  this  law  will  be  modified  in 
the  same  manner  as  the  law  of  Marginal  Utility  for  objects  of 
consumption  is  modified  by  the  principle  of  complementary 
values.  Just  as  the  details  of  a  feast  or  of  dress,  or  an  insig 
nificant  fraction  of  our  total  wealth,  themselves  without  appre 
ciable  instrumental  value,  may  acquire  value  as  part  of  a  unique 


328  Valuation:   its  Nature  and  Laws 

and  harmonious  totality,  so  acts  such  as  those  described  above, 
in  themselves  also  without  appreciable  value  for  social  participa 
tion,  may  acquire  complementary  value  as  a  part  or  a  sign  of 
a  harmonious  personality,  or  as  a  mark  of  a  coherent  social  class 
or  group. 

IV.  THE  OBJECTIVE  PARTICIPATION  VALUE  OF  DISPOSITIONS 
AS  DEDUCED  FROM  THE  LAWS  OF  SOCIAL  SYMPATHY— THE 
LAWS  OF  SOCIAL  SYNERGY 

i.  Objective  Participation  Value. 

The  preceding  studies  have  had  as  their  object  the  deter 
mination  of  the  laws  governing  the  changes  in  the  individual's 
feelings  of  over-individual  value,  immediate  and  mediate  or 
instrumental,  as  conditioned  by  the  extension  of  the  range 
of  sympathetic  participation.  The  question  now  arises  whether 
from  these  facts  it  is  possible  to  deduce  the  laws  of  objective, 
over-individual  value,  conceived  as  objective  laws  abstracted 
from  the  individual  subjects  of  the  feelings  of  value.  We  have 
maintained  that  the  individual's  feeling  of  value,  in  so  far  as  it 
is  impersonal,  reflects  the  actual  social  value  of  the  dispositions 
he  thus  impersonally  judges.  If  this  is  true,  the  quantitative 
laws  governing  his  feelings  of  over-individual  value  must  corre 
spond  to  similar  laws  of  social  value.  Only  on  this  condition 
would  the  norms  and  standards  of  impersonal  judgment,  pre 
sently  to  be  considered,  correspond  to  actual  social  values. 

The    objective    participation    value    of   a   disposition    is    a 
function  of  two  factors,  the  quantity  of  the   demand   and  the 
quantity  of  the  supply.     Both  of  these  factors  are  determined 
by  the  feelings  and  feeling-dispositions  of  the  individuals  par 
ticipating  in  the  collective  will  which  creates  the  demand  or 
supply.     These  feelings  of  the  individuals  are,  however,  as  we 
have  seen,  modified  in  certain  specific  ways  by  extension  of 
the  range  of  the  sympathetic  feeling ;  and  with  this  modification 
there  is  a  corresponding  change  in  the  individual's  desire  or 
disposition  to  participate.     From  this  it  may  be  easily  seen  that 
there  is  a  direct  relation  between  the  laws  governing  the  indi 
vidual's  feelings  and  feeling-dispositions  and  the  objective  value 
of  those  dispositions  as  determined  by  the  interaction  of  the 
individuals.     More  explicitly  stated,  the  social  demand  and  sup 
ply,  and  the  objective  values  of  which  they  are  the  determining 


Objects  of  Impersonal  Over- Individual  Value      329 

factors,  are  in  some  sense  the  product  of  the  individual's  dis 
position,  considered  as  modifying  the  demand  or  supply,  as  the 
case  may  be.  As  the  intensity  of  the  individual's  feeling  of 
over-individual  value,  expressed  either  in  the  degree  to 
which  he  participates  in  the  social  demand  or  in  acts  supplying 
the  desired  dispositions,  is  determined  by  the  degree  to  which 
the  dispositions  are  universalised,  so  the  intensity  of  the  over- 
individual  demand,  when  conceived  as  abstracted  from  the  in 
dividuals  participating,  must  be  determined  by  the  degree  to 
which  the  disposition  is  universalised,  i.e.,  by  its  quantity. 
Our  problem  is  now  to  state  the  law  of  this  dependence  more 
definitely,  and  to  formulate,  on  the  basis  of  the  principles  govern 
ing  this  interaction,  the  law  of  the  participation  value  of  over- 
individual  objects. 

2.  The  Law  of  Marginal  Participation  Value. 

A  disposition  or  quality  of  an  individual  may,  as  we  have 
seen,  have  a  twofold  value — the  participation  value  which  it 
has  as  the  condition  of  wider  social  intercourse,  i.e.,  as 
contributing  to  common  over-individual  ends,  and  the  per 
sonal  value  which  it  has  as  expressive  of  a  total  personality. 
The  latter  is  complementary  to  the  participation  value.  From 
the  point  of  view  of  the  present  study,  we  are  interested  in  the 
individual's  disposition  merely  as  it  affects  the  quantum  of  the 
social  good.  Participation  through  acts  expressive  of  dispo 
sitions  may  then  be  conceived  as  additions  to  or  subtractions 
from  this  good.  The  supply  may  be  increased  in  two  ways. 
Either  there  may  be  increase  in  the  amount  of  the  disposition 
of  the  individuals  involved,  as  displayed  in  increase  of  the 
energy  of  participation,  or  there  may  be  increase  through  the 
addition  of  individuals  to  the  group.  There  may  be  increase 
in  the  intensity  or  extension  of  the  supply.  Corresponding 
to  this  objective  social  supply  there  is  a  social  over-individual 
demand.  What,  then,  is  the  effect  of  changes  in  the  supply 
brought  about  by  these  two  factors  upon  the  quantity  of  the 
collective  demand?  This  collective,  over-individual  demand 
is,  like  the  supply,  a  variable  quantity,  and  a  quantity  which 
is  determined  by  the  desire  of  the  individuals  concerned  for 
acts  with  participation  worth,  that  is,  for  expressions  which 
may  be  the  inducing  grounds  of  sympathetic  projection.  This 
demand  in  the  individuals  may  be  satisfied  by  increase  either 


33°  'Valuation :    its  Nature  and  Laws 

in  the  intensity  of  sympathetic  participation  or  in  its  extension. 
We  must,  then,  consider  the  effect  of  each  factor  in  the  supply 
upon  the  demand  of  the  individuals. 

It  is  clear,  in  the  first  place,  that,  according  to  our  law  of 
sympathetic  projection,   the  effect  of  expansion  of  the  group 
as  a  factor  in  the  increase  of  supply  is  to  lower  the  intensity 
of  emotional  participation  in  the  individuals,   and,   therefore 
the  intensity  of  the  demand.     The  effect  of  increase  of  intensity 
of  the  supply  is  of  the  same  nature,  but,  the  situation  being 
somewhat    more    complex,    it    requires    closer    analysis.     Let 
us  suppose   that   in   some   manner,   how  it  does  not  matter— 
whether    through     imitation     or    more     conscious     choice— a 
disposition    which    has    participation    value    increases    in    an 
individual   or    a    number    of    individuals  considerably  beyond 
the    average    intensity.      Will    not    such    an    increase    in    the 
supply    bring   with    it,    through    suggestion    and    imitation,    a 
corresponding   increase  in   demand?     In   other  words,  cannot 
such  a  disposition  increase  in  intensity  and  extension  simul 
taneously  ? 

At  first  sight  it  would  appear  so,  but  upon  reflection  doubts 
arise.  For,  according  to  our  analysis  of  sympathetic  projection, 
a  variation  from  the  average  intensity  of  the  disposition  pre 
supposes  a  variation  from  the  normal  in  attitude,  and  the  greater 
this  variation  from  the  normal,  the  smaller  the  group  in  which 
sympathetic  participation  is  possible.  Clearly,  then,  additions 
to  the  supply,  through  increase  above  the  normal  of  the  amount 
of  disposition  in  individuals,  involves  new  group  segregations. 
The  demand  would  be  increased  in  individuals  and  groups, 
but  such  increase  in  intensity  above  the  normal  would  be  secured 
only  by  the  limitation  of  the  extensity  of  the  demand. 

From  these  facts  it  would  appear  that  we  are  justified  in 
inferring  that  the  over-individual  value  of  these  objects  is  gov 
erned  by  a  law  analogous  to  the  law  of  Marginal  Utility  in  the 
sphere  of  economic  values.  Ehrenfels  has  formulated  the  same 
law,  and  developed  it  in  a  somewhat  similar  fashion.1  He 
characterises  it  as  the  law  of  Grenz-frommen,  to  distinguish 

1  Ehrenfels,  System  der  Werttheorie,  Book  II,  chap,  in,  especially  p.  86. 

When  we  examine  more  closely  the  nature  of  these  over-individual  objects  or  goods, 
Lhe  participation  value  of  which  determines  their  morality,  we  see  that  all  the  con 
ditions  necessary  for  the  application  of  such  a  law  exist.  For  the  application  of  the 
economic  law  of  Marginal  Utility  to  any  object  of  condition  worth  it  is  necessary:  (i) 
that  the  object  be  limited  in  amount ;  (2)  that  it  have  capacity  of  substitution ;  and 
(3)  that  the  desire  corresponding  to  the  object  have  the  tendency  to  highest  possible 
activity.  An  examination  of  the  processes  of  sympathetic  projection  which  create 


Objects  of  Impersonal  Over- Individual  Value      331 

it  from  the  law  of  Grenz-nutzen  in  the  sphere  of  economic  goods. 
But  since  we  have  denned  social  good,  of  which  the  moral  sense 
is  the  reflection,  wholly  in  functional  terms  of  participation, 
we  may  preferably  speak  of  the  law  of  Marginal  Participation 
Value.  This  law,  it  is  further  apparent,  is  but  the  resultant, 
in  the  sphere  of  objective  social  values,  of  the  law  of  Diminish 
ing  or  Limiting  Value  operative  in  the  sympathetic  participation 
of  individuals. 


3.  The  Laws  of  Social  Synergy. 

If  the  line  of  reasoning  which  has  led  to  the  formulation 
of  this  law  is  sound,  we  are  justified  in  concluding — thus  far 
upon  merely  theoretical  grounds — that  the  individual's  feeling 
of  over-individual  value,  and  his  judgments  of  value,  precisely 
in  so  far  as  they  are  impersonal  and  reflect  this  feeling,  correspond 
directly  to  the  objective  social  participation  value  of  the  dis 
positions  judged.  Our  thesis  has  been  that  moral  worth  judg 
ments,  in  their  aspect  both  of  obligation  and  imputation,  are 
impersonal  judgments  upon  acts,  and  reflect  the  participation 
value  of  these  acts  for  over-individual  ends.  An  analysis  of 
these  judgments,  to  which  we  shall  presently  turn,  will  show 
this  correspondence.  It  will  become  evident  that  the  conscious 
ness  of  over-individual  value  presupposed  by  these  judgments, 
reflects  the  working  of  this  law,  i.e.,  that  the  norms  and  limits 
presupposed  by  these  judgments  reflect  this  law  of  participa 
tion  value.  In  the  meantime,  and  as  a  preliminary  to  this 
analysis,  we  may  here  develop  certain  important  consequences 
of  the  law  significant  for  our  later  studies.  These  consequences 
are  twofold,  and  may  be  properly  described  as  corollaries  from 
the  law  of  Marginal  Participation  Value.  This  law,  together 
with  its  corollaries,  may,  then,  be  described  as  the  laws  of  Social 
Synergy,  for  they  are  the  laws  governing  those  social  processes 
of  sympathetic  interaction  and  valuation  which  have  been 
characterised  as  social  synergy. 

these  over-individual  objects  shows  that  all  these  conditions  are  in  force.  The  desire  for 
the  object  springs  out  of  the  tendency  to  participation,  and  this  tendency,  in  its  primitive 
organic  form,  seeks  the  highest  possible  activity  both  in  intensity  and  extension.  The 
amount  of  the  disposition  is  always  limited,  both  in  extension  and  intensity,  by  the 
conditions  of  sympathetic  projection;  and, 'finally,  as  we  have  seen,  while  dispositions 
in  their  personal  or  limited  group  reference  are,  under  certain  special  circumstances, 
without  capacity  of  substitution,  no  specific  disposition  is  as  such  ultimately  indispensable 
for  social  participation. 


Valuation  :   its  Nature  and  Laws 

(a)  Corollaries  from  the  Law  of  Marginal  Participation  Value. 
In  the  first  place  the  working  of  this  law  results  in  the 


the  minimu      and  maximum  demanded  of  him  will 


* 


<* 

In  the  second  place,  we  shall  find  that  the  inevitable  con 
sequence  of  this  law  of  social  synergy  is  social  differe™n  and' 
group  segregat.cn.     The  tendency  to  complete  unive  saUsatio 

tlZcTT'ft0  h  JS  ™Tnt  in  "very  "da!  di^osufon  °or 

lency     is    followed    by    loss    of   participation    value      This 

arrest  of  social  demand  is  followed   by  value  mlvement   to 


,  ethical  or  esthetic.     In  the  former  it  is  through  detach- 
'01  i"divMuation  of  «-  object  or  /erson  that 


-  on 

ueH        S  °rSamsatlon  and  unification  about  an  ideal  or 

Irte  newelT        r"1  Other  ^"P5'  Thr°USh  these  cont^ 
rise  new  complementary  group  values,  which  are  imputed  over 

and  above  the  actual  value  which  the  object  would  haveL 
virtue  merely  of  the  functioning  of  the  primary  law  of  soda" 
synergy,  and  which  therefore  modify  the  norml  and  L  ts  of 

™  e  t°wTnh  JUdgTent'     °nly  ^  the  Consideration  of 

phenomena  m  their  inter-relations  shall  we  be  able 

C°nCrete   S0dal     UdmentS  °f 


«..»      j  «j.<j.cmcjiiLS    Of 

imputation. 


Objects  of  Impersonal  Over-Individual  Value      333 

(6)  Norms  and  Limits  of  Participation  Value — Social  Value 

Movements. 

The  first  consequence  of  the  application  of  this  law  to 
social  values  is  that  it  enables  us  to  distribute  conceptually 
these  social  worths  about  three  critical  points.  The  working 
of  this  law,  the  principle  of  Grenz-frommen,  as  Ehrenfels  calls 
it  when  applied  to  social  worth  dispositions,  results  in  a  mutation 
of  value  in  which  three  phenomenal  phases  may  be  distinguished, 
as  aspiring,  normal,  and  outlived  values  (Ehrenfels's 1  aufstrebende, 
normale,  und  nberlebte  Werthe).  These  we  may  describe  in  the 
following  way.  An  aspiring  social  value  is  one  the  intensity  of 
demand  for  which  in  a  given  social  group  is  great,  corresponding 
to  a  limited  expansion  or  diffusion  in  the  social  consciousness. 
A  normal  value  may  be  described  as  one  in  the  case  of  which 
the  intensity  of  the  demand  and  the  extent  of  its  diffusion  are 
more  nearly  equal.  In  the  outlived  value  the  diffusion  has 
become  so  great  that  decrease  in  intensity,  and  finally  loss  of 
value  as  it  approaches  universality,  follow.  We  have  here  the 
stages  of  a  social  value  movement,  from  social  passion  and 
emotion  to  habit  and  indifference,  analogous  to  the  similar 
stages  in  the  instrumental  valuation  of  objects  by  individuals. 

This  law  of  the  mutation  of  social  values  is  most  apparent 
in  the  superficial  changes  in  fashions.  In  the  case  of  fashions 
in  clothes  and  manners,  the  value  is  chiefly  found  in  the  functional 
significance  of  the  object  for  social  participation  and  contrast, 
and  the  value  movements  are  correspondingly  rapid  and  super 
ficial.  The  period  of  intensification  of  personal  and  group 
values  through  novelty  and  contrast  is  quickly  followed  by 
imitation  and  universalisation  until,  when  the  objects  become 
"  common,"  their  value,  which  has  been  conditioned  largely  by 
temporary  demand,  expires  and  is  finally  outlived — and  new  forms 
of  distinction  and  contrast  become  necessary.  The  working  of 
the  law  is  seen  here  at  its  purest  because  the  greater  part  of  the 
demand  for  the  object  or  act  is  an  expression  of  their  value  for 
immediate  social  participation,  the  more  indirect  value  they  may 
have  for  specific  individual  and  social  ends  being  relatively  little. 
This  is  seen  in  the  fact  that  when  the  mode  is  past  the  greater 
part  of  their  social  esteem  and  economic  value  or  price  has 
gone.  It  is  the  paradox  of  fashion  that  when  it  is  most 
prevalent,  when  the  demand  is  at  its  height,  its  decay  has 

1  Ibid.,  Book  II,  chap.  Ill,  par.  17. 


334  Valuation:   its  Nature  and  Laws 

already  set  in.  Up  to  that  point  it  is  the  extension  that  stimu 
lates  imitation  and  creates  its  value ;  when  that  point  is  reached 
it  is  this  same  extension  which  destroys  its  value  and  leads  to  its 
wholesale  abandonment.  It  is  just  because  of  this  emptiness  of 
all  meanings  except  the  commonness  of  superficial  contagion 
that  the  pursuit  of  fashion  is  so  essentially  self-defeating. 

When  we  turn  to  the  more  important  region  of  the  socially 
desirable  dispositions  and  common  ideals  of  men,  we  find  that, 
while  they  are  more  permanent  conditions  of  social  participa 
tion—the  ends  which  they  serve  being  more  ultimate  and  perma 
nent—nevertheless,  they  are  subject  to  the  same  law.  The  value 
movement,  it  is  true,  is  extended  over  a  longer  time,  and  is, 
therefore,  more  difficult  to  detect.  Moreover,  the  social  par 
ticipation  value  is  much  more  complicated  with  fundamental 
personal  worths,  and  the  dependence  of  its  intensity  upon  its 
social  expansion  is  not  so  clearly  marked.  Nevertheless,  analysis 
enables  us  to  distinguish  the  two  elements. 

The    most    patent    illustrations    of    these  value  movements 
are   to  be   found  in   the  sphere  of  social  and  political  ideals 
and     shibboleths,    although    similar    laws     govern     the    more 
ultimate    moral   ideals    which    underlie    them.     The    ideals    of 
"  liberty,  fraternity,  equality  "  of  the  French  Revolution  may 
be  said  to  have  gone  through  these  phases.     In  contrast  to  the 
opposing  ideals  of  the  old  regime  they  were  first  aspiring,  and 
participation  in  them  was  accompanied  by  intense  personal  and 
group  worths.    The  normal  stage  was  reached  when  they  became 
the    presupposition    of   judgments    of   morals    and    legislation. 
They   had   then   become   diffused   enough  to  find  institutional 
expression,  and  were  still  felt  intensely  enough  to  make  them  a 
moral  force.     At  this  point,  the  specific  dispositions  for  which 
these  ideals  with  their  suffused  emotion  stood  (the  disposition 
to  acknowledge  manhood  suffrage,  etc.)  having  secured  practi 
cally  universal  assent,  they  are  taken  for  granted.     Social  habit 
appears,   and   the   usefulness   of  the  ideals  for  social   partici 
pation  decreases.     They  begin  to  be  outlived,  and  their  demands 
are  not  strongly  felt.     The  old  symbols  become  empty  words, 
lacking  the  social  emotion  of  the  aspiring  value  or  the  social 
sentiment  which  accompanies  habit.     They  do  not  correspond 
to  actual  values,  and  therefore  lack  real  obligatory  force.     It 
may,  of  course,  be  said  that  the  spirit  of  these  ideals  is  far  from 
realised,  and  that  they  will  constantly  take  on  new  forms.   This  is 
true,  but  each  new  form  means  the  differentiation  of  a  new  specific 


Objects  of  Impersonal  Over- Individual  Value      335 

disposition  and  a  new  concrete  ideal,  and  the  intensity  with  which 
that  ideal  is  felt  will  again  depend  upon  the  same  empirical  con 
ditions,  the  degree  of  expansion  which  the  ideal  has  attained. 

All  this  may  be  easily  misunderstood — may  be  taken  to  mean 
that  the  ideals  which  have  undergone  this  transmutation  into 
"  things  taken  for  granted,"  or  social  habits,  have  lost  all  relation 
to  actual  social  values.  This  is  far  from  the  truth.  They  have 
become  the  platform  of  implicit  assumptions  upon  which  new 
values  may  be  built  up.  Emerson's  epigram  "that  culture  is 
the  measure  of  things  taken  for  granted"  may  well  be  applied 
here ;  we  may  say  that  civilisation  is  the  measure  of  socially 
desirable  dispositions  or  habits  taken  for  granted.  But  when 
all  this  is  admitted,  it  is  still  true  that  on  the  platform  of  things 
taken  for  granted  new  ideals  suffused  with  emotion  must  arise  if 
conscious  social  values  are  to  continue.  A  potential  value  can 
become  actual  and  dynamic  only  by  becoming  a  felt  value.  In 
order  to  continue,  value  must  forever  be  taking  on  new  forms. 

This  brings  us  finally  to  the  question  of  the  possible  existence 
of  certain  social  ideals  and  dispositions,  underlying  the  more 
superficial  values,  which  are  said  to  have  absolute  value  and 
to  escape  the  social  value  movement  here  described.  In  the 
preceding  case  it  was  said  "  the  spirit  remains,  even  if  the  form 
changes,"  and  it  is  not  only  a  popular  assumption,  but  also  in 
many  quarters  a  philosophical  postulate,  that  there  are  certain 
fundamental  acts  and  innermost  dispositions  to  participate 
which  have  absolute  and  unconditioned  value.  But  when 
we  examine  the  acts  or  dispositions  in  question,  we  discover 
that  in  every  case  they  have  acquired  this  unique  sanctity 
only  by  abstraction  from  concrete  reality.  The  assumption 
of  absoluteness  has  had  its  ground  either  in  a  narrowness  of 
historical  and  social  perspective — in  which  case  the  feeling 
of  absoluteness  has  attached  itself  to  a  concrete  act  or  disposi 
tion  which  is  in  reality  neither  universal  nor  eternal,  or  else, 
perhaps,  in  a  process  of  logical  abstraction— in  which  case  the 
object  is  so  abstract  that  no  actual  felt  value  whatever  is  at 
tached  to  it,  and  the  postulate  of  universality  and  eternity 
cannot  be  challenged.  This  illusion  is  furthered  by  the 
fact  that  different  dispositions  pass  through  these  phases 
of  value  movement  in  widely  varying  periods  of  time,  and 
that  some  dispositions  remain  in  the  normal  phase  through 
long  periods,  so  that  it  is  difficult  to  determine  whether 
there  has  been  any  appreciable  movement  toward  the  loss  of 


336  Valuation  :   its  Nature  and  Laws 

intensity  which  characterises  the  outlived  value.  Moreover, 
actual  changes  in  worth  dispositions  may  be  perfectly  consonant 
with  long-continued  retention  of  the  old  name. 

Such  a  conclusion  evidently  has  an  important  bearing  upon 
our  conception  of  moral  values.  Underlying  the  normative 
law  of  "  universalisation,"  whatever  its  form,  whether  utili 
tarian  or  idealistic,  whether  in  the  Kantian  form  or  in  Fichte's 
modification  ("  act  so  that  the  maxim  of  thy  conduct  may  be 
come  for  thee  an  eternal  law"),  is  the  assumption  that  moral 
values  escape  the  laws  inherent  in  the  temporal  character  of  values. 
But  the  difficulties  in  such  a  conception  are  not  to  be  hidden. 
Universalisation  of  a  given  disposition,  or  even  indefinite  increase 
of  the  supply,  must  involve  such  a  modification  of  the  demand  as  in 
turn  to  change  the  actual  objective  social  value  of  the  disposition. 
To  act  as  though  the  maxim  of  one's  act  were  an  eternal  law  is  to 
act  as  though  frequency  of  repetition  would  have  no  effect  upon 
its  value,  an  assumption  which  experience  does  not  allow  us  to 
make  with  respect  to  the  objective  values  of  any  object.  What 
ever  validity  such  an  assumption  may  have  for  the  worth  experi 
ence  of  the  individual,  it  cannot  be  taken  as  a  norm  or  measure 
of  actual  social  values. 

With  the  postulate  of  absolute  value  in  its  logical  or  axio- 
logical   aspect   we   are   not   here   concerned.     That    "  value   is 
eternal,"  in  the  sense  that  continuity  of  valuation  is  presupposed 
in  every  value  judgment,   is  the  necessary  postulate  of   every 
judgment  of  value;  that  it  is  universal  in  the  sense  that  my 
value  judgment  is  in  some  way  continuous  with  the  judgments 
of  all  other  subjects  of  valuation  is  an  equally  necessary  postulate. 
But  to  infer  from  this  that  the  participation  value  of  any  con 
crete  disposition  or   act   is   either   eternal  or   universal   is   un 
warranted.     Such  an  inference  would  be  justified  only  in  case 
we  could  show  that  for  the  instrumental  participation  value  of 
certain  dispositions  there  are  no  substitutes.     But  beyond  the 
idea  that  whatever  possesses  actual  social  value  must  persist  in 
one  form   or   another,   we   cannot   pass;    and  of   these   future 
forms  we  cannot  form  any  definite  ideas,  for  experience  shows 
us   a  continual   readjustment    according   to   the   laws    already 
described.     That  many  of  these  social  values  may,  as  ideals, 
and  therefore  as  centres  of   organisation   for   individuals  and 
groups,  acquire  the  meaning  of  practical  absolutes  is  a  possi 
bility  we  have  already  suggested,  and  will  consider  more  fully 
presently,  but  as  social  values  they  are  always  relative. 


Objects  of  Impersonal  Over- Individual  Value      337 

(c)  Social  Values  as  Reflected  in  the  Individual. 

Assuming  the  truth  of  this  schematic  picture  of  the  mutation 
of  social  values,  of  the  value  movement  that  every  social  value 
undergoes,  and  therefore  also  of  the  distribution  of  values  in  any 
given  cross-section  of  the  social  consciousness,  how  does  this 
situation  reflect  itself  in  the  individual's  consciousness  of  over- 
individual  values  ?  It  is  clear  that  values  in  all  three  stages 
will  be  represented.  Those  dispositions  which  belong  to  the 
class,  normal  values,  will  constitute  a  central  region  which 
we  may  describe  as  "  moral "  worth  dispositions.  They 
represent  a  degree  of  constancy  of  habit  in  worth  judg 
ment  which  other  dispositions  cannot  attain.  This  constancy 
has  two  aspects.  On  the  one  hand,  these  moral  worths 
represent  the  most  completely  universalised  dispositions,  and 
consequently  the  expectation,  or  demand  for  participation 
in  them  is  relatively  universal.  On  the  other  hand,  the  intensity 
of  participation  expected  is  relatively  constant,  and  represents 
the  norm  of  expectation  in  judgments  of  obligation  and  im 
putation.  As  a  result  of  this  element  of  constancy  in  the  normal 
moral  disposition,  the  judgments  which  spring  out  of  this  norm 
approach  most  closely  to  the  impartial  impersonal  ideal  which 
constitutes  the  standpoint  of  morality. 

About  this  region  of  "moral"  values  as  a  centre,  gather  the 
aspiring  and  outlived  values  which,  in  contrast,  may  be  described 
as  ethical  or  quasi-moral  worths.  They  are  so  named  because, 
while  social  and  over-individual  in  their  reference,  they  lack  the 
impersonal  reference  of  the  impartial  spectator.  They  are  limited 
group  values  which  shade  over  into  personal  worths. 

The  aspiring  values  are  supra-normal  worths  in  that  they 
represent  great  intensity  with  limited  expansion.  To  take  as 
an  illustration  the  new  values  taught  by  Christianity  in  its 
early  days,  within  the  Christian  fellowship  itself  the  de 
mand  for  manifestation  of  its  virtues  exceeded  the  normal, 
while  it  was  clearly  felt  that  a  different  standard  must  be 
recognised  for  those  outside.  In  the  case  of  such  ideals  the 
expectation  is  not  "  standardised  "  ;  individual  variation  is  more 
marked.  But  for  this  very  reason  the  intensity  with  which 
the  individual  participates  is  also  greater  than  in  the  case  of 
normal  values — in  direct  proportion  to  the  absence  of  the 
virtues  in  others.  Moreover,  the  value  imputed  to  the  indi 
vidual  thus  participating  is  proportionally  greater. 


338  Valuation:    its  Nature  and  Laws 

The   outlived  social  value,  on  the  other  hand,  is  below  the 
normal,  in  that  expansion  so  outruns  intensity  that  participa 
tion  is  no  longer  emotional  but  purely  intellectual  and  formal, 
in  so  far  as  the  mass  of  individuals  is  concerned.     The  social 
demand,  as  reflected  in  the  individual,  is,  therefore,  normally 
much  less  intense  than  in  the  case  of  the  other  classes  of  worth 
dispositions,  the  aspiring  and  the  normal.    Thus  there  are  certain 
conventional  acts  and  standards,  such,  for  instance,  as  church^ 
going  in  some  communities  and  certain  forms  of  charity,  which 
were  at  one  time  motived  by  a  vital  and  immediately  felt  dis 
position  to  participate  in  common  worship  or  in  the  common 
good,  but  which  are  now  kept  going  merely  by  an  intellectual 
recognition  of  their  formal  instrumental  value  for  social  solidarity. 
A  good  test  of  an  outlived  value  is  just  this  conscious  effort  to 
keep  it  up  when  the  real  disposition  for  which  it  stood  has  passed 
away  or  has  been  transferred  to  some  other  act.     The  loss  of 
belief  in  these  forms  of  piety  and  charity  does  not  necessarily 
mean   weakening   of   the   energy   or  will  to  participate   which 
underlies  them.     They  simply  draw  in,  and  contract  to  more 
ideal  and  personal  forms,  awaiting  perhaps  the  time  of  a  new 
embodiment  in  a  social  ideal  which  shall  again  caU  forth  feelings 
of  reality. 

An  outlived  value  makes  itself  felt,  therefore,  in  this  way. 
As  a  positive  ideal  it  is  still  upheld,  and  is  often  the  more 
eloquently  preached  the  more  its  vitality  wanes,  but  its  non- 
observance  is  less  and  less  noted  and  disapproved  until  finally 
it  is  entirely  detached  from  the  direct  relation  to  feeling  and  will 
which  formerly  gave  it  life.  In  the  consciousness  of  every  indi 
vidual  there  will,  therefore,  be  certain  faint  social  obligations 
which  reflect  outlived  values. 

In  general,  then,  as  has  already  been  stated,  the  individual's 
consciousness  of  over-individual  values  will  reflect  the  actual 
social  values  of  these  objects,  and  his  disposition  to  participate 
and  to  demand  participation  on  the  part  of  others  will  be  modified 
in  the  ways  described.  But  we  should  not  overlook  the  fact 
that,  as  in  the  case  of  the  aspiring  values,  so  in  the  outlived, 
the  working  of  this  general  law  of  value  movement  may  be 
modified.  The  outlived  value  may  become  the  ideal  or  the  centre 
of  intrinsic  values  for  individuals  and  groups,  and  may  acquire 
a  new  participation  value  through  contrast  and  opposition, 
as  in  the  case  of  reactionary  and  radical  groups.  An  outlived 
social  value,  such  as  class  ideals  of  honour  and  bravery  belonging 


Objects  of  Impersonal  Over- Individual  Value       339 

to  a  military  type  of  civilisation,  may  survive  in  an  industrial 
regime,  the  monastic  ideals  of  humility  and  contemplation  may 
survive  in  a  time  when  they  correspond  to  no  widespread  social 
demand,  and  through  this  very  contrast  and  opposition  they 
may  become  the  objects  of  intense  loyalty  and  of  emotional 
participation.  For  such  reactionary  individuals  and  groups 
they  again  become  aspiring  values.  The  worth  thus  acquired 
is,  however,  an  ethical,  personal,  or  group  worth,  and  the  de 
mand  is  no  longer  impersonal  and  over-individual.  To  the 
radical,  on  the  other  hand,  these  same  socially  outlived  values 
become  the  stimulus  for  the  development  of  new  ideals  and 
values,  and  in  seeking  freedom  from  the  old  conventional  senti 
ments  and  ideals  he  thereby  develops  the  new.  This  leads  us 
to  the  phenomenon  of  social  differentiation  and  its  bearing  upon 
the  development  of  values. 

(d)  Social  Differentiation — Group  Segregation. 

The  second  consequence  of  the  law  of  Marginal  Participation 
Value,  social  differentiation,  is  closely  connected  with  the 
phenomena  of  social  value  movements  which  we  have  been 
considering.  The  tendency  to  complete  universalisation  and 
expansion  which  is  inherent  in  social  dispositions  is  followed 
by  loss  of  participation  value  and  by  value  movements  to  new 
objects.  In  these  movements  appear  new  group  formations 
and  contrasts,  with  acquirement  of  complementary  value  through 
these  contrasts.  In  the  sphere  of  objective  social  values  also 
we  find,  within  certain  limits,  a  modification  of  the  law  of  Mar 
ginal  Participation  Value  similar  to  the  modification  of  instru 
mental  values  in  other  spheres, — through  acquirement  of  intrinsic 
complementary  value.1 

Social  differentiation,  group  segregation,  consists  in  the 
unification  and  organisation  of  a  group  about  some  ideal  end. 
It  is  the  product  of  two  factors,  the  tendency  to  complete  ex 
pansion  and  homogeneity  inherent  in  imitation  and  sympathetic 
participation,  and  the  tendency  to  individuation  and  intensi 
fication  of  sympathetic  feeling  through  contrast.  The  first 
factor  produces  social  habit.  Aspiring  values  become  normal 
and  normal  values  become  outlived.  But  this  process  is  held 
in  check  by  individuation  of  the  group  through  idealisation 
and  contrast.  The  processes  here  involved  are  in  principle  the 

1  Chap,  vi,  pp.  171  ff. 


340  Valuation :  its  Nature  and  Laws 

same  as  those  employed  in  the  idealisation  of  the  person.  The 
disposition  in  question  is  assumed  to  be  universalised  within  a 
group,  a  class,  a  race,  a  party,  etc.,  and  this  homogeneity  is 
deepened  through  contrast  with  opposing  groups.  Such  ideal 
construction  may  go  so  far  as  to  individuate  the  opposing  groups 
about  the  negative,  evil  dispositions  and  ends,  in  which  case, 
by  a  well-known  contrast-effect,  the  negative  elements  are  ex 
truded,  and  the  unity  of  the  group  idealised.  We  have  already 
seen  how  this  individuation  and  isolation  of  the  group  is  to  a 
degree  possible  in  organic  sympathy.  For  short  periods,  and 
in  connection  with  fundamental  desires  and  passions,  the  soli 
darity  of  the  group  may  be  complete,  and  the  sympathy  may 
be  intensified  by  group  contrast.  But  such  isolation  is  only 
temporary,  and  the  presumption  of  homogeneity  is  likely  soon 
to  lose  its  force  and  to  prove  illusory.  More  important  in 
creating  permanent  assumptions  of  group  solidarity  are  the 
idealisations  of  the  group  through  aesthetic  and  religious  con 
structions,  which  we  shall  presently  study  in  detail.  Through 
these  ideal  constructions  groups  are  organised  about  the  aspiring 
and  outlived  values,  and  these  objects  acquire  an  intrinsic 
meaning  for  the  individual  group  which  is  quite  apart  from  their 
objective  instrumental  value. 

This  is  but  a  special  application  of  the  general  law  according 
to  which  instrumental  values,  on  becoming  intrinsic,  acquire 
complementary  value.  As  the  law  of  Marginal  Utility  for  objects 
of  instrumental  value,  more  especially  for  wealth  in  general, 
was  seen  to  be  modified  in  certain  significant  ways,  i.e.,  a 
quantity  of  wealth  acquires  a  complementary  intrinsic  value, 
not  found  in  the  instrumental  value  of  the  parts,  so  we  find 
a  similar  modification  of  the  law  of  Marginal  Participation  Value. 
The  valuation  of  a  sum  of  money  as  a  whole,  where  the  separate 
instrumental  judgments  are  suppressed,  where  its  indefinite 
applicability  to  condition  and  personal  worths  is  assumed, 
and  where  it  is  referred  immediately  to  the  personality,  gives  to 
the  sum  of  money,  as  a  unity,  an  intrinsic  value  which  may 
greatly  exceed  its  actual  instrumental  value.  In  a  similar 
way,  when  an  over-individual  good,  a  social  disposition  with 
participation  value,  is  identified  with  a  group,  and  assumed  to 
be  universalised  in  that  group,  it  acquires  an  intrinsic  worth 
which  goes  beyond  the  limits  of  instrumental  valuation.  Emo 
tional  participation  is  extended  beyond  the  limits  set  by  the 
working  of  the  law  of  instrumental  judgments.  Doubtless 


Objects  of  Impersonal  Over- Individual  Value      341 

the  specialised  dispositions  and  ideals  of  special  groups  and 
classes  all  have  had,  or  still  have,  certain  instrumental  values 
for  society  at  large.  The  specialised  dispositions  of  military 
classes,  the  ideals  of  free  expression  of  the  artist  class,  the  ideals 
of  contemplation  and  renunciation  of  religious  orders,  all  have 
instrumental  values  for  society.  But  in  every  case  their  value 
is  conditional  upon  their  not  being  universalised. 

There  can  be  no  doubt  that  the  condition  of  the  highest 
realisation  of  participation  value  is  group  segregation.  How  far 
this  individuation  or  isolation  of  a  group  may  go  in  the  direction 
of  creating  absolute  permanent  worths,  and  what  are  the  limits 
of  the  process,  we  shall  have  occasion  to  consider,  but  that 
it  is  a  factor  to  be  reckoned  with  in  accounting  for  the  phenomena 
of  actual  social  values,  and  the  individual's  participation  in  them, 
is  clear.  For,  not  only  does  it  come  in  to  modify,  at  least  tem 
porarily,  the  inevitable  tendency  of  all  social  values  to  pass  into 
social  habit  where  emotional  participation  ceases,  and  thus  to 
maintain  in  the  form  of  group  worths  the  class  described  as 
outlived,  longer  than  would  otherwise  be  possible,  but  it  also, 
by  its  intensification  of  aspiring  worths,  enhances  their  value. 
Moreover,  it  affords  the  conditions  for  the  intensification  of 
personal  worths,  or,  from  a  more  objective  point  of  view,  for 
the  development  of  great  personalities. 

If  we  may  make  use  of  a  biological  analogy,  with  a  distinct 
consciousness  of  its  purely  suggestive  value,  this  group  isola 
tion  may  be  compared  to  isolation  in  the  sphere  of  natural 
selection.  There  it  is  generally  recognised  that,  in  order  that 
a  new  variation  may  be  fixed  in  the  species,  isolation  of  the 
species,  either  through  environmental  conditions  or  the  im 
possibility  of  breeding  with  other  species,  must  be  assumed. 
Otherwise  the  variation  would  soon  be  lost  again  through  pan 
mixia.  In  a  somewhat  similar  manner,  in  order  that  there  may 
be  continuance  of  intrinsic  valuation  of  definite  and  fixed  atti 
tudes,  isolation  of  groups  is  necessary,  or  at  least  contributory. 
Here  we  have  the  springs  of  class  jealousy,  once  group  segrega 
tion  has  taken  place.  Social  pan-mixia,  the  breaking  down  of 
class  barriers,  makes  impossible  that  fixity  and  contrast  of  ideals 
which  constitute  the  condition  of  many  personal  and  group  values. 
Such  class  jealousy  often  connects  itself  with  the  preservation  and 
isolation  of  ideals  and  standards  which,  from  the  point  of  view  of 
wider,  more  impersonal  judgment,  seem  fictitious,  as,  for  instance, 
in  the  case  of  the  seemingly  fine  spun  and  arbitrary  notions  of 


342  Valuation  :   its  Nature  and  Laws 

honour  which  characterise  certain  classes.  They  are  indeed 
fictitious,  but  only  in  the  sense  that  many  personal  values  are,  in 
that  they  are  ideal  constructions  made  in  the  interest  of  con 
tinuity  of  valuation.  As  the  primitive  tribesman  finds  himself 
in  the  arbitrary  and  conventional  tribal  marks  which  adorn  his 
brother's  person,  so  individuals  of  a  later  civilisation  find  in 
intrinsic  and  conventional  attitudes  those  contrasts  and  oppo 
sitions  which  deepen,  if  they  do  not  broaden,  the  feeling  of 
participation. 

It  is  not  necessary  to  insist  at  any  length  upon  the  fact  that 
we  are  not  here  concerned  in  the  least  with  any  evaluation  of 
these  phenomena,  with  the  determination  of  the  teleology  or 
dys-teleology  of  social  differentiation.  Whether  salvation  consists, 
as  with  Tolstoi,  in  wiping  out  these  fictitious  values  and  getting 
back  to  those  which  are  most  universal  and  primitive,  however 
that  may  be  conceived  to  be  possible,  or  as  with  Nietzsche,  in 
affirming  still  more  distinctly  these  same  segregating  attitudes,  or 
as  with  Guyau,  and  in  a  more  scientific  way  with  Simmel,  in 
bringing  about  an  ordered  continuum  of  social  values  in  which 
the  individual  can  easily  pass  from  one  to  the  other,  is  a  problem 
of  social  philosophy  and  practice.  Our  interest  is  merely  in  the 
psychological  processes  involved  in  social  differentiation.1 

4.  The  Limits  of  Participation  Value  of  Dispositions — The 
Question  of  Absolute  Social  Values. 

The  consideration  of  these  laws  of  social  synergy— the  law 
of  Marginal  Participation  Value,  and  its  two  corollaries  with 
reference  to  value  movement  and  group  formation  and  differ 
entiation — would  seem  to  lead  to  the  conclusion  that  the 
social  over-individual  value  of  qualities  and  dispositions  of 
persons,  no  less  than  of  economic  goods,  is  always  relative, 
the  degree  of  value  being  relative  to  the  quantity  of  the  dis- 

1  The  phenomena  of  social  differentiation  have  ordinarily  been  studied  from  the 
objective  point  of  view,  as  an  extra-psychic,  social  fact,  part  of  the  order  of 
nature.  Viewed  in  this  way,  apart  from  the  value  consciousness  of  the  individuals 
involved,  it  has  been  studied  in  two  aspects  :  (a)  causally  and  genetically,  as  the 
product  of  sub-conscious  forces  of  selection,  defined  now  as  economic,  now  as 
biological,  now  as  a  combination  of  the  two, — as  a  product,  in  other  words,  of  a 
specification  of  functions  ;  or  (b)  logically,  or,  perhaps  better,  ideologically,  in  that 
the  thinker  seeks  to  rationalise,  and  sanction,  social  segregation  by  showing  that 
such  differentiation  is  instrumental  to  the  realisation  of  some  abstractly  defined  good- 
according  to  one  or  other  of  the  hypotheses  most  in  vogue,  the  idealistic  or  the 
hedonistic,  either  the  fullest  ideal  life  or  a  maximum  of  pleasure  or  utility.  With 
neither  of  these  aspects  of  the  problem  are  we  primarily  concerned.  Both  types  of 
explanation  are,  strictly  speaking,  extra-psychological. 


Objects  of  Impersonal  Over- Individual  Value      343 

position  or  the  degree  of  its  universalisation.  This,  if  true,  would 
exclude  the  possibility  that  any  disposition,  no  matter  how  funda 
mental,  should  acquire  absolute  social  value.  The  very  fact 
that,  as  social,  the  value  is  instrumental  and  not  intrinsic  would 
necessitate  this  consequence.  From  this  it  would  follow  that 
judgments  of  value  of  the  individual,  in  so  far  as  they  are  im 
personal  and  moral  and  reflect  actual  over-individual  value, 
would  be  always  relative  and'  never  absolute,  as  in  the  case  of 
personal  worths. 

Nevertheless,  while  this  conclusion  seems  to  be  inevitable — 
and  we  shall  find  it  substantiated  by  an  analysis  of  the  concrete 
phenomena  of  impersonal  moral  judgments  of  obligation  and 
imputation,  we  cannot  accept  the  conclusion  without  further 
analysis.  It  is  at  least  conceivable  that,  as  in  the  case  of  personal 
worths,  so  here,  a  disposition  having  social  value  may  become 
intrinsically  valued,  and  thus  acquire  absolute  complementary 
value.  If  it  is  impossible  that  a  social  good  should  attain  abso 
lute  participation  value,  and  that  the  moral  obligation  to  that  good 
should  be  unconditional,  it  is  still  conceivable  that  the  abstract 
moral  point  of  view  might  be  so  modified  by  other  activities  of 
valuation,  as,  for  instance,  aesthetic  and  religious,  that  the 
instrumental  value  might  become  intrinsic  and  acquire  absolute 
complementary  value.  In  the  sphere  of  personal  values  the 
ethical  reaches  an  absolute  moment  through  aesthetic  characteri 
sation.  There,  we  found,  in  the  very  processes  of  characterisa 
tion  are  contained  the  necessary  presuppositions  of  absolute 
personal  worths,  the  aesthetic  isolation  of  the  individual,  the 
suppression  of  instrumental  judgments  and  repose  in  the  object. 
The  hypothesis  of  absolute  personal  worths  was  then  found 
substantiated  in  the  actual  judgments  of  personal  obligation 
and  merit.  May  it  not  be  that  in  the  processes  of  social  partici 
pation  and  ideal  construction  there  are  similar  activities  creating 
absolute  values  ? 

In  the  foregoing  study  of  social  sympathy  and  its  consequent 
ideal  constructions,  we  have  already  had  occasion  to  call  atten 
tion  to  the  fact  that,  while  the  normal  law  of  social  sympathy, 
as  determined  by  its  conditions,  is  that  with  the  increase  of  the 
range  of  sympathy  the  degree  of  sympathy  decreases,  neverthe 
less,  this  law  is  modified  by  group  isolation  and  contrast.  Simi 
larly,  it  appeared  that  through  this  group  isolation  and  contrast 
the  conditions  are  realised  which  make  possible,  within  limits  at 
least,  the  individuating  reconstruction  of  the  group  and  the 


344  Valuation :    its  Nature  and  Laws 

acquirement  of  complementary  value.  The  question  for  con 
sideration  now  is,  how  far  this  individuating  construction  may 
go  in  this  direction. 

There  are  two  types  of  social  ideal  construction  or  idealisation 
which  create  complementary  values  and  extend  the  range  of 
social  participation  and  valuation,  the  aesthetic  and  the  religious. 
In  the  beginning  strictly  limited  in  their  scope,  confined  to  the 
function  of  enhancement  of  group  and  racial  values,  the  range 
of  their  constructions  has  expanded  until  some  aesthetic  and 
religious  idealisations  lay  claim  to  absolute  and  universal  validity, 
on  the  one  hand  as  expressing  the  purely  and  simply  "  human,"  on 
the  other  as  having  reached  the  divine.  How  far  these  construc 
tions  are  able  to  create  absolute  social  values  we  must  now  decide. 

Taking  up  the  problem  from  the  side  of  aesthetic  idealisation, 
there  can  be  do  doubt,  it  would  seem,  that  aesthetic  feeling 
does  extend  immensely  the  capacity  for  social  participation — 
both  in  intensity  and  expansion.  In  the  chapter  on  Value 
Movement *  we  had  occasion  to  criticise  one  view  of  the  origin 
and  nature  of  aesthetic  attitudes  and  constructions  which,  because 
of  their  great  significance  for  the  expansion  of  social  sentiment 
and  its  maintenance  on  a  high  level  of  intensity,  sought  the  con 
ditions  of  aesthetic  expression  wholly  within  the  social.  While 
we  were  compelled  to  look  for  the  presuppositions  of  the  move 
ment  toward  the  aesthetic  in  the  individual,  and  thus  to  criticise 
the  merely  social  theory  of  its  origin,  we  did  not  ignore  its  sig 
nificance  as  a  vehicle  for  the  extension  of  social  sympathy. 
In  smaller  groups,  and  on  the  level  of  emotional  contagion — 
where  the  inducing  conditions  of  sympathetic  participation 
are  largely  perceptual,  as  in  the  dance  of  primitive  peoples— 
this  function  of  art  is  very  much  in  evidence.  On  the  higher 
ideational  level  also,  and  in  much  larger  social  groups,  national 
and  racial,  where  the  sentiment  shared  is  a  funded  meaning  of 
some  great  over-individual  ideal,  there  may  arise  something 
which  may  quite  properly  be  described  as  a  racial  assumption 
of  the  reality  of  that  ideal  and  of  its  complete  expansion. 
In  that  racial  assumption  the  individual  may  feel  an  absolute 
over-individual  worth,  and  may  realise  the  moments  of  inner 
peace  and  sublimity.  All  this  may  be  admitted,  and  yet,  when 
we  examine  more  closely  the  forms  in  which  this  great  group 
faith  or  illusion  incorporates  itself,  the  inherent  incapacity  of 
the  aesthetic  to  make  social  worths  absolute  shows  itself. 

1  Chap,  vii,  pp.  221  f. 


Objects  of  Impersonal  Over- Individual  Value      345 

The  very  condition  of  aesthetic  repose  is  individuation 
and  isolation.  Only  thus  can  an  object  acquire  immanental 
worth,  only  thus  are  the  instrumental  judgments  which  other 
wise  determine  the  worth  of  the  object,  and  which  give  us  only 
relative  value,  suppressed.  But  this  individuation  and  isola 
tion  is  possible  only  when  the  racial  ideal,  the  virtue  or  capacity, 
is  incorporated  in  some  great  individual,  the  "hero,"  or  in  some 
relatively  small  group.  The  meaning  of  the  ideal  in  all  its 
fullness  can  be  expressed  only  by  limiting  its  expansion.  Thus 
all  great  racial  art  which  approximates  to  the  expression  of 
absolute  worths  is  essentially  monarchical  or  aristocratic.  A 
democratic  art,  in  the  sense  that  it  may  be  shared,  is  possible — 
it  may  be  actually  prevalent  and  generally  appreciated, — but  in 
the  sense  that  it  represents  only  that  which  is  common  and 
undifferentiated  in  human  nature,  it  is  a  contradiction  in  terms. 

An  interesting  proof  of  this  position  from  the  negative  side 
is  to  be  found  in  the  incapacity  of  aesthetic  social  ideal  con 
structions  to  create  illusion.  Utopian  pictures  of  a  society  in 
which  the  social  worths  or  worth  dispositions,  which  now  have 
value  merely  instrumentally  with  reference  to  social  need  and 
demand,  are  universalised — where  communism  not  only  of  goods 
but  of  ideals  abounds,  where  altruism  reigns,  although  no  longer 
needed,  and  justice  without  injustice  to  be  righted — leave  us 
strangely  cold.  The  reasons  for  this  incapacity  of  impersonal 
social  ideals  for  the  acquirement  of  intrinsic  immanental  worth 
are  to  be  found,  not  only  in  the  fact  that  such  constructions 
must  remain  abstract  and  conceptual,  but  also  because  it  is 
impossible  for  us  to  assume  their  existence,  or  at  least  rest  in  the 
assumption,  with  any  degree  of  belief.  The  emotional  con 
ditions  of  belief  are  wanting,  and  intellectually,  the  assumption 
of  the  indefinite  applicability  of  these  dispositions  to  social 
ends,  when  subjected  to  scrutiny,  will  not  maintain  itself.  It  is 
only  when,  as  in  the  case  of  the  religious  ideal  of  the  Kingdom 
of  Heaven,  the  ideal  is  made  supernatural — when,  i.e.,  the  assump 
tion  underlying  the  aesthetic  contemplation  is  frankly  detached 
from  empirical  conditions  and  is  grounded  in  a  supernatural 
personality,  contrasted  with  the  entire  system  of  nature,  that 
belief  is  possible. 

This  incapacity  of  purely  impersonal  over-individual  worths 
for  aesthetic  idealisation  does  not,  however,  prevent  them  from 
being  the  object  of  absolute  immanental  worth  for  individuals 
and  groups.  The  relatively  complete  expansion  of  an  aspiring 


346  Valuation  :    its  Nature  and  Laws 

worth,  or  even  an  outlived  social  worth,  within  homogeneous 
groups,  may  create  the  illusion  of  complete  universalisation. 
The  conditions  here  are  favourable  to  the  isolation  and  indi- 
viduation  of  the  group  which  would  make  possible  such  an 
assumption.  The  very  contrast  and  opposition  of  such  groups, 
as  Utopian  societies  and  saintly  orders,  afford  the  conditions 
for  such  a  degree  of  intensification  and  expansion  of  these  worths 
within  the  group  as  to  arrest  critical  existential  and  instru 
mental  judgments,  and  to  favour  the  assumption  of  actual 
realisation.  These  conditions  are  indeed  favourable  to  the 
realisation  of  personal  worths  of  elevation  and  inner  peace, 
but  they  are  then  no  longer  impersonal. 

Religious  construction  is  closely  connected  with  aesthetic  in 
that  it  projects  its  social  worths  into  ideal  personalities,  but 
it  shows  this  important  difference  that,  whereas  the  aesthetic 
construction  is  all  directed  toward  repose  in  a  worth  already 
realised,  toward  an  hypostatisation  of  the  immanental  tendency 
in  worth  experience,  the  religious  construction  is,  in  its  purest 
form,  directed  toward  making  absolute  the  transgredient 
moment.  As  such,  it  projects  the  over-individual  worth  into 
a  person  or  persons  with  whom  the  individual  or  society  is  in 
volitional  relations.  This  difference  appears  in  the  different 
role  which  the  negative  moment  plays  in  the  two  constructions. 
In  the  aesthetic  constructions  the  tendency  is  simply  to  eliminate 
or  ignore  opposing  tendencies  as  illusion-disturbing  moments. 
In  the  religious  consciousness,  before  it  is  affected  by  philo 
sophical  reflection,  the  tendency  is  to  intensify  the  conscious 
ness  of  the  reality  of  the  ideal  by  opposing  it  to  certain  negative 
forces  to  which  equal  reality  is  ascribed.  In  both  cases  value 
is  intensified  by  contrast  and  detachment,  but  the  result  is 
attained  in  different  ways. 

Now,  in  so  far  as  religious  construction  is  a  social  phenomenon, 
it  is  clear  that  it  is  simply  an  extension  of  sympathetic 
projection.  If  it  is  true,  as  comparative  religion  seems  to  in 
dicate,  that  "  fear  first  made  the  gods,"  if  the  first  projection 
beyond  the  social  group,  in  over-social  persons  or  forces,  is 
that  of  the  negative  or  opposing  moment,  such  a  fact  is  easily 
understood  on  the  basis  of  our  analysis.  The  dispositions  with 
participation  value  remain  longest  instinctive  and  intra-social 
because  they  constitute  the  attitudes  and  dispositions  which 
through  sub-social  forces  were  selected  and  fixed  as  the 
instinctive  basis  of  participation.  But  once  the  negative 


Objects  of  Impersonal  Over- Individual  Value       347 

dispositions  have  been  projected,  the  positive  or  good  dis 
positions  with  participation  value  soon  follow.  From  this 
point  on  the  process  is  fairly  clear.  The  persons  of  the  gods 
in  whom  the  worth  dispositions  are  projected  correspond  at 
first  directly  with  group  distinctions.  Tribal  or  national 
gods  are  the  good  gods,  the  bad  deities  being  the  gods  of  the 
enemy.  The  highest  form  of  religious  construction,  on  the 
social  side,  is  attained  when  the  racial  limits  are  transcended 
and  the  god  becomes  one  god  and  the  god  of  all  men.  It  would 
seem,  then,  that  in  the  religious  construction  we  have  the  possi 
bility  of  the  universalisation  of  over-individual  values,  and  at 
the  same  time  of  emotional  participation  of  the  individuals 
in  these  values,  which  is  the  real  condition  of  absolute  worth. 
If  in  the  religious  consciousness  group  limits  may  really  be  trans 
cended,  and  if  in  the  ideal  constructions  of  religion  we  have  an 
object  for  universal  emotional  participation,  we  have  a  situation 
of  the  greatest  possible  significance,  for  we  have  here  the  possi 
bility  of  an  absolute  over-individual  demand  for  participation, 
and  such  participation  would  have  absolute  not  relative  value. 
Does  the  religious  consciousness  really  transcend  these  limits  ? 

In  order  that  there  may  be  emotional  social  participation 
in  these  over-individual  constructions,  two  conditions  seem 
to  be  requisite  :  (i)  the  personalisation  of  the  deity,  and  (2) 
the  opposition  of  the  deity  to  negative  tendencies,  preferably 
personalities.  The  religious  consciousness,  if  it  is  to  be  both 
emotional  and  social,  cannot  transcend  these  limits.  I  mean 
by  this  to  say  that  there  may  be  a  certain  type  of  emotional 
participation  of  the  individual  in  a  being  conceived  to  transcend 
these  limits,  but  the  moral  and  social  sphere  has  been  left  behind. 
Attempts  have  been,  and  are  constantly  being  made,  in  the  in 
terests  both  of  purely  intellectual  and  of  worth  continuity,  to 
transcend  these  limits  ;  and  these  efforts  take  the  form  of  a  panthe 
istic  monism  in  which  the  fundamental  notes  are  universalisation 
of  some  worth  attribute  and  abstraction  from  anthropomorphic 
personal  and  group  limits.  It  is  with  these  constructions  only 
in  so  far  as  they  are  worth  constructions,  and  with  their  capacity 
for  becoming  objects  of  absolute  worth,  that  we  are  here  con 
cerned.  When  viewed  in  this  aspect,  as  a  worth  construction 
which  shall  at  the  same  time  be  completely  over-personal  and 
the  object  of  worth  feeling,  the  process  is  seen  to  be  self-defeating. 

The  facts  which  show  this  may,  for  the  sake  of  emphasis, 
be  put  in  the  form  of  a  dilemma,  which  may  be  described  as  the 


348  Valuation :   its  Nature  and  Laws 

dilemma  of  all  pantheistic  worth  constructions.  When  the 
worth  attribute,  whether  it  be  pure  reason — and  reason  in  these 
constructions  is  always  a  worth  attribute — or  some  ineffable 
form  of  experience  which  transcends  reason,  is  thus  universalised 
and  made  absolute,  one  of  two  things  happens.  Either  it 
ceases  to  be  a  worth  construction  with  any  intrinsic  meaning 
whatever,  and  becomes  a  purely  instrumental  intellectual  con 
struction,  or  else  it  passes  from  the  sphere  of  impersonal  worth 
objects  into  the  region  of  the  subjective  and  personal.  In  either 
case  it  has  transcended  the  region  of  the  moral.  In  the  great 
historic  pantheistic  constructions  this  emotional  logic  is  every 
where  apparent.  Hinduism  can  attain  the  purely  impersonal 
worth  object  only  through  atheism,  and  the  object  thus  created 
can  remain  an  object  of  feeling  only  by  becoming  identified 
with  the  self,  as  in  the  famous  phrase,  "  That  art  thou  !  "  The 
attempt  to  make  absolute  the  over-individual  worth  construction 
reduces  the  social  moral  world  to  illusion,  and  in  the  moment 
of  attainment  it  is  no  longer  an  impersonal  but  a  personal  worth 
of  inner  peace  that  is  achieved.1 

We  are  justified,  then,  in  concluding  this  study  with  the 
statement  that  religious  construction  in  its  social  aspect  may, 
like  the  purely  aesthetic  construction,  enhance  the  consciousness  of 
over-individual  value,  but  it  cannot  transcend  the  laws  inherent 
in  social  construction.  Religion  may  give  complementary 
values  to  social  ends,  and  intensify  the  feeling  of  social  obli 
gation,  but  to  do  so  it  must  remain  dualistic  and  anthropo 
morphic.  This  dualism  is  overcome  in  moments  of  contempla 
tion  and  faith,  but  the  condition  of  these  experiences  is  Einfiih- 
lung  in  its  aesthetic  individuating  form,  not  impersonal  partici 
pation.  The  personal  and  impersonal  values  fuse  in  an  absolute 
intrinsic  value.  As  immediate,  it  is  over-personal  and  over- 
social  in  its  meaning.  But  it  still  remains  a  personal  value  in 
the  sense  that  it  is  only  as  a  practical  absolute,  as  the  limit  of 
a  series  of  personal  experiences,  that  it  has  axiological  meaning 
and  validity. 

Whether  this   implies   that   personality  in   its  aesthetic   and 

1  The  intellectual  love  of  Spinoza  and  the  cosmic  emotion  of  Clifford  are  also  doubt 
less  real,  or  at  least  possible,  experiences,  in  which  absolute  immanental  worth  is 
attained,  but  it  is  a  personal  worth  with  which  we  are  concerned  in  both  cases.  Such 
emotion  does  not  attach  to  the  purely  impersonal  world  order  when  abstracted  from 
the  subject  which  constructs  and  individuates  it.  The  mechanical  world  order,  the 
immensities  of  space  and  time  are,  as  such,  worthless.  It  is  only  when  the  soul 
whispers  to  itself,  in  varied  dialects  to  be  sure,  the  magical  and  mystical  words,  "  That 
art  thou"  that  the  "that"  acquires  worth. 


Objects  of  Impersonal  Over- Individual  Value       349 

religious  form  is  the  highest  category  of  worth  experience  we 
need  not  here  inquire— since  our  interest  is  confined  to  the 
dynamics  of  valuation.  If  it  is,  it  must  take  up  into  it  all  the 
meanings  of  the  social  consciousness.  What  we  must  insist 
upon  is  that  religious  and  aesthetic  experiences  have  axiological 
validity,  and  are  not  merely  empty  mysticism,  only  in  so  far  as 
they  retain  as  their  content  and  indispensable  presuppositions, 
the  meanings  acquired  in  personal  participation.  It  is  as  means' 
of  enlargement,  not  loss,  of  personality  that  they  have  signifi 
cance. 


CHAPTER   XII 

OVER-INDIVIDUAL  VALUES  (Continued). 

I.  INTERPRETATION  OF  CONCRETE  MORAL  JUDGMENTS  IN  THE 
LIGHT  OF  THE  PRECEDING  ANALYSIS  AND  THEORY 

The  Problem. 

WE  come  now  to  the  study  of  the  concrete  phenomena  of  moral 
judgment  which  must  test  the  truth  of  the  preceding  theory — 
the  analysis  of  judgments  of  moral  obligation  and  imputation.1 
If  moral  values  are  actual  values,  determined  by  processes  of 
social  participation  and  by  the  laws  of  social  synergy,  then 
these  judgments  should  reflect,  in  their  qualitative  and  quantita 
tive  aspects  alike,  the  expectations  and  demands  created  by 
these  processes.  The  situation  here  is  entirely  similar  to  that 
which  presented  itself  in  the  sphere  of  personal  worths.  As  a 
result  of  our  study  of  their  genesis,  certain  laws  were  developed, 
and  these  were  shown  to  be  reflected  in  the  phenomena  of  per 
sonal  obligation  and  imputation.  A  theory  of  the  nature  and 
laws  of  moral  values  has  likewise  been  developed,  and  the  phe 
nomena  of  moral  judgment,  if  we  succeed  in  isolating  them, 
should  reflect  these  laws.  More  specifically,  since  moral  values 
are  the  objective  participation  values  of  dispositions,  and  since 
these  values  are  by  their  very  nature  subject  to  the  law  of  Mar 
ginal  Participation  Value,  we  should  expect  the  moral  judgment 
to  reflect  in  its  quantitative  aspects  the  operations  of  this  law. 
An  analysis  of  the  empirical  laws  of  moral  judgment  will  show 
this  hypothesis  to  be  justified. 

i.  The  Object  of  Moral  Judgments — The  Morally  Qualified  Act. 

The  prerequisites  of  such  an  empirical  study  are,  as  we  have 
seen  in  a  similar  analysis  of  personal  worths,  the  isolation  of 

1  For  similar  studies  of  these  phenomena,  and  also  for  some  of  the  terminology  used 
in  this  discussion,  the  reader  may  be  referred  to  Meinong,  Psych.  -Ethisch.  Unter- 
suchungen  zur  Werttheorie,  chap,  ill  of  Part  II  (Vom  moralischen  Sollen)  ;  also 
Ehrenf'els,  System  der  Werttheorie,  Part  II,  pp.  195-205  ;  also  Simmel,  Einleitung  in 
die  Moralwissenschaft,  Vol.  II,  p.  323. 

35° 


Over- Individual  Values  351 

the  object  of  judgment,  the  fixation  of  the  terms  in  which  esti 
mation  of  moral  value  takes  place,  and  the  definition  of  the 
presupposed  demand  in  terms  of  its  norms  and  limits — that  is, 
the  limits  within  which  the  estimation  moves. 

The  definition  of  the  morally  qualified  object,  as  already 
given,  isolates  the  phenomena  in  a  preliminary  way.  The 
morally  qualified  act  we  found  to  be  an  act  which  expresses  a 
disposition  to  impersonal  participation,  the  response  to  an 
impersonal  demand.  The  morally  qualified  judgment  upon 
such  an  act  is  one  in  which  abstraction  is  made  from  all  sub 
jective  and  personal  elements,  and  which  reflects  a  disposition 
to  judgment  as  determined  solely  by  the  over-individual  de 
mand.  It  is  the  judgment  of  the  "  impartial  spectator."  Further 
studies  have,  however,  caused  us  to  modify  this  definition 
somewhat.  If,  as  has  appeared,  over-individual  demands, 
to  be  felt  at  all,  must  be  related  to  the  concrete  interests  of 
a  group  and  can  never  be  wholly  abstract  and  universal,  then, 
strictly  speaking,  there  is  no  purely  impersonal  participation 
and  no  wholly  impartial  spectator.  The  morally  qualified 
act  and  the  morally  qualified  judgment,  as  at  first  defined, 
would  alike  be  ideal  limits  never  actually  realised.  This  con 
clusion  we  may  admit.  The  moral  value  of  an  act  is  always 
an  actual  social  value,  and  the  demand  presupposed  by  moral 
obligation  toward  such  an  act,  and  by  moral  judgment  upon  it, 
always  represents  a  concrete,  relative  value  determined  by  the 
laws  of  social  synergy.  Moral  values  are  actual,  not  ideal, 
as  in  the  case  of  personal  values.  To  this  statement  the  "  ideal 
society  "  for  which  the  reformer  works  seems  to  present  a  con 
tradiction.  The  values  there  are  ideal,  and  at  the  same  time 
apparently  moral  and  social.  Nevertheless,  the  contradiction 
is  only  apparent.  In  so  far  as  they  are  ideal  they  are  per 
sonal  and  intrinsic.  The  reformer's  utopia  may  be  intrinsically 
desirable,  but  it  cannot  yet  be  determined  whether  the  ideal 
can  be  actualised.  Only  those  ideals  which  have  already  been 
at  least  partially  actualised,  and  have  taken  form  in  an  over- 
individual  demand,  institutional,  legal,  or  social,  can  form  a 
standard  for  an  impartial  judgment  of  the  actual  moral  value 
of  an  act.  Thus  even  the  reformer  who  works  for  ideal  con 
ditions  would  hesitate  to  judge  the  men  of  his  time  by  any  other 
than  the  best  actualised  standards.1 

1  Even  when  the  "radical"  promulgates  ideals  and  standards  in  opposition  to  the 
social  valuations  of  his  time,  and  seeks  to  organise  groups  about  these  ideals,  he  does 


352  Valuation:   its  Nature  and  Laws 


2.  The  Terms  of  Estimation  of  Moral  Value — Egoism  and  Altruism. 

All  estimation  of  relative  value,  we  have  repeatedly  seen, 
takes  place  in  terms  of  two  variables  :  affirmation  of  conative 
tendency  and  its  arrest,  which  stand  for  the  positive  and  negative 
moments.  In  the  case  of  impersonal  judgment  upon  impersonal 
participation,  the  positive  factor  is  the  tendency  to  participate 
in  over-individual  trends,  to  contribute  to  over-individual 
goods,  while  the  negative  moments  are  the  "  condition  "  and 
"  personal "  worths  which  may  arrest  this  tendency  to  par 
ticipate.  In  estimating  the  participation  value  of  the  act  we 
measure  its  positive  altruistic  quality  in  terms  of  the  strength 
of  the  egoistic  tendencies  sacrificed  for  the  altruistic. 

It  will  be  observed  that  in  this  social  or  moral  reckoning 
the  distinction  between  condition  and  personal  worths,  funda 
mental  in  personal  imputation,  becomes  irrelevant,  and  both 
are  subsumed  under  the  general  term  of  egoistic  tendencies  and 
dispositions,  a  clear  indication  of  the  process  of  abstraction 
by  which  the  impersonal  point  of  view  is  reached.  The  dispo 
sitions  described  as  altruistic  may  include  attitudes  or  tendencies 
to  participate  which  display  different  degrees  of  impersonality. 
Thus  one  may  sacrifice  individual  ends  for  the  good  of  another 
individual,  for  the  good  of  a  limited  group,  or  for  relatively 
abstract  social  ends.  But  the  essential  of  such  participation 
is  that  there  be  some  contrast  between  individual  and  over- 
individual  good — otherwise  we  have  purely  personal  relations 
and  personal  worths.  The  extreme  of  altruism,  as  understood 
in  this  reckoning,  is  the  case  where  the  object  or  person  for  which 
the  sacrifice  is  made  is  so  remote  from  our  personal  sympathies 
that  the  participation  is  wholly  unemotional. 

Moral  imputation,  as  distinct  from  personal  and  ethical, 
is  therefore,  strictly  speaking,  judgment  upon  the  act  as  thus 
qualified.  An  act  is  said  to  be  "  correct,"  deserving  or  blame 
worthy,  according  to  the  relation  between  the  egoistic  and 
altruistic  tendencies  expressed  by  the  act.  These  qualifications 
of  the  act,  and  the  degrees  of  emphasis  with  which  they  are 

not  seriously,  in  ordinary  social  relations  at  least,  judge  his  fellows  by  these  standards. 
Except  for  purposes  of  pedagogical  effect  (as  in  the  case  of  Bernard  Shaw,  perhaps),  he 
does  not  in  actual  situations  go  beyond  the  normal  demand.  His  sense  of  humour,  if 
not  of  justice,  and  his  intellectual  apprehension  of  the  principle  of  "  economy  of  the 
truth"  (especially  when  the  truth  is  so  very  inner  and  ideal)  is  usually  sufficient  to  guard 
him  against  such  fallacies  of  worth  judgment.  Nietzsche's  personal  and  social  relations 
are  excellent  illustrations  of  this  fact. 


Over- Individual  Values  353 

predicated,  are,  moreover,  reducible  to  a  function  of  these 
positive  and  negative  moments.  But  while  moral  judgment  is 
primarily  upon  the  act,  and  upon  the  act  as  instrumental, 
that  is,  as  contributing  to  over-individual  worths,  the  moral 
judgment  may  easily  pass  over  into  the  ethical  and  personal. 
Altruism,  as  we  have  seen,  may  be  judged  both  as  a  personal 
and  a  social  worth ;  and,  while  in  the  concrete  worth  experience 
the  two  are  not  kept  distinct,  yet  it  is  necessary  for  our  scientific 
purposes  that  they  should  be. 

(a)  The  Amount  of  Altruistic  Disposition  as  Measured  by  the 

Character  of  the  Individual  Good  Sacrificed. 

If  now  we  turn  to  a  more  specific  examination  of  this  reckon 
ing  in  terms  of  egoism  and  altruism,  we  find  that  the  dispositions 
thus  described  may  vary  in  several  significant  aspects,  each 
type  of  variation  affecting  the  moral  judgment,  which  is  a  function 
of  the  two  variables.  In  the  first  place,  there  is  the  amount  of 
the  disposition  to  participate,  as  measured  by  the  character  of 
the  condition  and  personal  worths  sacrificed.  Condition  worths 
may  vary  along  the  whole  scale  from  the  existence-minimum  to 
the  wholly  worthless,  personal  worths  from  the  characterisation- 
minimum,  which  is  without  capacity  of  substitution,  to  the  most 
superficial  personal  quality.  As  an  illustration  we  may  take  the 
case  of  the  rescue  of  a  human  life  (an  act  of  high  participation 
value)  :  (a)  by  the  sacrifice  of  a  few  hours  of  ease  and  pleasure  ; 

(b)  by  the  risk  of  one's  good  name  ;    (c)  by  the  loss  of  all  one 
values  of  condition  and  person,  or  even  loss  of  life  itself.     If  we 
bring  no  other  elements  into  the  reckoning,  it  is  obvious  that 
the  degree  of  altruism  is  a  steadily  increasing  one  in  these  three 
cases,  and  that,  other  things  being  equal,  the  moral  value  of  the 
act  increases  accordingly. 

(b)  As  Measured  by  the  Character  of  the  Over-Individual  Good  for 
which  Sacrifice  is  Demanded. 

A  second  factor  enters  into  our  estimation  when  we  take 
into  account  the  character  of  the  over-individual  demand  to 
which  the  given  act  constitutes  a  response.  This  demand 
differs :  (a)  according  to  the  nearness  or  remoteness  of  the  object 
from  personal  or  group  sympathies;  and  (b)  according  to  the 


2   A 


354  Valuation:   its  Nature  and  Laws 

degree  to  which  the  demand  for  the  object  has  become  uni- 
versalised.  The  latter  factor  we  may  call  the  "  coefficient  of 
projection  or  participation,"  by  which  is  meant  the  character  of 
the  demand  as  representing  a  normal,  aspiring,  or  outlived 
value. 

In  the  first  case,  the  nearness  or  remoteness  of  the  object 
determines  the  emphasis  of  our  judgment,  of  our  positive  judg 
ment  when  the  act  takes  place,  of  negative  when  the  act  is 
omitted.  Thus  sacrifice  of  condition  or  personal  worths  for 
one's  child  will  receive  less  praise  than  sacrifice  for  a  friend,  for 
an  unknown  man,  or  for  an  abstract  principle.  And,  in  the 
same  order,  the  failure  to  sacrifice  will  call  out  disapproval  or 
negative  judgment — strong  disapproval  in  the  case  of  failure  to 
sacrifice  for  one's  child,  less  and  less  disapproval  as  the  objects 
become  more  and  more  remote  from  the  emotional  participation 
of  the  individual.  Again  it  may  be  observed  that  our  judg 
ments  take  into  account  the  projection  co-efficient  of  the  senti 
ments,  participation  in  which  is  demanded.  Sacrifice  for  a 
value  which  has  already  attained  a  high  "  expansion-coefficient," 
large  social  recognition,  let  us  say  duty  to  the  family,  integrity 
in  business,  freedom  in  the  State,  is  rated  less  highly  than  sacri 
fices  for  worths  which  have  attained  less  expansion,  which  are 
not  normal,  but  rather  aspiring  values,  as,  for  instance,  new 
ideals  of  truth  or  social  justice  just  beginning  to  be  strongly  felt. 
It  would  appear,  then,  that  when  we  take  the  attitude  of  the 
impartial  spectator  we  cannot  abstract  from  a  consideration  of 
the  instrumental  value  of  the  sacrifice  itself,  i.e.,  from  the  relative 
importance  of  the  object  sacrificed  and  of  the  over-individual 
demand  to  which  the  sacrifice  is  a  response.  Altruism  as  a 
social  value  is  not  possessed  of  the  capacity  of  indefinite  increase  ; 
its  felt  importance  has  a  limit  determined  by  the  laws  of  social 
participation  already  developed. 

It  cannot  be  denied,  of  course,  that  absolute  sacrifice  is  actually 
demanded  of  ourselves  and  others,  that  an  act  which  is  so  insig 
nificant  as  to  be  below  the  threshold  of  instrumental  value  may 
not  only  become  so  important  as  to  demand  such  sacrifices,  but 
the  sacrifice  may  actually  acquire  absolute  value,  as  was  seen 
in  our  study  of  personal  worths.  It  is  important  to  recognise, 
however,  that  this  value  is  no  longer  merely  impersonal  and  over- 
individual,  that  which  it  has  for  the  impartial  spectator,  but  is 
a  complementary  value  which  is  acquired  through  reference  to 
the  concept  of  the  personality  or  to  quasi-personal  constructions 


Over-  Individual  Values  35  c 

of  the  group  or  the  nation,  as  an  indispensable  part  of  a  total 
complex.  The  merit  imputed  in  such  a  case  is  not  the  reflex 
of  an  over-individual  demand,  but  of  a  personal  demand  which 
arises  through  identification  of  the  attitude  with  the  personality 
The  importance  of  this  will  be  especially  apparent  when  we 
come  to  study  the  laws  of  preference  between  personal  and  over- 
individual  worths.1 

(c)  The  Significance  of  these  Terms  of  Estimation. 

When  the  results  of  this  analysis  are  properly  weighed 
they  are  seen  to  constitute  strong  grounds  for  the  hypothesis' 
that  moral  judgments  reflect  the  working  of  the  laws  of  social 
sympathy,  and  more  specifically  the  law  of  Marginal  Partici 
pation  Value.  Estimation  of  the  degree  of  moral  value  in  terms 
of  the  character  of  the  individual  good  sacrificed  and  of  the 
over-individual  end  for  which  it  is  sacrificed,  turns  upon  the 
degree  of  concreteness  and  immediacy  or  abstractness  and 
remoteness  of  the  goods  and  ends. 

For,  in  the  first  place,  this  imputation  of  value  in  degrees 
varying   with    the   remoteness    of   the    object   from   emotional 
sympathetic  participation,  is  precisely  what  we  should  expect 
m  the  light  of  the  law  that  intensity  of  sympathy  decreases 
with    increase    of    the    generality    of    resemblance.     It    means 
simply  that  our  sense  of  over-individual  value  is  relative  to 
our  capacity  to  represent  in  sympathetic  projection  the  affective- 
volitional  meaning  of  others.     Again,  the  variations  in  degree 
f  emphasis  of  the  judgment,  corresponding  to  the  differences 
in  social  sentiment  described  as  normal,  aspiring,  and  outlived 
values,   indicate  the  working  of    the    same  law,   according  to 
which  the  intensity  of  an  over-individual  trend  decreases  with 
its  expansion.     Where,  as  in  the  aspiring  value,  the  intensity  is 
in  excess  of  the  expansion,  there  the  emphasis  is  greater  than  in 
the  case  where  the  relation  of  intensity  to  expansion  is  more 
nearly  normal.     These  are  all  aspects  of  the  law  of  Marginal 
Participation  Value.2 

1  Chap.  xni. 


356  Valuation  :   its  Nature  and  Laws 


II.  THE  THRESHOLDS  AND  NORMS  OF  MORAL  JUDGMENT 

These  general  facts  with  regard  to  the  conditions  which 
determine  the  changes  of  emphasis  in  our  moral  judgments 
being  established,  we  may  now  turn  £o  a  quantitative  study  of 
these  changes.  And  if  the  preceding  analysis,  largely  qualitative, 
made  our  hypothesis  probable,  the  facts  now  to  be  considered 
make  this  probability  a  practical  certainty.  In  a  general  way 
we  have  seen  that,  if  no  changes  in  the  character  of  the  over- 
individual  object  be  introduced,  the  moral  value  imputed  to 
the  act,  within  certain  limits  which  will  presently  be  deter 
mined,  varies  directly  with  the  amount  of  sacrifice  of  condition 
and  personal  worths.  The  question  now  arises  whether  this 
variation,  its  laws  and  its  limits,  can  be  more  accurately  de 
termined — in  order  that  we  may  connect  them  with  the  psychical 
processes  which  condition  these  judgments,  and  compare  the 
region  of  impersonal  with  that  of  personal  worth  judgment. 

The  first  requisite  of  such  a  study  is  the  determination  of 
conceptual  points  or  thresholds  from  which  the  increase  and 
decrease  of  judgmental  emphasis  rise  and  fall.  Two  such 
critical  points  may  be  distinguished.  They  may  be  described 
as  the  Norm  of  Participation  or  the  "  correct,"  and  the  Partici 
pation-Minimum.  In  general  the  correct  represents  the  normal 
expectation,  fulfilment  of  which  is  accompanied  by  neither 
praise  nor  blame.  The  minimum  of  participation  represents 
the  smallest  .quantity  of  a  social  good  for  which  sacrifice  of 
individual  good  is  demanded.  These  conceptual  points  are 
in  principle  not  difficult  to  determine,  although,  owing  to  changes 
in  social  values,  they  are  not  always  easy  to  define  in  a  given 
concrete  situation. 

i.    The  Norm  of  Participation  or  the  "Correct" --The  Normal 
Expectation  of  Social  Participation. 

In  defining  the  region  of  the  "  correct  "  the  first  thing  to  be 
noted  is  that,  in  comparison  with  the  normal  threshold  of 
personal  worth,  it  is  much  cruder,  much  less  sensitive.  The 
sphere  of  the  correct  includes  wider  variations  than  the  corre 
sponding  sphere  in  personal  worth  feeling.  Praise  and  blame 
come  less  quickly  in  impersonal  than  in  personal  valuation. 
The  "  correct  "  in  wider  business  circles  allows  of  considerable 


Over- Individual  Values  357 

variation  from  the  standard  demanded  in  personal  intercourse. 
The  sacrifices  of  his  personal  interests  for  the  larger  interests  of 
his  profession  demanded  of  the  physician  as  correct,  may  seem 
quixotic,  or  be  wholly  unintelligible,  to  wider  business  groups.  The 
reason  is  not  far  to  seek.  Social  and  impersonal  participation  is 
instrumental,  and  a  display,  either  of  egoism  or  altruism,  does 
not  become  significant,  does  not  rise  above  the  threshold  of  the 
correct  into  the  regions  of  praise  and  blame,  until  its  importance 
can  be  felt  by  the  cruder  sense  of  the  multitude.  But  the  fact 
must  not  be  overlooked  that  the  region  of  the  correct  expands 
and  contracts  with  differentiation  or  fusion  of  sentiment,  with 
group  segregation  or  combination. 

The  region  of  the  correct  in  impersonal  moral  judgment 
represents,  then,  the  normal  disposition  to  participate,  the 
normal  sacrifice  of  egoistic  to  altruistic  interests,  whether  for 
individuals  or  for  social  ideals.  In  any  particular  case  of  moral 
judgment,  this  norm  of  sacrifice  is  strictly  relative  to  the  sig 
nificance  of  the  over-individual  object  or  ideal,  as  we  have 
defined  that  significance  above,1  for  which  the  sacrifice  is  made. 
The  disposition  normally  expected  is,  therefore,  a  direct  product 
of  the  laws  of  social  synergy  as  we  have  defined  them. 

2.  The  Participation-Minimum — The  Lower  Threshold. 

At  the  point  of  the  correct  there  is  no  imputation  of  praise  or 
blame.  To  remain  in  the  social  niveau  is  not  meritorious,  and, 
by  reason  of  the  crudeness  of  the  social  sense  at  this  point, 
slight  variations  in  excess  or  defect  of  the  normal  are  not  readily 
marked.  But  as  the  variations  increase  the  judgments  of  praise 
and  blame  appear.  As  the  altruistic  disposition  to  sacrifice 
increases,  so  within  certain  limits  the  judgments  of  praise  also 
increase.  This  increase  of  altruism  may,  as  we  have  seen,  be 
measured  in  two  ways :  either  in  terms  of  the  character  of  the 
over-individual  object  for  which  sacrifice  is  made,  or  in  terms 
of  the  character  of  the  individual  good  sacrificed.  Now,  measured 
in  either  way,  there  is  a  limit  to  the  praise  imputed  for  the  per 
formance  of  an  act  or  of  blame  for  its  omission.  This  limit  is 
the  participation-minimum  already  described,  the  minimum  of 
the  over-individual  good  for  which  sacrifice  is  demanded,  or, 
what  amounts  to  the  same  thing,  the  maximum  of  sacrifice  of  in 
dividual  good  demanded.  When  this  minimum  for  participation 

1  Cf.  above,  pp.  353  f. 


35 ^  Valuation  :   its  Nature  and  Laws 

is  reached,  in  so  far  as  the  praise  or  blame  is  purely  moral  or 
impersonal,  the  performance  of  the  act  is  not  praised  and  its 
omission  not  blamed. 

The  existence  and  function  of  this  participation-minimum 
appears  clearly  in  certain  concrete  cases.  The  demand  for  truth 
fulness,  respect  for  property,  and  benevolence,  are  more  or  less 
completely  universalised  sentiments.  But,  from  an  impersonal, 
impartial  point  of  view,  we  do  not  expect  a  man  to  sacrifice 
important  personal  interests,  or  ultimately  life  itself,  for  a 
merely  formal  adherence  to  an  insignificant  truth.  Neither 
do  we  expect  him  to  make  such  sacrifice  for  the  comfort  of 
a  total  stranger,  nor  to  suffer  pain  or  death  rather  than  ap 
propriate  an  insignificant  object  to  which  his  name  is  not  at 
tached. 


III.  JUDGMENTS  OF  IMPUTATION  AND  OBLIGATION— 
AS  RELATED  TO  THESE  NORMS  AND  LIMITS 

i.  Imputation  of  Praise  and  Blame. 

The  minimum  for  participation  represents  that  minimum  of 
the  over-individual  good  beyond  which  sacrifice  is  not  demanded 
by  the  impartial  spectator,  and  the  demand  for  sacrifice  falls 
off  as  the  good  approaches  this  minimum.  Any  exhibition 
of  altruistic  disposition  in  excess  of  this  demand  will  then  be 
supra-normal.  Let  us  see  how  the  judgments  of  praise  and 
blame  are  constituted  at  this  point. 

A  disposition  in  excess  of  the  normal  may,  for  reasons  which 
we  have  seen,  not  be  noticeable  at  first,  but  when  it  becomes 
apparent  it  calls  out  judgments  of  praise,  for  the  very  good 
reason  that  in  general  altruism  is  in  demand.  But  the  increase 
of  the  disposition  is  not  accompanied  by  a  corresponding  in 
crease  in  the  degree  of  praise.  Altruism,  as  a  social  good,  is 
not  susceptible  of  indefinite  increase.  As  the  excess  of  the 
disposition  in  proportion  to  the  object  becomes  more  and  more 
marked,  the  judgmental  emphasis  begins  to  fall  off,  and  finally 
a  point  is  reached  where  praise  passes  over  into  blame.  A  good 
illustration  is  that  of  the  mother  who  persists  in  sacrificing  im 
portant  personal  and  condition  worths,  health,  strength,  and 
her  own  interests  of  various  kinds,  for  minor  worths  of  the  child. 
In  such  a  case  the  moral  value  of  the  act  decreases  relatively  to 
the  excess  of  the  sacrifice  over  the  normal,  and  threatens  to 


Over -Individual  Values  359 

pass  over  into  the  region  of  the  blameworthy.  In  the  same 
manner  failure  to  sacrifice  is  accompanied  by  judgments  of 
blame,  but  with  less  and  less  emphasis  until,  as  the  minimum  is 
reached,  the  failure  becomes  negligible. 

How  different  all  this  is  from  personal  imputation  is  patent. 
There  it  was  precisely  at  these  points,  at  the  minima  of  char 
acterisation  (and  of  participation  also  when  the  over-individual 
good  is  identified  with  the  person),  that  absolute  worths  ap 
peared.  Thus  in  the  case  of  the  preceding  illustration  of  the 
mother,  if  we  participate  in  her  act  emotionally,  that  is,  isolate 
her  aesthetically  from  the  social  value-process,  an  absolute 
sacrifice  for  a  trivial  end  may  acquire  absolute  value.  In  such 
a  case  we  have  intrinsic  valuation  of  the  mother  as  such,  repose 
in  the  idea,  and  the  principles  of  intellectual  imputation  are 
transcended.  The  ethical  theorist  might  say — upon  reflection— 
that  the  reason  the  moral  judgment  takes  this  form  is  that,  as 
a  result  of  her  excessive  sacrifice  for  the  child,  other  services  to 
society  at  large,  and  even  to  the  child  itself,  are  made  impossible 
— more  important  values  of  the  same  kind,  are  sacrificed  to  lower. 
Such  a  logical  relation,  such  an  ultimate  harmony  of  ends  and 
norms,  may  conceivably  be  worked  out  by  a  philosophy  of  ethical 
values,  but  the  presuppositions  of  concrete  judgments  are  not 
derived  reflectively,  but  result  from  the  working  of  the  empirical 
laws  of  valuation. 

In  this  connection  another  fact,  already  noted  in  the  preceding 
chapter,  is  significant.  In  personal  characterisation  the  emo 
tional  accompaniments  of  the  act  are  relevant.  In  immediate 
personal  participation  we  infer  the  disposition  presupposed 
from  the  emotional  expressions  which  constitute  the  inducing 
grounds  of  our  sympathetic  participation.  Whether  the  act 
of  participation  is  accompanied  by  passive  participation,  lively 
sympathy,  apparent  readiness  to  offer  oneself,  or  enthusiastic 
sacrifice,  is  a  question  of  decided  moment.  A  person  with  sensi 
tive  feeling  for  personal  worths  would  prefer — and  rate  higher— 
an  insignificant  act  with  evidence  of  gracious  insight  and  feeling, 
to  a  much  more  important  act  done  with  a  sullen,  or  even  cool, 
sense  of  duty.  There  is  ample  room  for  the  "  pathetic  fallacy  " 
here,  yet  most  of  us  prefer  our  illusions  because  of  their  fruitful- 
ness  for  life  and  faith.  But  these  very  accompaniments,  so 
significant  in  emotional  imputation,  are  irrelevant  for  the  im 
personal  imputation  of  moral  judgment.  A  minimum  of  dis 
position  requisite  for  the  bringing  forth  of  the  normal  act  is 


360  Valuation  :   its  Nature  and  Laws 

demanded,  but  all  beyond  that  is  more  or  less  irrelevant.  I  say 
"  more  or  less,"  for  while  it  is,  in  a  measure,  taken  into  account, 
because  the  impersonal  judgment  is  never  pure,  nevertheless, 
the  value  imputed  to  these  manifestations  is  far  from  being  in 
proportion  to  the  excess  of  disposition  displayed.  When  a  man 
is  righteous  overmuch,  the  excess  is  but  grudgingly  acknowledged 
and  approved  by  social  judgment. 

2.   Moral  Obligation. 

When  we  turn  to  the  more  internal  aspect  of  the  over-individual 
demand,  as  it  finds  expression  in  the  individual's  sense  of 
impersonal  demand  or  obligation  to  participate,  we  find  the 
same  general  laws  at  work,  determining  the  conditions  of  the 
demand's  being  felt  at  all  and  the  intensity  with  which  it  is  felt. 
Oligation  being  but  the  reverse  side  of  imputation  of  praise 
and  blame,  this  is  what  we  should  expect ;  and,  while  it  is  per 
haps  more  difficult  to  isolate  impersonal  from  instinctive  and 
personal  obligations  than  to  distinguish  between  personal  and 
impersonal  imputation,  nevertheless,  the  phenomena  are  suf 
ficiently  distinguishable  to  enable  us  to  show  the  working  of 
these  laws,  the  laws  of  actual  over-individual  values. 

a)  It  Reflects  Actual  Objective  Participation  Values. 

In  conformity  with  the  nature  of  moral  values  as  actual, 
we  find  that,  in  order  that  impersonal  obligation  may  be  felt 
at  all,  there  must  be  as  a  necessary  presupposition  the  judgment 
of  the  existence  of  an  actual  over-individual  demand.  More 
over,  and  this  is  still  more  significant,  there  must  be  presupposed 
the  certainty,  probability,  or  at  least  possibility,  of  the  act 
in  question  being  instrumental  to  the  realisation  of  the  over- 
individual  social  end.  The  degree  of  obligation  decreases  accord 
ing  as  the  judgment  is  one  of  certainty,  probability,  or  mere 
possibility,  and  lapses  entirely  with  the  judgment  of  impossi 
bility.  This  condition  of  impersonal  obligation,  which  a  further 
analysis  of  the  facts  will  clearly  show  to  exist,  is  in  striking 
contrast  to  the  conditions  of  personal  obligation,  where,  as  we 
have  seen,  the  feeling  of  obligation  to  personal  ideals  is  not  so 
conditioned. 

A  vivid  characterisation  of  the  critical  point  at  which  over- 
individual  obligation  lapses  is  given  in  Robert  Louis  Stevenson's 


Over-Individual  Values  361 

fable  of  the  two  men  on  the  sinking  ship.  The  captain  of  the 
sinking  ship,  it  must  be  premised,  has  just  found  one  of  the 
hands  smoking  in  the  powder  magazine.  "  '  For  my  own  poor  part,' 
says  the  captain,  '  I  should  despise  a  man  who,  even  on  board 
a  sinking  ship,  should  omit  to  take  a  pill  or  to  wind  his  watch. 
That,  my  friend,  would  not  be  the  human  attitude  !  '  'I  beg 
pardon,  sir,'  said  Mr.  Spoker,  '  but  what  is  precisely  the  differ 
ence  between  shaving  in  a  sinking  ship  and  smoking  in  the 
powder  magazine  ?  '  'Or  doing  anything  at  all  in  any  con 
ceivable  circumstances  ?  '  cried  the  captain.  '  Perfectly  con 
clusive.  Give  me  a  cigar  !  '  Two  minutes  after  the  ship  blew 
up  with  a  glorious  detonation."  With  the  lapsing  of  all  possi 
bility  of  there  being  social  significance  to  the  act,  lapsed  all 
obligation  toward  the  ordinary  duties  of  life.  Personal  obliga 
tions,  the  captain  would  probably  admit,  still  remain.  Just  as 
one  may  not  be  nasty  in  the  dark,  so  when  courage  is  of  no  more 
avail,  we  do  not  expect  a  man  to  lose  his  personal  worth  of  man 
liness.  Obligations  of  the  instinctive,  appreciative  order,  of 
strength,  etc.,  also  still  remain.  But  with  the  lapsing  of  the  possi 
bility  of  realisation  of  over-individual  good,  all  social  obligation 
lapses. 

The  situation  is  brought  out  still  more  clearly  in  the  obligation 
to  vote  at  an  election.  The  feeling  of  obligation  is  undoubtedly 
strongest,  other  moments  being  neglected,  when  the  vote  is 
thought  to  have  almost  certainly  an  effect  upon  the  election. 
As  that  certainty  decreases,  the  feeling  of  obligation,  in  so  far 
as  it  is  impersonal,  also  decreases  until  a  point  is  reached  where 
it  may  lapse  entirely.  This  situation  may  arise  in  two  ways. 
Either  there  may  be  so  many  voting  the  same  way  as  to  make 
the  individual's  vote  negligible  ;  or  else,  through  corrupt  prac 
tices,  the  effect  of  the  vote  may  be  nullified.  If  the  individual 
is  absolutely  certain  that  his  vote  wiU  be  of  no  effect,  ordinarily 
his  feeling  of  social  obligation  lapses.  It  is  quite  possible,  of 
course,  that  personal  and  narrower  group  obligations  may  still 
persist.  He  may  owe  it  to  himself  to  fulfil  his  duty.  Loyalty 
to  a  class  ideal,  or  even  to  an  abstract  principle,  as  an  object 
of  personal  worth,  may  still  influence  him,  but  his  feeling  of 
obligation  loses  that  over-individual  reference  which  charac 
terises  impersonal  obligation.  This  accounts,  it  would  seem, 
for  that  peculiar  sense  of  futility  which  the  principle  of  "  per 
formance  for  performance's  sake  "  arouses  in  us  when  it  is  justified 

1  Quoted  from  Taylor,  The  Problem  of  Conduct,  p.  265. 


362  Valuation :   its  Nature  and  Laws 

merely  by  reference  to  future  and  indirect  instrumental  values. 
The  fear  that  failure  to  do  one's  abstract  duty  completely,  even 
when  there  is  no  point  in  doing  it,  may  lead  to  undermining  of 
a  good  habit,  seems  to  be  a  form  of  unworthy  timidity.  The 
robust  conscience,  to  use  Ibsen's  phrase,  lets  the  good  habit 
take  care  of  itself.  But  the  intrinsic  value  of  the  act  at  the 
moment — its  purely  personal  value — is  another  matter. 

(b)  It  Reflects  the  Norms  and  Limits  of  Objective 
Participation  Value. 

Moral  obligation  is  intimately  related  to  moral  praise  and 
blame.  It  is  the  more  subjective  aspect  of  the  demand  pre 
supposed  in  moral  judgments.  There  must,  accordingly,  be 
a  definite  relation  between  the  degree  of  strength  or  intensity 
of  the  feeling  of  obligation  and  the  degree  of  emphasis  of  moral 
judgments  of  praise  and  blame.  We  have  already  seen  that 
the  degree  of  praise  and  blame  is  determined  by  certain  norms 
and  limits,  the  norm  and  minimum  of  participation.  How  is 
the  intensity  of  obligation  related  to  these  ? 

In  studying  the  phenomena  of  imputation  it  was  found 
that  the  norm  of  participation  represents  the  zero-point.  To  re 
main  in  the  social  niveau  is  not  meritorious.  Those  who  perform 
the  normal  duties  of  life,  who  fulfil  the  normal  expectations, 
both  as  to  the  quality  and  amount  of  disposition  displayed, 
acquire  no  merit  except  through  contrast  with  exceptional 
demoralisation.  But  when  we  turn  to  the  more  subjective 
aspect  of  the  situation,  impersonal  obligation,  we  find  that 
it  is  precisely  at  this  region  of  the  "  correct  "  that  the  feeling  of 
impersonal  obligation  is  most  intense  and  emphatic.  Moreover, 
the  intensity  of  the  obligation  falls  off  as  the  quantity  of  the 
object,  the  social  good,  approaches  the  participation  minimum, 
or  as  the  object  passes  out  of  the  region  of  the  normal  into  that 
of  the  supra-normal,  where  the  values  are  ideal  and  aspiring, 
and  the  disposition  to  sacrifice  is  beyond  the  normal.  In  these 
cases  the  obligation  to  participate  is  less  and  less  intense ;  while, 
as  the  distance  from  the  region  of  the  correct  increases,  the 
neglect  or  refusal  to  participate  appears  more  and  more 
admissible. 

The  facts  of  the  moral  life  bear  out  this  analysis,  and  the 
reasons  for  the  facts  are  not  far  to  seek.  Even  a  limited  ob 
servation  convinces  one  that  in  those  cases  where  the  objective 


Over-Individual  Values  363 

social  demand  appears  in  its  purest  form,  in  those  cases,  namely, 
where  men  of  action  and  affairs  live  the  unreflecting  life  of  their 
day  and  class,  uncomplicated  by  more  general  and  ideal  re 
flections — it  is  the  acts  which  represent  the  "  sacred  average," 
in  some  cases  of  the  race,  in  others  of  a  class,  which  mark  the 
limits  of  social  obligation  and  effort.  It  is  what  "  one  does," 
and  even  more  emphatically,  what  "  one  does  not,"  in  other 
words  the  correct,  which  constitutes  the  whole  duty  of  man. 
To  the  achievement  of  this  necessary  minimum  all  conscious 
moral  effort  is  devoted,  the  rest  being  left  to  the  control  of  obli 
gations  of  an  instinctive  kind.  Strongly  confirmatory  of  this 
view  is  the  fact  that  in  more  developed,  as  well  as  in  more  primi 
tive  communities,  the  conventional  demands  of  ceremonial 
morality  and  of  etiquette  are  easily  confused  with  the  more 
strictly  moral.  It  has  been  said  that  one's  conscience  often 
pricks  him  more  severely  for  a  faux  pas  than  for  a  sin,  and, 
while  this  is  perhaps  not  quite  true,  it  is  still  true  enough  to 
indicate  that  it  is  at  the  correct,  the  habitual,  that  the  stress  of 
the  over-individual  demand  is  chiefly  felt,  often  irrespective  of 
the  question  of  the  ends  which  the  social  demand  subserves. 

The  reason  for  these  characteristics  of  impersonal  obligation 
is  to  be  found  in  the  fact  that,  in  the  main,  the  correct  or  the  par 
ticipation  norm  represents  the  indispensable  minimum  of  social 
participation  and  cohesion.  Its  value  is  instrumental  and  not 
intrinsic,  and  consequently  increase  of  the  disposition  above  the 
amount  demanded,  while  it  has  value,  becomes  progressively 
less  and  less  significant.  It  is  for  this  reason  also  that  law,  with 
its  legal  norms,  is  almost  wholly  concerned  with  giving  additional 
sanctions  to  the  average  normal  duties,  with  the  preservation  of 
the  indispensable  minimum  of  moral  or  altruistic  dispositions. 
The  objective  value  of  this  minimum,  experience  has  shown; 
but  it  would  be  impossible  to  predict  what  changes  in  value 
would  follow  the  enforcement  by  law  of  more  personal  and  in 
dividual  obligations.  We  may  conclude  then  that  moral  obli 
gation,  as  well  as  the  moral  judgments  of  praise  and  blame, 
reflects  actual  participation  values  as  determined  by  the  laws  of 
social  synergy. 

(c)  Moral  Obligation  is  Relative — not  Absolute. 

But  with  this  conclusion  we  find  ourselves,  it  cannot  be  denied, 
in  apparent  conflict  with  the  supposed  "  moral  sense,"  and  with 


364  Valuation :   its  Nature  and  Laws 

much  of  ethical  theory.  Moral  obligation  often  seems  to  reflect 
in  terms  of  feeling  an  absolutely  impersonal,  over-individual 
law,  universal  and  necessary.  Such  is  Kant's  categorical  im 
perative,  which  he  thinks  he  finds  attached  to  attitudes  and  dis 
positions  demanding  complete  universalisation,  not  realising, 
apparently,  that  the  condition  of  this  demand  being  felt  at  all 
is  precisely  the  lack  of  universalisation.  Clearly  in  such  a 
situation  one  of  two  things  must  be  true.  Either  our  entire 
theory  of  moral  values,  and  the  analysis  of  facts  upon  which  it 
rests,  must  be  at  fault,  or  else  this  supposed  deliverance  of  the 
moral  sense,  and  with  it  the  theory  based  thereon,  must  represent 
some  distortion  of  the  worth  consciousness,  some  misinterpre 
tation  of  the  facts  which  a  closer  analysis  will  discover. 

We  shall  scarcely  be  open  to  criticism  if,  after  that 
which  has  gone  before,  we  choose  the  second  alternative. 
Nor  is  it  difficult  to  show  that  we  are  here  concerned  with  a 
misinterpretation,  and  to  point  out  wherein  it  consists.  The 
moral  values,  as  distinguished  from  the  ethical  and  quasi- 
ethical,  belong  to  that  innermost  group  of  dispositions  described 
as  normal,  representing  social  habit,  i.e.,  the  maximum  of  ex 
pansion  consistent  with  the  minimum  of  intensity  necessary  to 
felt  impulsion.  As  social  habit,  the  feeling  of  obligation  is  not 
the  reflection  of  an  absolute  impersonal  law,  but  of  a  concrete 
social  synergy,  and  of  actual  objective  values.  As  such  the 
demand  is  limited  in  intensity  and  extension.  How,  then, 
does  the  concept  of  absolute,  or  perfect  and  unconditional, 
obligation  arise — that  is,  the  idea  of  a  demand  for  every  in 
dividual  and  for  an  individual  under  every  circumstance  ? 

The  idea  of  perfect,  in  the  sense  of  universal,  obligation 
rests  upon  an  abstract  ideal  construction.  A  certain  minimum 
of  a  given  disposition  is  demanded  of  all  the  participants  in  a 
given  social  group,  for  which  it  has  normal  participation  value. 
An  individual  who  feels  this  demand  idealises  it ;  he  assumes 
that,  were  it  universalised,  it  would  still  have  the  same  or  greater 
value.  In  this  assumption  he  is  further  influenced  by  the  fact 
that  for  these  fundamental  normal  attitudes  there  are  certain 
class  names,  such  as  the  cardinal  virtues,  which,  although  they 
include  under  them  in  the  course  of  time  greatly  varying  atti 
tudes,  nevertheless,  create  the  illusion  of  permanence  and  uni 
versality.  Through  the  working  of  the  laws  of  social  value 
movement  there  may  be,  as  in  the  case  of  the  change  in  meaning 
of  the  virtues  of  courage  and  chastity  from  Greek  to  Christian 


Over- Individual  Values  365 

times,  actual  change  in  attitude  with  relative  permanence  of 
name.1 

The  illusion  of  perfect,  in  the  sense  of  unconditional,  obli 
gation  arises  in  a  somewhat  similar  way.  The  act  which  has 
value  with  reference  to  a  certain  specific  end,  and  in  certain 
circumstances,  is  assumed  to  be  applicable  under  all  circum 
stances,  and  the  disposition  expressed  by  the  act  to  be  of  value 
in  any  amount.  Such  an  assumption  with  regard  to  personal 
values  which  are  intrinsic  is,  we  have  seen,  in  a  sense  justi 
fiable,  but  with  reference  to  instrumental  values  it  is  illusory. 

1  See  above,  p.  335. 


CHAPTER   XIII 
SYNTHETIC  PREFERENCE 

I.  THE  RELATIVE  VALUE  OF  DIFFERENT  CLASSES  OF 
WORTH  OBJECTS 

THERE   still  remains  a  chapter   in   our  investigation  of   these 
judgments  without  which  the  preceding  studies  would  be  fla 
grantly  incomplete,  a  chapter  which  is  in  a  sense  both  the  com 
pletion  and  the  test  of  our  preceding  analyses.     We  have  sought 
to  isolate  the  various  objects  and  standpoints  in  valuation,  to 
study  their  laws,  their  norms,  and  their  limits,  in  the  light  of 
their  psychological  genesis  and  of  the  character  of  their  psycho 
logical  presuppositions.    But  we  have  continually  recognised  that 
there  is  an  artificial  and  abstract  element  in  this  procedure, 
that   any  given  concrete   act  of  valuation  may  represent   the 
resultant  of  various  motives,  and  that,  moreover,  many  of  these 
concrete  acts  of  valuation  consist  of  preferences,  not  only  be 
tween  objects  of  the  same  class,  on  the  same  level  of  valuation, 
but  between  objects  on  different  levels.     We  have  analysed  the 
conditions  which  determine  the  relative  values  of  objects  within 
the  same  general  group,  but  it  is  not  yet  clear  that  we  can  infer 
without  further  study  the  laws  which  determine  the  relative 
values  of  the  different  groups.     To  this  question  we  must  now 
turn. 

Some  steps  in  this  direction  we  have  already  taken.  We 
have  analysed  the  fundamental  appreciative  distinctions  be 
tween  condition,  personal,  and  over-individual  impersonal 
values,  and  have  sought  to  account  for  these  distinctions  in 
terms  of  the  acquired  presuppositions  of  the  feelings.  Each 
of  these  classes  represents  a  meaning  acquired  in  some  process 
and  determined  by  certain  characteristic  presuppositions.  As 
instruments  of  analysis  we  made  use  of  the  concepts  of  the 
threshold,  and  of  capacity  for  continuous  valuation,  as  determined 

366 


Synthetic  Preference  367 

by  the  laws  of  valuation,  seeking  to  determine  these  laws  for 
the  different  levels.  As  a  result  we  found  that,  in  general, 
personal  values  represent  a  higher  level  of  meaning  than  con 
dition  worths,  and  over-individual  values  a  higher  level  than 
either.  The  question  now  arises  whether  the  acts  of  valuation 
which  consist  of  preferences  among  these  different  worth  objects 
can  be  explained  as  resultants  of  the  facts  and  laws  already 
developed,  or  whether  new  principles  must  be  called  in  to 
account  for  them.  This  may  be  described  as  the  problem  of 
Synthetic  Preference,  in  contrast  to  the  laws  of  analytic  prefer 
ence  developed  for  the  several  levels  of  valuation. 

The  characteristic  of  all  such  acts  of  preference  is  sacrifice. 
What  Bohm  Bauwerk  has  said  of  economic  goods  is,  mutatis 
mutandis,  and,  with  certain  limitations,  true  of  all  worth  ob 
jects.  "  We  find  occasion,"  he  says,  "  to  pass  judgments  of 
value  only  under  two  conditions  :  first,  when  it  is  a  matter  of 
letting  a  good  pass  out  of  our  possession  by  gift,  exchange,  or 
use  ;  and,  secondly,  when  it  is  a  question  of  adding  a  good  to 
our  possessions."  This  is  true  in  all  those  cases  where  the  judg 
ment  of  value  takes  place  after  conflict  of  motives,  although, 
as  we  have  seen,  there  are  certain  types  of  judgments  which 
merely  register  feelings  of  value  having  as  their  presuppo 
sitions  the  habits  or  implicit  assumptions  which  follow  upon 
adjustment.  But  in  all  cases  of  conflict,  estimation  of  value 
takes  place  in  terms  of  two  variables,  a  positive  and  a  negative 
factor.  The  question  of  preference  and  sacrifice  is  simple  enough, 
at  least  in  principle,  within  the  separate  spheres  of  values.  It 
is  simply  a  matter  of  more  or  less,  and  the  judgment  is  analytical. 
By  this  is  meant  that  the  judgment  or  preference  is  the  result 
merely  of  the  distinction,  according  to  their  degree,  between 
objects  or  qualities  within  the  same  general  class  of  values.  Thus 
in  the  sphere  of  condition  worths,  the  relative  value  of  different 
objects,  and  quantities  of  objects,  is  a  function  of  the  specific 
laws  of  valuation  in  this  sphere,  of  the  capacity  of  the  object 
immediately  to  satisfy  desire,  or  to  acquire  complementary 
values  through  rearrangement  and  association  with  other 
objects.  Similarly  in  the  other  spheres  of  personal  and  social 
values,  definite  and  specific  empirical  laws  determine  the  relative 
value  of  different  objects,  of  acts  and  dispositions.  When,  how 
ever,  it  becomes  a  question  of  sacrifice  of  an  object  of  one  type 
for  an  object  of  another  qualitatively  different  type,  the  problem 
is  more  complicated. 


368  Valuation  :   its  Nature  and  Laws 


II.  RATIONALISTIC  AND  MONISTIC  THEORIES  OF  PREFERENCE 
AND  SACRIFICE— CRITICISM — VOLUNTARISM  AND  SCEP 
TICISM 

The  history  of  ethical  theory  is  full  of  attempts  to  explain 
these  facts  of  preference  and  sacrifice  by  reference  to  some 
monistic  principle,  some  single  conception  of  end  or  good  out 
of  which  all  relative  values  spring,  and  to  which  they  may  be 
referred.  Now  with  these  conceptions  as  metaphysical  theories 
of  the  ideal  end  we  have  in  this  connection  no  concern.  It  is 
conceivable  that  a  philosophy  of  values  might  be  able  to  formu 
late  a  concept  of  a  single  ultimate  ideal,  the  mere  thinking  of 
which  would  include  its  obligatory  character,  its  normative 
objectivity,  in  the  sense  that  the  conception  of  "  the  most  perfect 
Being  "  was  said  to  include  its  existence.  But  it  does  not  neces 
sarily  follow  that  from  this  logical  obligation,  if  we  may  so  call 
it,  the  actual  felt  demands  can  be  deduced.  We  must  avoid 
carefully  the  confusion  between  obligation  as  a  logical  category 
and  as  an  actual  experience.  It  is  with  the  latter  phenomenon, 
with  the  empirical  facts  of  obligation  and  imputation  in  the 
more  complicated  form  of  conflicts  between  different  spheres 
of  values,  that  we  are  here  concerned,  and  our  interest  in  the 
different  theories  of  a  single  end  is  in  this  connection  confined 
to  the  one  question  :  to  what  extent  do  the  single  ideals  actually 
maintain  themselves  in  practical  situations,  to  what  extent  can 
our  actual  feelings  of  obligation  be  shown  to  refer  to  a  single 
conscious  end  ?  Should  they  turn  out  to  do  so,  the  idea  of  a 
science  of  ethics  would  so  far  find  confirmation.  But,  on  the  other 
hand,  even  if  a  single  conscious  end  should  be  shown  to  be  em 
pirically  untenable,  it  does  not  follow  that  a  science  of  ethics 
is  impossible,  nor  indeed  that  it  is  impossible  to  show  a  functional 
unity  in  the  processes  of  valuation.  To  this  more  ultimate 
question  we  shall  return,  but  for  the  present  we  must  examine 
these  theories  of  preference  in  detail. 

Viewed  from  the  empirical  standpoint,  as  hypotheses  for 
interpreting  the  concrete  facts  of  preference,  the  monistic  theories 
present  difficulties,  for  in  every  case  they  are  the  products  of 
abstraction  in  which  one  conceptual  construct,  developed  in 
the  course  of  the  concrete  processes  of  valuation,  is  abstracted 
from  these  concrete  processes,  and  taken  as  an  equivalent  for 
all  types  of  values.  Thus  Hedonism,  in  all  its  forms,  takes 


Synthetic  Preference  369 

the  abstract  equivalent  for  condition  worths,  quantities  of 
pleasure,  and  seeks  to  reduce  all  affective-volitional  meanings 
to  these  terms.  All  preferences  are  reduced  to  choices  between 
quantities  of  pleasure.  The  self-realisation  hypothesis  does  the 
same  thing  with  the  fundamental  concept  of  the  personal  level, 
striving  in  vain  to  show  a  reference  to  the  "  self  "  as  the  pre 
supposition  of  all  acts  of  preference.  Still  another  theory  finds 
the  solution  in  the  reduction  of  all  condition  and  personal  worths 
to  abstract  impersonal  ends. 

But  all  these  hypotheses  have  shown  themselves  more  and 
more  incapable  of  explaining  the  manifold  and  complicated 
phenomena  of  synthetic  preference — i.e.,  it  seems  impossible  to 
reduce  all  preferences,  all  resolutions  of  conflicts,  to  the  conscious 
introduction  of  any  one  of  these  ideals  as  norms ;  and  the  pro 
blem  has  accordingly  been  brought  to  ahead  by  the  absolute  denial, 
in  recent  discussions,  of  any  monistic  and  rational  principle  in 
terms  of  which  these  preferences  may  be  explained.  This 
denial  has  found  expression  in  two  forms :  in  the  revival  of  in- 
tuitionism,  of  which  Schwartz's  voluntaristic  type  is  a  good 
example  ;  and,  on  the  other  hand,  in  a  thorough-going  scepticism, 
of  which  Simmers  is  the  most  conspicuous  illustration. 

The  essential  point  in  Schwartz's  position  J  is  that,  while 
the  preference  which  takes  place  in  the  different  spheres  of 
condition,  personal  and  over-individual  worth  objects  is  analyti 
cal,  and  may  be  seen  to  be  determined  by  the  empirical  laws  of 
valuation,  preference  between  these  different  groups  can  be 
understood  only  as  synthetic,  as  due  to  an  immediate  unanalys 
able  judgment  that  personal  worths  shall  be  preferred  to 
condition  worths,  and  over-individual  to  personal  and  con 
dition  worths.  This  law  is  absolute  and  is  not  reducible  to 
any  empirical  laws  growing  out  of  the  genesis  and  character  of 
worth  objects.  Leaving  out  of  account  the  failure  of  this  law 
as  a  description  of  fact,  a  point  which  we  shall  consider  in  its 
proper  place,  it  is  sufficient  to  note  in  this  connection  that  such 
a  method  of  solving  the  problem,  by  magnifying  relative  ap 
preciative  distinctions  into  eternal  principles,  implies  a  failure 
to  recognise  the  essential  nature  of  these  distinctions  as  genetic 
and  relative. 

Scepticism  is  always  a  near  neighbour  to  intuitionism,  and 
the  scepticism  in  this  case  takes  the  form  of  denying  the  ex- 

1  Schwartz,  Psychologic  des  Willcns,  zur  Grundlegtntg  der  Ethik,  Part  II, 
chaps.  I  and  II. 

2  B 


37O  Valuation:   its  Nature  and  Laws 

istence  of  any  single  law  to  which  these  preferences  may  be 
referred.  The  two  monistic  conceptions  which  have  especially 
called  out  this  sceptical  attitude  are  the  concept  of  an  extra- 
experiential  or  metaphysical  Self,  of  which  these  different  worth 
objects  represent  different  stages  of  realisation,  and  that  of  hedon 
istic  utilitarianism  which  conceives  these  different  worth  objects, 
with  their  differences  for  appreciation,  as  reducible  to  a  common 
equivalent,  quantity  of  pleasure. 

The  inadequacy  both  of  the  Self-realisation  and  of  the  Hedon 
istic  hypothesis  is  shown  by  Simmel's  criticism  of  the  two 
concepts.1  The  keynote  of  his  criticism  of  the  unity  of  the 
personality  as  the  supreme  norm  of  worth  experience  is  to  be 
found  in  the  fact  that  he  sees  in  it  a  transference  of  the  supreme 
logical  category  of  epistemology  to  the  sphere  of  worth  experi 
ences  where  it  does  not  necessarily  apply.  To  infer  the  actual 
unity  of  the  ends  of  volition  from  the  logical  unity  of  the  sub 
ject  of  knowledge  is  to  go  beyond  the  worth  experience  itself. 
We  do  indeed  find  the  empirical  unity  of  the  subject,  or  self,  an 
object  of  desire  and  of  worth  judgment,  and,  within  limits, 
a  standard  of  values,  just  as  we  find  individuation  a  motive  in 
the  construction  of  worth  objects,  but  the  individual,  and  the 
processes  of  individuation,  are  empirical  and  not  logical.2  In 
like  manner  the  concept  of  pleasure  is  an  abstraction  from  our 
experiences  of  feeling  of  value,  an  abstraction  which  ignores  the 
appreciative  distinctions  in  feeling  to  such  a  degree  that  it 
no  longer  includes  the  appreciative  differences  of  depth  and 
breadth,  and  the  personal  and  impersonal  meanings  of  the 
feelings.  It  is  a  workable  equivalent  only  for  what  have  been 
described  as  condition  worths. 

It  is  undoubtedly  true  that  neither  of  these  conceptions 
will  establish  continuity  between  the  different  spheres  of  values. 
And  the  reason  is  that  in  both  cases  the  standards  of  a  par 
ticular  type  of  worth  judgment  are  abstracted  from  the  processes 
in  which  they  are  constructed,  and  applied  as  norms  to  a  differ 
ent  type  of  activity.  Within  a  certain  class  of  relative  judg 
ments,  concerned  with  condition  worths,  pleasure  is  a  well- 
founded  object  of  desire  and  quantity  of  pleasure  a  well-founded 

1  Simmel,  Einleitung  in  die  Moralwissenschaft,  Book  I,  chap.  IV,  Book  II,  chap.  VI. 

2  An  instructive  criticism  of  the  first  of  these  monistic  principles  is  to  be  found 
in   Taylor's   Problem  of   Conduct,    where  it   is  shown    at  what   points   actual   worth 
experiences   prove   refractory   to  such  reduction.      The  great  single  moments  of  ex 
perience,  both  of  supreme  assertion  and  supreme  negation,  do  not  lend  themselves  easily 
to  this  conception  of  Self-realisation. 


Synthetic  Preference  371 

standard  of  judgment.  The  ideal  of  unity  of  the  personality  is 
also  undoubtedly  the  presupposition  of  an  entire  group  of  worth 
judgments — of  personal  obligation  and  of  imputation  of  personal 
worth — and  as  such  constitutes,  within  those  limits,  a  well-founded 
object  of  desire  and  a  standard  of  judgment.  The  ideal  of  the 
sacrifice  of  the  subjective  and  personal,  it  is  equally  obvious, 
is  a  well-founded  ideal  within  the  limits  of  strictly  moral  judg 
ment.  But  no  single  one  of  these  conceptions  has  been  able  to 
supply  a  satisfactory  standard  for  synthetic  preference,  and 
scepticism  as  to  the  existence  of  such  a  standard  has  resulted. 
The  only  way  open  to  us,  then,  is  to  take  these  pre-scientific 
distinctions  between  condition,  personal,  and  over-individual 
values  as  heuristic  conceptions,  and,  by  empirical  analysis  of 
actual  preferences — the  conflicts  between  obligations  on  these 
different  levels  and  between  personal  and  impersonal  imputation, 
to  determine  to  what  extent  these  preferences  show  uni 
formities,  to  what  extent  higher  unities  and  continuities  of 
preference  may  be  established.  If  any  such  higher  laws  of 
preference  emerge,  it  may  be  expected  that  they  will  be  more 
general  expressions  of  the  empirical  laws  already  found  operative 
in  the  different  spheres. 

III.  ANALYSIS  OF  THE  FACTS  OF  SYNTHETIC  PREFERENCE  AS 
EXHIBITED  IN  JUDGMENTS  OF  OBLIGATION  AND  IMPU 
TATION 

The  point  at  which  Simmers  scepticism  becomes  most 
incisive  is  in  his  discussion,  in  the  last  chapter  of  his  work,  of 
the  conflicts  of  duty.  It  is  just  the  impossibility  of  accounting 
for  the  actual  resolutions  of  the  conflicts  between  our  different 
obligations  in  terms  of  any  single  highest  end,  any  monistic 
principle  of  valuation,  that  leads  to  the  denial  of  such  a  principle. 
In  particular  he  insists  that  we  have  no  standard  in  terms  of 
which  we  may  estimate  the  relative  importance  of  the  extensive 
and  intensive  aspects  of  obligation,  i.e.,  judge  between  the  duty 
which  is  more  intensive  because  more  personal  and  emotional, 
and  that  which  is  less  intensive  but  more  extensive,  because  more 
general  and  over-individual  in  its  reference.  Preference  be 
tween  an  extensive  duty  with  weak  intensity  and  a  narrower 
duty  with  marked  intensity  is,  he  maintains,  determined  by 
forces  in  the  darker  life  of  feeling  which  cannot  be  adequately 
expressed  in  terms  of  our  knowing  consciousness.  In  this 


372  Valuation:   its  Nature  and  Laws 

statement  of  the  difficulty  we  have  the  problem  of  our  own 
study  presented  in  our  own  terms,  that  is,  what  deter 
mines  our  choice  between  ideal  objects  of  condition,  personal, 
and  over-individual  worth,  where  degrees  of  obligation  can  be 
reduced  neither  to  differences  in  degree  of  intensity  of  feeling 
nor  to  the  explicitness  of  reference  of  the  object  to  the  self.  And 
the  same  difficulty  inheres  in  our  attempts  to  rationalise  our 
judgments  of  imputed  worth. 

Let  us,  then,  begin  our  study  by  returning  to  that  other 
form  of  denial  of  a  monistic  principle  of  continuity,  the  volun- 
taristic  intuitionism  of  Schwartz,  according  to  which  our  con 
sciousness  of  value,  as  expressed  in  feelings  of  obligation,  always 
demands  the  sacrifice  of  condition  to  personal  and  over-individual 
values,  and  the  sacrifice  of  both  condition  and  personal  to  over- 
individual  values.  Some  facts  of  experience  must  have  motived 
this  formulation,  otherwise  it  would  not  have  the  element  of 
truth  that  it  certainly  has.  For  it  contains  at  least  this  much 
truth — that  it  constitutes  a  broad  generalisation  of  the  facts.  It 
is  one  of  those  pre-scientific  formulations  which  precede  more 
detailed  analysis.  More  than  this,  it  corresponds  to  what  we 
have  found  to  be  the  genetic  levels  of  meaning.  In  general, 
normal  over-individual  values  have  greater  transgredient  refer 
ence  than  normal  personal  values,  and  personal  values  greater 
than  condition  worths.  In  general,  also,  we  find  this  fact 
reflected  in  our  feelings  of  obligation. 

But  when  we  seek  to  carry  out  this  generalisation  in  detail, 
we  find  that  it  is  by  no  means  an  accurate  picture  of  the  facts. 
If  it  were  so  it  would  mean,  as  Schwartz  admits,  that  our  sense  of 
obligation  would  demand  the  sacrifice  of  the  most  important 
personal  to  the  least  important  over-individual  good,  and  the 
greatest  condition  worth  to  the  smallest  personal  worth,  a  con 
sequence  which  the  analysis  of  actual  judgments  by  no  means 
bears  out.  It  is,  therefore,  precisely  at  this  point  of  the  limits 
that  the  critical  question  arises.  An  analysis  of  how  our  sense 
of  value  reacts  at  these  limits  or  points  of  conflict  of  obli 
gations,  will  not  only  have  the  negative  result  of  showing  the 
points  where  this  intuitional  formula  breaks  down,  but  will  also 
have  the  positive  result  of  disclosing  the  empirical  principles 
which  determine  the  actual  resolutions  of  such  conflicts. 


Synthetic  Preference  373 


i.  Conflicts  between  Personal  and  Impersonal  Obligations. 

Especially  enlightening  for  our  study  is  a  striking  illus 
tration  of  this  intuitionist  point  of  view  from  the  pen  of  Tolstoi.1 
With  his  customary  fondness  for  intense  and  clear-cut  assertions, 
he  raises  the  question  of  the  limits  of  self-sacrifice,  and  answers 
it  by  affirming  that  our  conscience  recognises  no  limit.  He 
presents  us  with  a  picture  of  beggars  coming  one  after  the  other 
to  the  house  of  the  moralist,  and  receiving  all  his  money,  food, 
and  shelter.  At  last  there  appears  a  disreputable  tramp,  who 
asks  for  the  moralist's  last  bundle  of  straw,  the  final  barrier 
between  him  and  the  certainty  of  death.  Should  the  moralist 
give  it  ?  Yes,  not  only  share  it,  but  give  it  completely.  To  all 
questions  of  limits  or  compromise  Tolstoi  answers  with  a  de 
termined  "  No  !  "  Even  the  minimum  of  existence,  and  all  the 
personal  worths  which  it  involves,  must  be  sacrificed,  be  the 
consequences  what  they  may.  There  can  be  no  doubt  that  there 
are  many  who  would  honestly  feel  no  such  obligation,  and  yet 
this  is  not  the  extremest  form  in  which  the  demand  might  be 
stated.  Conceivably  the  last  bundle  of  straw  might  be  asked 
merely  for  some  minor  purpose  of  the  beggar— not  for  the  pro 
tection  of  his  life  and  health.  Or  it  might  be  asked  as  a  charity 
for  one  who  is  not  only  far  removed  from  the  immediate  sym 
pathies  of  the  giver,  but  one  with  whom  he  would  never  come  in 
contact,  and  yet  the  demand  would  be  the  same  in  principle. 

Assuming  that  for  some,  as,  for  instance,  Tolstoi  himself, 
this  would  be  a  real  obligation,  for  many  others  it  certainly 
would  not.  And  if  this  be  granted,  it  would  appear  that,  not 
only  does  the  a  priori  principle  itself  fail  of  universality,  but  that 
this  fundamental  difference  in  worth  feeling,  at  such  a  point, 
affords  some  basis  for  that  scepticism  which  denies  the  possi 
bility  of  our  finding  any  explanation  for  preferences  of  this  sort. 
The  first  inference  we  may  admit,  but  the  second  only  in  case 
the  laws  which  we  have  already  developed  fail  to  give  us  any 
clues  to  the  explanation. 

In  attempting  to  find  these  clues  we  must  first  recall  the 
fact,  frequently  insisted  upon,  that  any  concrete  practical 
attitude  of  obligation  or  imputation  is  a  complex,  the  total 
force  of  which  is  analysable  into  several  motives.  The  sense 

Tolstoi,  Acte  der  Selbst-Opferung,  Wiener  Rundschau,  October  I,  1899.     Quoted 
from  Kreibig,  Psychologische  Grundlegung  ernes  Systems  der  Wert-theorie,  p.  152. 


374  Valuation:   its  Nature  and  Laws 

of  worth  which  Tolstoi  expresses  in  this  extreme  judgment  of 
obligation  may  very  well  represent,  not  only  his  sense  of  the 
moral  or  participation  value  of  his  act,  the  over-individual 
value,  and  therefore  obligation,  but  also  a  sense  of  its  personal 
worth,  and  therefore  personal  obligation  for  him.  We  should 
in  such  a  case  have,  not  a  pure  over-individual  demand  as  the 
determinant  of  the  judgment,  but  such  a  demand  as  modified  by 
complementary  personal  worths.  Such  a  preference  would  then 
be  at  least  partially  explicable  upon  our  principles. 

We  may  best  approach  this  question  by  taking  the  other 
side  of  the  apparent  antinomy,  that  of  the  man  who  would,  not 
feel  such  obligation.  This  we  should  have  less  difficulty  in 
understanding  psychologically.  In  the  extreme  case,  as  the 
situation  was  presented — where  the  beggar  is  far  away  and 
the  object  which  means  life  or  death  to  the  giver  is  of  only 
secondary  importance  to  the  recipient,  the  participation  value 
of  the  act  is  so  remote  and  indirect  that  it  is  practically  below 
the  threshold  of  worth  feeling.  It  is  below  the  participation- 
minimum,  as  we  have  defined  it.1  On  the  other  hand,  the 
value  of  the  object  for  the  individual  sacrificing — since  it  is 
the  minimum  of  existence  and  constitutes  the  indispensable  pre 
supposition  of  other  acquired  values,  is  practically  absolute. 
If,  then,  the  laws  of  "  condition  "  and  of  over-individual,  social 
values  were  alone  operative,  there  can  be  no  question  which 
way  the  preference  would  go.  In  that  case,  too,  where  the 
bundle  of  straw  is  also  the  minimum  of  existence  for  the  beggar, 
the  obligation  would  still  not  be  felt  by  many,  and  with  perfect 
honesty,  because  of  the  remoteness  of  the  beggar  from  their 
sympathy,  and  because  of  the  absolute  value  of  the  object 
sacrificed.  In  this  case,  as  well  as  in  the  first,  the  value  of  the 
over-individual  object  is  far  below  the  minimum  of  participation, 
the  amount  of  the  over-individual  good  for  which  the  sacrifice 
is  normally  demanded,  while  the  sacrifice  here  demanded  is  far 
above  the  maximum  of  altruism  expected  in  the  normal  working 
of  over-individual  demand.  The  judgment  of  the  man  who 
honestly  does  not  feel  the  obligation  to  such  extreme  sacrifice 
seems  to  be  in  accord  with  the  normal  laws  of  valuation. 

Does  it  not  then  seem  that  such  a  preference  as  Tolstoi's, 
such  a  feeling  of  obligation,  must  be  irrational,  or  at  least  supra- 
rational,  in  the  sense  that  it  transcends  all  the  empirical  laws 
of  worth  feeling  ?  Some  extreme  voluntarists  have  so  held, 

1  Chap,  xi  ,  pp.  357  f. 


Synthetic  Preference  375 

finding  in  self-denial  a  mystery  which  transcends  all  empirical 
explanation.  But  this  seems  as  hasty  an  inference  as  is  the 
claim  of  the  extreme  rationalist  to  explain  these  experiences  in 
terms  of  a  conscious  rational  ideal.  It  is  just  cases  like  these 
which  give  us  the  clue  to  our  empirical  analysis.  We  have 
already  seen  in  an  earlier  chapter l  how  altruism,  as  measured 
in  terms  of  sacrifice  of  condition  and  personal  for  over-individual 
worths,  may  itself  become  a  personal  worth,  and,  as  such,  is 
indefinitely  Steigemngsfahig,  susceptible  of  indefinite  increase, 
when  as  an  impersonal  worth  it  is  not.  It  is  quite  possible  that 
the  supreme  worth  of  tragical  elevation  enters  as  a  determining 
moment  in  such  an  act  of  preference.  In  fact  it  would  seem  that 
Tolstoi  is  himself  a  case  very  much  in  point.  James,  in  his 
study  of  Tolstoi's  conversion  to  his  extreme  altruism  with  its 
accompanying  religious  feeling,  lays  stress  upon  the  preceding 
moment  of  satiety  and  loss  of  value  upon  the  part  of  other 
objects.  "  A  well-marked  case  of  anhedonia,"  he  calls  it, 
"  a  passive  loss  of  appetite  for  all  life's  values."  2  We  have, 
then,  in  the  absolute  intrinsic  value  ascribed  to  altruism,  ap 
parently,  merely  the  substitution  of  one  personal  worth  for 
another,  and  not  really  the  sacrifice  of  all  condition  and  personal 
worths  to  the  smallest  over-individual  worth.  If  Tolstoi  rea 
lised  the  relative  participation  value  of  such  acts,  i.e.,  felt  their 
demand  merely  in  the  degree  which  would  follow  upon  the 
normal  working  of  the  ordinary  laws  of  sympathetic  participa 
tion,  there  would  be  a  limit  to  the  over-individual  obligation 
to  sacrifice. 

(c)  Confirmation  of  this  Explanation  by  the  Judgments  of  the 
"Impartial  Spectator." 

When  we  turn  to  the  judgments  of  the  impartial  spectator 
upon  such  acts  of  sacrifice,  we  find  this  view  substantiated. 
There  is  a  fundamental  difference  between  what  we  may  demand 
and  what  we  may  admire  in  these  matters.  The  ordinary  man 
would  say,  I  suppose,— if  he  could  become  articulate  and  if  he 
would  consent  to  use  the  terminology  of  the  present  discussion, 
'  I  cannot  help  imputing  emotionally  absolute  worth  to  such 
a  personality  when  he  thus,  by  an  act  of  extreme  self-sacrifice 
for  an  insignificant  over-individual  worth,  displays  such  strength 
and  singleness  of  disposition,  but  I  am  at  the  same  time  unable, 

1  Chap,  x,  p.  301.  2  James,   Varieties  of  Religious  Experience^  p.  149. 


376  Valuation:   its  Nature  and  Laws 

when  I  take  the  impersonal  point  of  view  of  moral  judgment 
upon  the  act,  on  the  basis  of  its  participation  value,  to  demand 
it  of  every  person.     Nor  is  the  moral  value  which  I  impute  to 
it  in  any  sense  proportional  to  the  strength  of  the  disposition 
displayed.     For    when    I    take    the    impersonal   standpoint    in 
judgment,  I  become  aware  of  the  fact  that  such  absolute  self- 
sacrifice  means  the  sacrifice,  not  only  of  condition  and  personal 
worths,    but    of    other    over-individual   values    of   which    the 
existence  and  self-realisation  of  the  individual  in  question  are 
presuppositions.     The  individual  is  the  meeting-point  of  various 
over-individual  demands  ;    in  him  inhere  various  group  worths, 
of  family,  state,  knowledge,  art,  etc."     It  would  seem,  then, 
that  in  his  judgment  of  the  participation  value  of  an  act,  the 
impartial   spectator   cannot    avoid   taking   into   account   what 
may  be   described   as   the  "  personality-coefficient  "  of  the  in 
dividual  sacrificing  and  of  the  individual  for  whom  the  sacrifice 
is  made,  that  is,  the  relative  significance  of  the  system  of  values 
for  which  each  stands.     But  when  this  is  once  admitted,  it  is 
clear  that  the  a  priori  principle  of  preference   enunciated  by 
Tolstoi  is  not  reflected  in  our  judgments  of  imputation. 

2.  Conflicts  of  Condition  and  Personal  Worths. 

The  actual  resolutions  of  the  conflicts  between  personal 
and  condition  worths  seem  equally  refractory  to  subsumption 
under  one  general  principle.  The  proposition  that  all  condition 
worths  should  be  sacrificed  to  personal  worths  also  falls  short  of 
being  an  accurate  picture  of  our  real  feelings  of  obligation. 
That  it  describes  in  a  general  way  our  feeling  of  the  relative 
value  of  the  two  classes  of  objects  is  beyond  doubt,  but  here 
again,  when  we  examine  the  limiting  cases,  difficulties  arise. 

In  order  to  understand  the  situation  properly,  we  must  recall 
the  fact  that  there  are  instinctive  obligations  on  the  level  of 
simple  appreciation  or  simple  condition  worths,  prior  to  activities 
of  characterisation,  and  that  it  is  in  the  region  of  the  existence 
minimum  that  these  obligations  arise.  It  is  a  question,  there 
fore,  of  preference  between  two  kinds  of  obligation,  not  between 
mere  desire  unqualified  by  obligation  and  the  feeling  of  obliga 
tion,  and  here  again  we  may  be  disposed  to  say  with  Simmel  that 
the  determining  forces  lie  beyond  the  ken  of  our  knowing  con 
sciousness. 

But  to  consider  first  the  facts  themselves,  it  seems  fairly 


Synthetic  Preference  377 

clear  that  objects  of  condition  worth,  in  so  far  as  they  are  of 
relative  and  instrumental  value,  are  normally  felt  to  be  of  less 
value  than  personal  worths;  they  should  always  be  sacrificed 
to  the  least  important  personal  worth.  They  have  more  capacity 
for  substitution  and  have  less  capacity  for  continuous  valuation 
than  personal  worths.  Such  obligation  as  may  attach  to  merely 
instrumental  and  condition  worths  is  only  indirect,  and  is  de 
pendent  upon  their  relation  to  the  necessaries  of  existence  or  to 
personal  worths.  But  when  it  comes  to  a  choice  between  the 
minimum  of  existence  and  minor  personal  worths  with  capacity 
of  substitution  (more  or  less  external  personal  worths,  such  as 
pride  in  one's  position,  name,  beauty,  etc.),  our  judgment 
is  not  wholly  unequivocal.  When  this  type  of  personal  obliga 
tions  comes  into  conflict  with  fundamental  condition  worths, 
with  obligations  which  arise  in  the  struggle  for  existence, 
we  are  disposed  to  think  that  such  personal  values  are  more  or 
less  fictitious,  or  at  least  have  capacity  of  substitution,  and 
should  therefore  be  sacrificed.  While  we  may  admire  absolute 
sacrifice  of  condition  worths  for  such  personal  worths,  in  that 
thereby  a  new  personal  worth  is  revealed,  we  cannot  demand  it 
for  the  reason  that  such  personal  worths  are  relative  and  not 
absolute.  They  are  not  without  capacity  of  substitution.  If 
they  are  lost  more  fundamental  personal  worths  may  take  their 
place.1  The  minimum  of  existence,  on  the  other  hand,  is  without 
capacity  of  substitution. 

The  real  test  of  the  formula  appears  when  the  conflict  lies 
between  absolute  personal  and  absolute  condition  worths, 
between  the  minimum  of  existence  and  the  minimum  of  charac 
terisation.  Illustrations  in  point  would  be  conflicts  between 
starvation,  or  extreme  bodily  pain,  and  honour.  Extreme  bodily 
pain,  for  example,  approaches  the  psychological  absolute  of 
supreme  evil  in  the  sphere  of  condition  worths.  Otherwise 
expressed,  in  the  moment  of  extreme  bodily  pain,  or  in  imagina 
tion  of  it  with  belief  in  its  imminence,  cessation  or  removal  of 
the  evil  has  absolute  worth,  constitutes  the  existence-minimum. 
Let  us  suppose  that  a  person,  subjected  to  cruel  torture,  has 
sacrificed,  one  after  the  other,  minor  personal  worths  which 
are  not  without  capacity  of  substitution,  until  finally  nothing 
but  the  sacrifice  of  his  honour,  it  may  be  by  the  betrayal  of 
a  comrade  or  by  obedience  to  a  demand  to  recant,  will  purchase  the 
relief  which  now  has  for  him  absolute  value.  We  expect  him 

l.  Chap,  x,  pp.  295  f. 


378  Valuation:   its  Natitre  and  Laws 

to  hold  out,  for  we  believe,  however  the  belief  may  have  been 
created,  that  such  persistence  is  possible  and  has  taken  place. 
If  the  personal  worth  triumphs  we  hail  such  triumph  as  heroic, 
and  the  worth  imputed  to  the  personality  is  absolute.  But 
suppose  it  does  not  ?  Our  reaction  in  that  case  is  not  instinctive 
and  unequivocal.  We  have  already  seen  in  our  study  of  per 
sonal  worths  a  that  at  this  point  two  reactions  are  empirically 
possible.  When  the  minimum  of  characterisation  is  sacrificed, 
either  all  personal  relation  is  abrogated,  the  person  is  beneath 
contempt,  and  our  attitude  passes  into  one  of  purely  impersonal 
moral  or  judicial  judgment,  or  else  the  attitude  of  personal 
respect  and  admiration  passes  into  one  of  profound  organic 
sympathy  and  pity  which  may  amount  to  a  sanctioning  of  his 
act.  We  may  admire  such  extremes  of  sacrifice,  but  we  do  not 
demand  them  universally.  Our  feeling  of  obligation  seems  to  be 
similarly  equivocal  in  such  situations.  There  are  some  indi 
viduals  for  whom  this  extreme  of  sacrifice  would  be  obligatory, 
others  for  whom  it  honestly  would  not  be.  The  significance  of 
these  facts  seems  to  be  that  in  the  limiting  cases  of  conflict 
between  condition  and  personal  worths  the  general  norm  of 
preference  may  break  down.  Our  judgments  are  not  unequivocal 
and  a  priori. 

IV.  CONCLUSIONS  FROM  THE  ANALYSIS  OF  SYNTHETIC 
PREFERENCE 

i.  Summary  of  Results. 

The  preceding  examination  of  the  facts  of  synthetic  preference 
leads  to  certain  conclusions,  in  the  light  of  which  it  is  possible 
to  formulate  a  theory  of  more  general  philosophical  significance. 
Returning  to  the  question  with  which  the  analysis  started, 
the  grounds  are  now  apparent  for  the  negative  conclusion  that, 
whatever  continuity  there  may  be  in  the  processes  of  valuation, 
it  cannot  be  shown  to  spring  from  the  consciousness  of  any 
single  rational  end  or  principle.  There  is  no  single  ideal  which 
shows  itself  to  be  ultimate  in  the  sense  that  it  always  consti 
tutes  the  controlling  factor  in  the  solution  of  these  conflicts. 
In  the  second  place,  it  seems  equally  certain  that  there  is  no 
universal  a  priori  law  governing  these  preferences.  While 
the  norms  and  ideals  of  the  different  genetic  levels  constitute 

1  Chap,  x,  pp.  300  f. 


Synthetic  Preference  379 

a  continuous  series  of  values,  the  appreciative  distinctions  are 
not  ultimate  in  the  sense  that  they  maintain  themselves  ab 
solutely  in  the  limiting  cases  of  conflict. 

Nevertheless,  our  study  has  not  been  wholly  barren  of  positive 
results.  Modest  as  our  insight  is  in  comparison  with  what  is 
claimed  by  those  theories  of  the  standard  which  we  have  been 
considering,  still  we  have,  up  to  a  point  at  least,  learned  some 
thing  of  the  principles  operative  in  synthetic  preference.  In  so 
far  as  these  concrete  and  individual  experiences  can  be  ration 
alised  at  all,  they  seem  to  be  resultants  of  the  laws  of  valuation 
already  developed.  They  reflect,  in  a  general  way  at  least,  the 
relative  capacities  of  the  different  objects  for  continuous  valua 
tion  and  for  substitution,  and  therefore  the  acquired  meaning 
of  the  objects.  By  the  method  of  ethical  experimentation  we 
have  employed,  in  which  it  was  sought  to  test  the  relative  capacity 
of  the  different  ideals  to  persist  in  the  face  of  arrest,  and  to  dis 
cover  the  range  and  limits  of  our  judgments  of  obligation  and 
imputation,  we  have  secured  results  which  agree  in  the  main 
with  the  laws  of  acquirement  of  value. 

But  that  these  values  are  acquired,  that  the  distinctions 
between  the  different  groups  of  worth  objects  are  acquired  and 
not  ultimate,  is  apparent  in  the  breakdown — at  the  limits— 
of  the  so-called  a  priori  law  of  preference.  At  these  points  the 
distinctions  tend  to  lapse.  In  the  case  where  the  distinctions 
are  wholly  within  the  Ego  (between  worths  of  condition  and 
person),  there  seem  to  be  supreme  moments  of  assertion  of  the 
will  where  the  distinction  disappears,  where  the  ideal  objects  and 
their  acquired  meanings  and  obligations  fall  away,  giving  place 
to  simple  immediacy.  Personal  ideals  may  acquire  absolute 
value,  that  is,  may  become  practical  absolutes  in  the  moment  of 
tragical  elevation.  But  so  may  a  supreme  necessity  of  organic 
life.  In  the  one  case  it  is  the  minimum  of  characterisation, 
in  the  other  that  of  existence,  which  has  absolute  value. 
Whichever  alternative  triumphs  in  a  conflict,  we  cannot  properly 
speak  of  a  victory  either  of  condition  or  personal  worth,  for 
the  conceptual  distinction  between  the  idea  of  pleasure  and  of 
the  self  simply  does  not  exist.  Similarly,  the  man  who,  in  the 
supreme  moments  of  preference  between  personal  and  over- 
individual  worths,  chooses  the  absolute  personal  worth,  is  no 
longer,  strictly  speaking,  an  egoist.  Nor  is  he  who  chooses 
the  over-individual  worth,  with  its  moment  of  tragical  elevation, 
an  altruist.  The  relative  distinction  between  egoism  and 


380  Valuation:   its  Nature  and  Laws 

altruism  lapses.  Again  we  are  face  to  face  with  an  absolute 
moment  of  simple  immediacy  in  which  relative  distinctions 
vanish. 

On  every  hand,  then,  we  are  forced  to  the  conclusion  that 
the  supreme  moments  of  affirmation  or  abnegation  trans 
cend  the  relative  distinctions  which  the  intellect  makes  in 
the  service  of  the  will,  and  consequently  in  part  elude  our  con 
ceptual  descriptions.  The  acquired  distinctions  and  meanings 
of  these  ideal  objects  are  appreciative  descriptions,  volitional 
norms,  formulated  in  the  interest  of  continuity  of  appreciation 
and  valuation.  This  is  the  important  point.  The  monistic 
theories  we  have  been  considering  fail  to  take  sufficiently  into 
account  the  elements  of  discontinuity  which  the  analysis  of 
value  judgments  discloses.  They  invariably  seek  to  deduce 
the  concrete  norms,  or  psychologically  determined  presuppo 
sitions  of  value  judgments,  from  logical  conceptions  of  the 
relation  of  the  individual  to  the  universal,  i.e.,  from  ultimate 
logical  presuppositions.  The  philosophy  of  supreme  moments,  to 
which  we  have  been  led  by  our  analysis,  seems  to  negative  such 
conclusions.  Rather  does  it  appear  that  the  moments  of  abso 
lute  worth  experience  may  be  equally  those  of  self-affirmation 
and  self-abnegation,  and  ultimately  moments  in  which  even  these 
intellectualistic  conceptions  lapse.  Viewed  phenomenally,  the 
activity  of  the  will  seems  to  be  an  oscillation  between  supreme 
moments  within  the  different  spheres  of  values,  and  a  principle 
which  should  explain  them  would  have  to  give  us  the  laws  of 
these  oscillations,  the  larger  concept  which  would  comprehend 
them  in  a  higher  unity. 

2.  The  Bearing  of  these  Results  upon  Larger  Conceptions. 

The  attempt  to  formulate  such  a  larger  philosophical  con 
ception  does  not  lie  within  the  province  of  this  study,  but  we 
may  offer  in  conclusion  certain  suggestions,  as  much  by  way  of 
avoiding  misunderstanding  as  for  the  sake  of  completeness. 

Throughout  this  entire  study,  including  the  present  chapter, 
the  standpoint  and  methods  of  empirical  analysis  of  valuation 
have  been  consistently  maintained.  In  so  far  as  the  monistic 
theories  of  the  ideal  are  considered,  it  is  with  the  object 
of  determining  their  function  as  conscious  ideals,  as  actual  pre 
suppositions  or  norms  of  judgment  and  feeling — not  as  attempts 
to  characterise  the  ultimate  logical  presupposition  of  valuation 


Synthetic  Preference  381 

required  by  a  philosophy  of  values.  That  such  a  philosophy 
of  values  is  possible,  or  that  the  single  ultimate  presupposition 
upon  which  valuation,  when  thus  logically  viewed  must  rest, 
can  be  characterised,  we  need  not  deny.  In  order  that  valua 
tion  shall  be  not  only  describable  in  terms  of  its  empirical  con 
ditions,  but  also  intelligible  in  the  light  of  its  ultimate  meaning, 
such  an  axiological  re-reading  of  our  judgments  of  value  in  the 
light  of  their  ultimate  logical  presuppositions  is  probably  neces 
sary.  All  that  we  have  here  been  concerned  to  show  is  that  no 
unity  and  continuity  of  conscious  ideals  and  norms  is  discover 
able,  no  single  content  has  for  feeling  unconditional  value. 

That  the  logical  presupposition  of  all  valuation  must  be  a 
single  incontestable  or  unconditional  value,  follows  from  the  logical 
unity  of  the  subject  of  the  value  judgment,  and  from  the  claim 
of  the  value  judgment  to  objectivity.     But  from  this  logical 
unity  of  the  subject  we  cannot  pass  to  the  empirical  unity  of 
conscious  ends  and  of  felt  values  ;  from  the  logical  postulate  of 
an  unconditioned  value  we  cannot  pass  to  the  unconditioned 
value  of  any  concrete  content.     Such  a  transition  is  made  im 
possible,  we  have  seen,  by  the  equivocal  character  of  the  actual 
feelings  of  obligation  and  of  intrinsic  appreciation  or  approval 
and  disapproval  in  the  limiting  cases,  and  by  the  oscillation 
between  the  ideals  of  self-realisation  and  self-abnegation.     But, 
despite  this  discontinuity  of  conscious  ideals  and  of  empirically 
derived  norms,  there  can  be  no  question  of  the  functional  unity 
and  continuity  of  valuation  itself.     Even  in  those  cases  in  which 
the  outcome  of  the  conflict  is  equivocal,  and  in  which  the  acquired 
distinctions  break  down,   the  choice  is   always   of  the  nature 
of  a  falling  back  upon  a  value  which  is  without  capacity  of 
substitution,  and  which  represents  the  necessary  presupposition  of 
continuity  of  volition  and  valuation — in  other  words,  is  a  practical 
absolute  in  our  sense  of  the  term.     It  is  conceivable,  therefore, 
that,  while  no  single  ideal  can  be  taken  to  represent  absolutely 
the  whole  end  and  meaning  of  this  continuity — for  the  reason 
that  no  conceptual   construction   can,   because  of  its  empirical 
origin,  be  universalised  and  applied    continuously,  and  while, 
consequently,    no  single  norm  embedded  in  such  ideal  can  be 
applied  in  every  empirical  situation,  it  may  still  be  possible  to 
formulate  a  metaphysical  conception  of  this  functional  unity  and 
continuity  which  shall  be  compatible  with  the  actual  plurality 
of  conscious  ends  and  of  empirical  norms  or  presuppositions. 
Accordingly — thus  it  is  argued  by  the  monistic  philosophies 


382  Valuation:    its  Nature  and  Laws 

of  value — we  cannot  dispense  with  the  concept  of  a  meta 
physical  or  met-empirical  will,  whether  personal  or  impersonal 
and  over-individual,  if  the  functional  unity  and  continuity 
of  valuation  is  to  be  intelligible,  the  realisation  of  such  a  will 
being  presupposed  in  every  realisation  of  finite  ends  and  in  every 
particular  judgment  of  value.  The  facts  here  examined  do 
not  prove  the  impossibility  of  such  logical  unity  and  continuity, 
but  merely  that  the  attempt  to  formulate  that  implicit  pre 
supposition  in  terms  of  conscious  ideals  must  remain  incom 
plete.  The  breakdown  of  the  distinctions  between  personal 
and  condition  worths,  between  egoism  and  altruism,  simply 
means  that  the  content  of  the  ideal  has  been  conceived  too 
narrowly,  and  that  out  of  these  conflicts  and  oppositions  a 
larger  ideal  arises  which  includes  these  distinctions.  The 
question  must,  therefore,  assume  this  form — not  is  there  a 
single  conscious  purpose  to  which  all  other  ends  are  subordinate, 
but  rather  is  there  not,  for  reflective  evaluation  at  least,  some 
absolute  intrinsic  value  logically  presupposed  in  these  empirical 
ends,  to  which,  as  an  ultimate  presupposition,  the  implicit 
assumptions,  the  empirically  derived  norms,  in  the  various 
judgment  situations,  may  be  logically  reduced  ? 

When  the  problem  is  thus  stated  the  situation  is  materially 
changed.  For  a  monistic  philosophy  as  thus  contemplated, 
there  is  support  in  our  actual  experiences  of  value.  The  presence 
in  our  experience  of  what  we  have  described  as  practical  absolutes 
indicates  clearly  that  such  an  absolute  unconditioned  value  is 
presupposed,  and  that,  while  this  value  is  not  realised  in  any 
empirical  content  conceived  as  end,  it  is  realised  in  moments  of 
intrinsic  appreciation  where  the  empirical  will  comes  to  rest  in 
the  assumption  of  complete  realisation  through  identification 
with  the  met-empirical  will. 

It  is  not  our  purpose  to  pursue  these  suggestions  as  to  the 
possibility  of  a  philosophy  of  values  further  in  this  connection. 
How  the  ultimate  postulate  of  valuation  is  to  be  defined — 
whether  as  self-realisation  or  as  realisation  of  an  impersonal 
over-individual  will,  whether  as  "  will  to  power  "  or  as  will 
to  "  Selbsterhaltung  der  Welt " — or  whether  indeed  the  functional 
unity  and  continuity  of  values  presupposed  in  the  empirical 
experiences  of  feeling  and  will  can  be  defined  in  any  such  abstract 
terms,  is  a  question  which  for  our  present  purposes  may  be  left 
unanswered.  In  the  following  chapter  we  shall  undertake  a 
consideration  of  the  entire  axiological  question  of  the  evaluation 


Synthetic  Preference  383 

of  values,  of  the  meaning  of  the  distinction  between  subjective 
and  objective  values,  of  the  meaning  of  the  presupposition  of 
reality  implicit  in  every  feeling  and  judgment  of  value,  and  of 
the  ways  in  which  that  presupposition  is  actualised  and  acknow 
ledged.    From  this  larger  view  of  the  problem  we  may  again  return 
to  this  point.     The  object  of  these  paragraphs  has  been  merely 
to  show  the  bearing  of  the  results  of  our  empirical  analysis  upon 
this  question,  and  here  the  conclusion  of  importance  is  this. 
While  a  monistic  philosophy  of  values  is  possible,  it  must  be 
such  as  to  allow  us  to  take  the  facts  of  actual  value  judgment 
at  their  face  value.     We  must  not  distort  them  by  subsuming 
them  all  under  one  empirically  derived  ideal  and  norm,  for  they 
are  'practically  discontinuous.     There  are  situations  where  con 
dition  worths  have  supreme  value  without  the  consciousness 
that  they  are  the  necessary  condition  of  personal  and  over- 
individual  values.     Personal  values  may  have  supreme  worth 
without  the  consciousness  that  they  are  the  means  to  the  realisa 
tion  of  social  ends.     There  are  cases  where  altruistic  acts  are 
chosen  without  the  idea  that  thereby  the  self  is  realised.     In 
general  the  principle  must  be  recognised  that  the  feelings  of 
value— of   obligation   and   of   intrinsic   appreciation,    approval, 
and  disapproval— are  not  to  be  conceived  as  determined  by  the 
logical  relations  of  subordination  to  ultimate  ends,  but  as  feelings 
attaching  directly  to  specific  content,   and  as  conditioned  by 
empirically    derived    demands.     In    the    light    of    an    ultimate 
postulate  these  demands  may  be  interpreted  and  made  intelligible, 
but  they  cannot  logically  be  deduced  from  a  single  end. 


CHAPTER    XIV— (Conclusion) 
VALUATION  AND    EVALUATION 

I.  THE  PROBLEM— RESTATEMENT  OF  THE  AXIOLOGICAL  POINT 

OF  VIEW 

WHATEVER  else  the  course  of  this  investigation  may  or  may 
not  have  succeeded  in  bringing  to  light,  it  has  at  least  led  us 
to  a  point  where  we  may  see  the  justification  of  the  concept 
and  method  of  a  general  theory  of  value  as  outlined  in  the  in 
troductory    chapter.     Through    the    genetic    treatment    of    the 
different    types    of  value   judgments  and   their  laws,   a   more 
intensive  analysis  of  the  facts  of  worth  experience  and  a  more 
comprehensive    view   of   their  inter-relations  have  been  made 
possible.     Every  advance  in  intensive  analysis  and  in  compre 
hensive  correlation  should  bring  with  it  greater  power  of  in 
terpretation  ;     and  this   test   of   fruitfulness  has   been  applied 
at  various  points,  with  the  result  that  many  types  of  value 
judgment,    hitherto    not    sufficiently    understood,    have    been 
explained  in  terms  of  the  general  laws  of  valuation. 

But  such  increase  of  insight  as  we  may  have  gained  cannot 
have  failed  to  bring  more  fully  to  consciousness  the  other  aspect 
of  a  general  theory  of  value  described  as  axiological,  to  have 
emphasised  its  importance,  and  to  have  made  clearer  the  nature 
of  the  problem.  In  our  introductory  chapter  we  sought  to 
state  this  problem,  and  in  doing  so  we  examined  critically  a 
certain  view  according  to  which  the  two  problems  of  description 
and  evaluation  are  wholly  unrelated.  Between  the  feeling 
of  value  as  an  experience  of  the  individual,  conditioned  by 
psychically  derived  and  determined  presuppositions,  and  the 
judgment  of  value  with  its  claim  to  objectivity,  and  therefore 
its  logical  presupposition  of  a  world  of  unconditioned  values, 
there  is,  it  is  held,  no  common  ground.  According  to  this 
view  our  task  was  properly  concluded  in  the  preceding  chapter 
and  our  inability  to  discover  any  absolutely  unconditione 

384 


Valuation  and  Evaluation  385 

values  constitutes  a  condemnation  of  our  entire  method.  At 
this  point  a  philosophy  of  values  must  begin  with  an  entirely 
different  purpose  and  method,  its  task  being  to  discover  and 
to  formulate  the  single  absolute  value  logically  presupposed 
in  all  specific  forms  of  value,  and  from  this  to  develop  deduc 
tively  "  a  closed  system  of  pure  values,"  uncontaminated  by 
any  admixture  of  empirical  feeling  and  will.  Without  raising 
any  question  as  to  the  abstract  possibility  or  impossibility  of 
any  such  ideal — although  after  our  study  of  actual  valuation 
it  cannot  fail  to  appear  to  rest  upon  a  false  reading  of  the  mean 
ing  of  the  claim  to  reality  and  objectivity  in  value  judgments, 
we  may  easily  see  that  this  is  not  the  nature  of  the  problem 
as  it  presents  itself  to  us.  At  numerous  points  we  have  seen 
specifically — what  was  stated  in  general  terms  at  the  beginning 
—that  the  question  of  validity  or  evaluation  is  in  some  way 
closely  bound  up  with  the  facts  and  conditions  of  valuation, 
and  that  the  axiological  problem  rises  directly  out  of  the  psycho 
logical.  The  problem  of  evaluation  being  the  adjustment  of  the 
implicit  claims  to  reality  which  our  feelings  of  value  with  their 
presumptions,  judgments,  and  assumptions  make,  it  is  necessary 
to  interpret  those  claims  in  terms  of  their  empirical  origin  and 
conditions. 

It  is,  therefore,  with  a  much  more  concrete  body  of  questions 
that  axiology,  as  we  have  conceived  it,  has  to  deal, — with  those 
questions,  namely,  which  arise  when  in  practical  situations 
we  seek  to  distinguish  between  the  subjective  and  objective 
value,  between  the  founded  and  unfounded  judgment  of  value. 
Such  problems  are  everywhere  present,  not  only  in  practical 
reasoning  but  in  the  sciences  of  value.  It  was  in  fact  with 
a  view  to  their  solution  that  the  distinctions  between  actual 
and  imputed,  real  and  ideal,  subjective  and  objective,  values 
have  arisen  in  practical  judgments,  and  have  been  developed 
in  scientific  usage.  But  their  significance  is  still  more  clearly 
seen  in  all  those  discussions,  as  for  instance  between  individual 
ism  and  socialism,  about  ethical  and  social  ideals,  where  the 
dispute  rests  upon  certain  assumptions  as  to  the  possibility  of 
economic  or  moral  motives,  the  truth  or  falsity  of  certain  ideals 
—whether  they  are  founded  or  unfounded.  More  specifically, 
then,  we  have  the  problem  of  illusions  and  fallacies  of  value 
judgment,  a  problem  which  came  to  the  surface  at  various  points 
in  our  special  studies.  ^Esthetic  and  quasi-aesthetic  appreci 
ations,  personal  and  social  ideals,  religious  hypostatisations — 
2  c 


386  Valuation :   its  Nature  and  Laws 

all  these  give  rise  to  certain  ideal  and  imputed  values.  In 
what  sense  are  these  values  real  ?  Or  are  they  indeed  real  at 
all  ?  Now  from  the  standpoint  of  psychological  process,  all 
these  intrinsic  ideal  and  imputed  values  arise  in  the  processes 
of  actual  valuation,  and,  as  such,  are  in  a  sense  real.  They  all 
rest  upon  assumptions  which  have  the  feeling  of  reality.  The 
question  is  whether  from  some  other  point  of  view  they  may 
not  be  wholly  subjective  and  ideal,  fictitious  and  illusory,  as 
opposed  to  real.  What  is  the  standpoint  of  ultimate  judgment 
in  such  cases— how  shall  the  ultimate  criterion  of  reality  be 
defined  in  a  way  which  shall  include  predicates  of  worth  as  well 
as  of  truth,  attributions  of  value  as  well  as  of  fact  ? 


II.   REFLECTIVE  EVALUATION — NORMATIVE  AND  FACTUAL 

OBJECTIVITY 

i.  Analysis  of  Axiological  Distinctions. 

These  are  questions  of  reflective  evaluation,  and  any  solution 
of  them  must  involve  the  development  of  the  meanings  and 
implications  of  the  various   axiological  distinctions  which  re 
flective  evaluation  introduces.     The  origin  and  nature  of  these 
distinctions  we  have  already  seen.     Analysis  of  the  value  judg 
ment   (Chapter  II)  has  shown  us  that,  while  the  judgment  is 
assertorial,  there  is  always  some  implication  of  relation  of  values 
to  reality.     But  while  judgments  of  value  presuppose  reality, 
while  they  presume,  assume,  or  judge  their  objects  to  be,  or  to 
exist,  it  is  not  always  clear  in  what  sense  that  existence  is  claimed. 
Unreflective  value  judgments  are  not  unequivocal,  and  it  is  for 
the  purpose  of  removing  the  various  equivocations  which  arise 
that  the  distinctions  between  subjective  and  objective,  intrinsic 
and  instrumental,  actual  and  potential,  actual  and  imputed,  real 
and  ideal  or  imagined  values,  are  introduced.    These  distinctions 
we  have  already  found  of  use  as  guides  to  the  analysis  of  pre 
suppositions  of  feelings  of  value,  and  with  their  help  we  were 
able  to  determine  the  subject  and  object  and  the  dispositional 
and  actual  presuppositions  of  different  types  of  value  judgment. 
But  their  implications  are  ultimately  normative,  and  it  is  with 
the  development  of  these  implications  that  we  are  here  con 
cerned. 

The  whole  problem  is  bound  up,  therefore,  with  the  question 


Valuation  and  Evaluation  387 

of  the  ultimate  meaning,  in  all  its.  extension  and  intension, 
of  the  presupposition  of  reality  implicit  in  all  feelings  and  judg 
ments  of  value.  What  are  the  possible  meanings  of  reality 
as  employed  in  reflective  evaluation,  and  what  is  the  common 
logical  core  of  all  these  meanings  ?  The  answer  to  these  ques 
tions  will  enable  us  to  see  the  relation  of  values  to  fact  and 
truth,  of  normative  to  factual  and  truth  objectivity,  as  they 
appear  in  actual  judgment.  From  a  study  of  these  relations 
we  shall  then  be  enabled  to  understand  the  ultimate  meaning 
of  the  postulate  of  realisation,  and  to  develop  specific  criteria 
for  determining  the  extent  to  which  it  is  fulfilled. 

2.  Meanings  of  Existence  and  Truth. 

As  to  the  first  aspect  of  the  problem  we  may  get  our  starting- 
point  by  retracing  the  processes  of  reflective  evaluation,  and 
developing  the  implications  of  the  axiological  distinctions  there 
employed.  Reflective  evaluation  consists  in  the  clarification 
of  the  meaning  of  reality  implicit  in  judgments  of  value  by  the 
development  of  explicit  existential  and  truth  meanings.  Through 
the  development  of  these  distinctions,  subjective  valuation  is 
controlled,  this  element  of  control  being,  as  we  have  seen  in 
the  Introduction,  the  practical  significance  of  normative  ob 
jectivity. 

In  their  first  and  more  limited  meaning,  the  terms  actual 
and  real,  when  applied  to  values,  signify  that  the  presupposition 
of  reality — presumption  or  assumption,  as  the  case  may  be, 
is  directly  or  indirectly  convertible  into  already  completed 
factual  or  truth  judgments,  independent  of  the  value  judgment. 
The  terms  potential,  ideal,  and  imputed,  on  the  other  hand, 
mean  that  the  values  thus  described  have  presuppositions 
not  thus  convertible,  or  at  least  not  wholly  convertible,  into  such 
judgments.  What  are  the  meanings  of  existence  and  truth 
employed  in  such  evaluations  ? 

(a)  Existence:  Outer  and  Inner. 

The  first  and  most  obvious  meaning  of  the  presuppo 
sition  of  reality  is  that  of  physical  as  distinguished  from  psychical 
existence.  If  this  is  what  the  presupposition  means,  then  in 
order  that  the  value  may  continue  it  is  necessary  that  this 
specific  meaning  of  the  presupposition  of  reality  shall  be  ful- 


388  Valuation :   its  Nature  and  Laws 

filled.  If  this  meaning  cannot  be  realised,  if  conversion  into 
the  object  cannot  take  place,  then  the  value  is  subjective  and 
imaginary.  In  the  case  of  the  so-called  "  condition  "  worths, 
both  immediate  and  mediate  or  instrumental,  this  is  what  the 
distinction  between  subjective  and  objective  values  means. 
In  order  to  satisfy  directly,  or  to  be  instrumental  to  more 
ultimate  satisfactions,  the  object  must  have  physical  existence. 
A  feeling  of  the  imagination  has  in  this  case  merely  repre 
sentative,  not  actual,  value. 

Again,  for  some  purposes  of  normative  control  a  value  is 
said  to  be  actual  and  objective  when  its  object  exists — not  in 
the  physical  sense,  as  an  object  of  sense  perception,  but  still  in 
the  outer  sense  of  being  the  object  of  a  demand  external  to  and 
independent  of  the  will  of  the  subject.  When  a  desire,  ex 
pectation,  or  demand  of  the  individual  subject  finds  fulfilment 
in,  or  is  continuous  with,  the  demand  of  others,  individuals 
or  social  groups,  the  value  is  said  to  be  actual.  Sympathetic 
participation  in  the  feelings  and  wills  of  others  gives  rise  to 
the  construction  of  dispositions ;  and  the  assumption  of  ex 
istence  in  this  case  means  outer  existence  in  wills  other  than 
our  own.  Such  is  the  meaning  of  existence  implied  in  all  judg 
ments  in  which  a  quality  is  said  to  belong  to,  or  to  be  possessed 
by,  an  individual,  and  it  has  its  psychological  correlate  in  the 
concept  of  disposition.  Such  also  are  the  economic  and  moral 
social  values  determined  by  processes  of  demand  and  supply. 
The  value  exists,  and  in  a  sense  only  in  will  or  mind,  but  in 
the  collective  will  or  mind  as  distinguished  from  the  individual. 
A  subjective  value  is  said  to  be  actual,  to  have_  objective 
grounds,  when  it  is  in  some  sense  continuous  with,  or  con 
vertible  into,  the  social  value.  From  this  point  of  view,  as 
we  have  already  seen  in  our  analysis  of  impersonal  judgments, 
whatever  of  the  individual  and  personal  feeling  and  feeling 
dispositions  is  convertible  directly  or  mediately  into  a  supply 
for  that  demand,  has  actual  value,  in  that  it  is  founded  in  an 
existential  judgment.  All  else  is  irrelevant,  and  is  described 
as  ideal  and  imputed  value. 

These  are  the  chief  meanings  of  the  predicate  of  existence, 
but  there  is  still  a  third  which  is  equally  important  in  distinguish 
ing  the  subjective  from  the  objective,  the  real  from  the  unreal 
in  valuation.  In  this  case  the  distinction  is  within  the  individual 
subject.  The  demand  which  is  acknowledged  as  objective 
and  as  a  norm  for  the  control  of  the  fleeting  subjective  experi- 


Valuation  and  Evaluation  389 

ences,  is  not  "outer"  in  either  of  the  preceding  senses.  It  is 
an  inner  demand  which  represents  organised  and  permanent 
dispositions  as  over  against  temporary  desires  and  feelings. 
Any  form  of  will  which  has  become  ineradicable,  any  expectation, 
demand,  or  assumption  which  is  incontestable,  acquires  a  nor 
mative  objectivity  which,  in  contrast  to  the  desires  and  feelings 
which  it  controls,  makes  it  an  existent  which  must  be  taken 
into  account.  It  is,  accordingly,  merely  this  persistence,  con 
tinuity,  or  control  which  is  acknowledged  when  the  predicate 
of  existence  has  this  meaning  of  inner  reality.  x 


(b)  Truth:    Outer  and  Inner. 

The  preceding  meanings  of  reality  include  all  those  cases 
where  the  claim  to  objectivity  is  acknowledged  in  predicates 
of  existence — where  continuance  of  value  requires  the  possibility 
of  direct  conversion  of  presumption  or  assumption  into  existen 
tial  judgment.  There  are,  however,  other  cases  where  satisfaction 
of  this  claim  to  objectivity  does  not  require  this  direct  and 
immediate  conversion,  but  only  acknowledgment  of  truth  as 
objective.  What  is  the  meaning  of  the  objectivity  of  truth  ? 
Here,  as  in  the  case  of  the  predicate  of  existence,  there  are 
several  meanings  which  we  may  conveniently  classify  as  "  outer  " 
and  "inner"  truth. 

By  "  outer  "  truth  is  understood  in  ordinary  speech  "  cor 
respondence  of  idea  with  reality,"  reality  being  taken  in  the 
special  sense  of  existence.  Propositions  said  to  be  true  or 
false  in  this  sense  are  general  propositions  about  existents 
of  the  physical,  social,  or  individual  worlds,  or  connections, 
causal  and  other,  among  these  existents.  The  concept  of  cor 
respondence  includes  the  further  assumption  that,  while  the 
ideas  themselves  are  not  existents,  they  are  founded  upon 
existents,  and  are  hypothetical  in  the  sense  that  their  truth 
is  conditioned  upon  the  assumption  of  the  existence  of  the  ob 
jects  about  which  the  propositions  are  made.  The  external 
control,  though  remote,  is  assumed  to  be  real.  While  to  ground 
the  objectivity  of  a  value,  it  is  sufficient  to  say  that  the  idea  or 
concept  to  which  the  feeling  of  value  attaches  is  thus  true, 
still  there  is  always  implied  in  the  conception  of  outer  truth 

1  For  a  similar  discussion  of  the  meanings  of  existence,  from  the  standpoint  of 
Epistemology,  see  Baldwin,  Thought  and  Things,  vol.  I,  chapter  X,  §§  9,  10,  II.  Also 
his  discussion  of  "  primary,  secondary  and  tertiary  conversion,"  in  vol.  II,  chapter  III. 


390  Valuation :   its  Nature  and  Laws 

the  belief  that  the  judgment  of  truth  will  lead  to,  or  may  be 
converted  into,  the  judgment  of  existence.  .  Outer  truths  of  this 
sort  are  in  general  the  causal  laws  of  the  physical,  social 
economic,  and  individual-psychical  spheres. .  They  are  organised 
and  retrospective  propositions  about  existents,  and  many  ideals 
and  anticipations  of  value  must,  in  order  to  make  good  their 
claim  to  objectivity,  conform  to  these  laws. 

But  there  are  certain  meanings  of  truth  not  exhausted 
in  this  description,  cases  in  which  truth  is  not  "  outer "  in 
the  sense  that  it  claims  correspondence  with  external  exist 
ents.  Here  truth  is  said  to  be  merely  an  internal  relation 
among  ideas,  correspondence  of  idea  with  idea,  identity  and 
lack  of  contradiction.  The  claim  to  objectivity  in  this  case 
is  interpreted  as  logical  consistency,  necessity,  and  universality. 
It  is  sometimes  held  that  this  meaning  of  the  objectivity  of 
truth  is  also  the  ultimate  meaning  of  the  claim  to  objectivity  in 
values.  Whether  this  is  true  or  not  we  have  still  to  consider, 
but  we  may  at  least  admit  that  there  are  certain  values,  e.g.,  of 
knowledge,  where,  in  order  that  the  presupposition  of  reality 
may  be  fulfilled,  they  must  conform  to  this  demand. 

3.   The  Relation  of  Normative  to  Factual  and  Truth 
Objectivity. 

Evidently,  after  this  analysis,  the  next  question  is — to 
what  extent  normative  objectivity  is  identical  with  factual 
and  truth  objectivity,  to  what  extent,  in  other  words,  the  terms 
real  and  objective,  when  applied  to  the  values  of  objects,  have 
the  same  meaning  as  when  applied  to  objects  apart  from  valu 
ation.  There  can  be  no  question  that  at  some  points  they  are 
identical.  At  others  they  are  closely  related,  and  at  still  others 
they  may,  perhaps,  be  independent.  The  answer  to  this  ques 
tion  involves  the  whole  problem  of  the  relation  of  judgments 
of  value  to  judgments  of  truth  and  fact. 

(a)  Normative  and  Factual  Objectivity. 

There  are  some  cases  where  normative  and  factual  objec 
tivity  are  clearly  identical.  Here  continuance  of  subjective 
value  and  feeling  of  reality  requires  that  the  presumption  or 
assumption  of  existence  created  by  subjective  dispositions 
shall  be  convertible  into  an  existential  judgment.  When  the 


Valuation  and  Evaluation  391 

home-sick  mariner  imagines  his  desired  haven  near,  this  feeling 
the  imagination  has  representative  value,  but  it  will  not 
continue  unless  the  assumption  develops  into  existential  judg 
ment,  either  through  perception  or  by  inference  from  obser 
vations  by  means  of  his  nautical  instruments.  Similarly, 
in  the  case  of  Gaunilo's  Island,  the  mere  formation  of  the  image 
and  assumption  of  its  existence  is  not  sufficient;  there  must 
be  conversion  into  existential  judgment  of  the  first  type,  in 
order  that  it  may  have  real  value. 

In  the  second  place,  it  is  clear  that  normative  objectivity 
is  frequently  identical  with  factual  objectivity  of  the  second 
type.     Wherever  the  reality  of  a  value  is  conditioned  by  the 
belief  that  the  object  desired  is  also  the  object  of  desire  and  will 
on  the  part  of  others,  individuals  or  social  groups,  there  factual 
objectivity   is   implied   in    the   normative.     Thus   in   all   cases 
where   the   value  judgment   of   the    individual    lays   claim   to 
objectivity  in  the  sense  of  the  impersonal  economic  or  "moral" 
judgment,    this   judgment    can  receive   its  validity   only  from 
an  existential  judgment  of  the  second  type,  which  predicates 
of  the  value  an  existence  in  social  demand  and  supply  independeni 
of    the    individual.     "  Normal "    exchange    values    or    prices, 
"  normal  "  moral  or  participation  values,  are  both  facts  and 
norms— facts  in  that  they  have  a  kind  of  outer  existence  in 
dependent  of  the  subject,  and  norms  because  the  subject  must 
take  this  outer  reality  and  control  into  account.     They  have 
normative    objectivity    precisely    because    they    have    factual 
objectivity.     In  so  far  as   actualisation   of  subjective  feelings 
of  value  is  conditioned  by  exchange  or  participation  in  social 
activities,  economic  or  moral,  the  individual's  judgment  of  the 
value  of  the  object  or  disposition  must  conform  to,  and  be  ful 
filled  in,  the  objective  social  value.     Such  norms  we  may  de 
scribe  as  instrumental;    and  they  must  have  factual  existence 
in  the  sense  that  they  are  founded  upon  social  demand. 

(b)  Normative  and  Truth  Objectivity. 

In  general  it  may  also  be  said  that,  when  the  value  is  instru 
mental,  normative  objectivity  is  identical  with  truth  objectivity 

the  outer  type,  or  with  propositions  about  relations  among 
physical,  social,  and  individual-psychical  existents.  Whenever 
the  actualisation  of  any  ideal  or  anticipation  requires  the  appli 
cation  of  physical  objects  or  participation  in  economic  and  social 


39 2  Valuation :   its  Nature  and  Laws 

processes,  this  ideal  must  conform  to  the  general  laws  or  truths 
of  these  spheres.  All  instrumental  values  presuppose  first  direct 
conversion,  and  if  that  is  not  possible,  indirect  conversion  into 
existential  judgments. 

But,  as  we  have  said,  there  are  other  values,  ideals,  and 
anticipations  which  transcend  the  claims  of  this  kind  of  outer 
truth,  and  which,  in  order  to  be  valid,  do  not  require  this 
direct  or  indirect  conversion  into  existences.  Such  are  the 
ideal  values  of  ethical,  aesthetic,  and  religious  experience.  The 
assumptions  of  ultimately  desirable  ethical  dispositions,  and  of 
objective  beauty  as  distinguished  from  the  subjectively  effective, 
are  ideals  of  this  sort.  Developed  though  they  may  have  been 
in  social  interaction,  and  retaining,  as  they  undoubtedly  do, 
a  secondary  instrumental  value  for  social  participation,  they 
are,  nevertheless,  now  primarily  significant  as  the  conditions 
or  presuppositions  of  the  continuity  of  individual  and  per 
sonal  values.  Inner  truth  and  reality  such  ideals  have ; 
they  are  ideal  constructions  which  are  realised  or  fulfilled  in 
the  individual  experience  which  they  control.  They  are  not 
empty  imagery,  mere  objects,  for  they  have  a  funded  meaning 
or  value  which  can  be  converted  into  actual  feelings  of  value. 
But  their  presupposition  of  reality  cannot  be  wholly  converted 
into  existential  and  truth  judgments.  There  are  aesthetic 
values  which  do  not  claim  the  physical  existence  of  their  object. 
There  are  personal  ethical  ideals  the  objectivity  of  which  is 
claimed  in  every  feeling  of  obligation,  in  every  judgment  of  im 
putation  of  which  they  are  the  grounds,  and  yet  this  normative 
objectivity  does  not  imply  that  they  are  actually  realised,  or 
even  capable  of  complete  realisation,  in  specific  individuals 
or  societies. 

We  are  thus  led  to  the  conception  of  "  inner  truth,"  as  the 
last  of  the  pre-formed  and  organised  definitions  of  the  meaning 
of  the  presupposition  of  reality.  Surely  here  value  and  truth 
are  ultimately  and  completely  identical ;  all  values  presuppose 
at  least  inner  truth  or  validity.  It  all  depends  upon  whether 
truth,  when  thus  defined  abstractly  and  retrospectively,  can  also 
be  defined  comprehensively  enough.  The  most  comprehensive 
definition  of  abstract  intrinsic  truth  attainable  is,  as  we  have 
seen,  that  of  logical  consistency,  with  its  assumption  of  uni 
versality  and  necessity.  Can  the  postulate  of  valuation  be 
identified  with  this  logical  postulate  ?  Or  is  there  a  last  meaning 
of  inner  truth  not  definable  apart  from  the  judgment  of  value 


Valuation  and  Evahiation  393 

in  which  it  is  implied  ?  We  have  already  suggested  doubts  of 
this  identity  of  truth  and  value,  and  a  fuller  examination  of  the 
question  establishes  these  doubts  more  firmly. 

This  lack  of  identity  may  be  seen  at  two  points.  On  the 
one  hand,  such  ethical,  aesthetic,  and  religious  ideals,  though 
still  claiming  inner  truth,  often  appear  in  an  indeterminate 
form,  which  makes  them  not  comparable  with  abstract  con 
ceptions,  a  fact  especially  apparent  in  aesthetic  and  religious 
symbolic  truth.  It  may  be  said  that  this  fact  means  merely 
that  the  implicit  claim  is  never  pushed.  But  this  is  hardly  true, 
for  the  reason  that  the  presupposition  of  reality  is  satisfied  with 
out  pushing  the  claim;  and  in  case  the  attempt  is  made,  for 
extrinsic  reasons,  to  turn  the  reality  into  abstract  truth,  and 
the  attempt  to  do  so  is  unsuccessful,  the  feeling  of  inner  truth 
still  persists.  On  the  other  hand,  this  same  lack  of  identity 
is  apparent  from  the  fact  that  logical  necessity  does  not  always 
include  realisation  of  inner  truth.  It  by  no  means  follows,  as 
we  have  seen  in  the  preceding  chapter,  and  shall  see  more  clearly 
in  the  sequel,  that  such  logical  necessity  includes  reality 
for  the  will  and  for  appreciation  generally.  Nor  does  logical 
contradiction  necessarily  mean  incompatibility  for  volition. 
Intellectual  conviction  does  not  always  include  realisation. 

We  are  therefore  forced  to  conclude,  either  that  certain  forms 
of  value  judgment  are  independent  of  truth  judgments,  or  else 
that  such  truth  as  they  presuppose  is  not  definable  in  retrospec 
tive  or  logical  terms.  In  the  former  case,  appreciation  includes 
reality  unconditionally,  and  there  is  no  criterion  other  than 
this  immediate  appreciation.  In  the  latter  case,  there  is  still 
an  implicit  hypothetical  reference  to  judgments  of  existence  and 
truth  not  reducible  to  any  of  the  retrospective  formulations 
or  organisations  of  experience  thus  far  described.  The  latter 
is  probably  the  true  interpretation  of  the  situation.  While 
tese  ideal  values,  in  order  to  make  good  their  claim  to  norma 
tive  objectivity,  do  not  demand  conversion  into  existential 
and  truth  judgments  of  the  types  defined,  they  still  have  a  re 
ference  to  matter  of  fact  and  truth.  Just  what  this  hypothetical 
reference  may  be,  can  be  determined  only  after  the  postulate 
of  valuation  itself  is  developed  in  all  its  meaning.  In  either  case 
however,  this  much  is  certain— there  is  only  partial  identity 
between  normative  and  factual  objectivity,  between  the  axio- 
logical  and  epistemological  predicates.  The  distinction  be 
tween  subjective  and  objective  values  is  one  which  arises  within 


394  Valuation :   its  Natztre  and  Laws 

the  function  of  valuation  itself  with  a  view  to  the  control  of 
this  function.  The  axiological  predicates  of  existence  or  truth 
get  their  meaning  from  their  place  in  this  function,  and  this 
meaning  is  always  relative  to  the  special  intent  of  the  sub 
jective  experience  to  be  controlled.  Every  value  is  in  a  sense 
real.  An  unreal  value  is  a  contradiction  in  terms.  The  only 
question  is  as  to  how  this  reality  shall  be  explicitly  acknow 
ledged  and  characterised  in  terms  of  reflective  definition. 

4.  Proof  of  this  Conclusion  in  the  Value  Judgments  of 

Religion. 

Religious  values  furnish  an  admirable  illustration  of  the 
truth  of  this  analysis  of  the  relation  of  judgments  of  value 
to  judgments  of  truth  and  fact — of  their  relative  independence 
at  certain  points.  They  are,  as  Hoffding  has  well  said,  value 
judgments  of  the  second  degree,  in  that  they  express  the  feelings 
which  arise  from  the  consideration  of  the  fate  of  primary  per 
sonal  and  social  values  in  reality ;  they  express  the  demand  for 
conservation  of  values  already  acquired.  As  a  consequence  of 
this  character,  the  judgment  of  value  is  bound  up  originally 
with  existential  judgments,  the  real  meaning  of  which  is  not 
always  at  first  apprehended. 

The  object  of  religious  belief  and  devotion  is  psychologically 
an  appreciative  construction  externalised  and  given  "  outer  " 
existence  in  the  manner  already  described  (Chapter  XI).  The 
presupposition  of  reality  is  acknowledged  in  existential  judg 
ments  of  the  most  elementary  type.  As  a  consequence,  for 
the  unsophisticated  religious  consciousness,  appreciative  and 
existential  meanings  are  often  inextricably  mixed.  Literal 
paradises  and  hells  arise  out  of  this  state  of  primitive  undifferen- 
tiation  ;  the  heights  and  deeps  of  our  experiences  of  value  are  so 
confused  with  temporal  and  spatial  magnitudes  that  belief  in 
the  reality  of  the  values  is  made  to  rest  upon  the  belief  in  the 
physical  existence  of  the  symbols.  But  when  this  first  fusion 
—and  confusion — is  broken  up,  as  it  ultimately  is  when  the 
interpretation  of  the  presupposition  of  reality  in  this  way  is  no 
longer  possible,  i.e.,  when  the  object  is  neither  perceptually  verifi 
able  nor  continuous  with  other  truth  judgments,  a  readjustment 
of  reality-meanings  takes  place.  A  restoration  of  the  value  in 
the  form  of  a  conception  of  symbolic  or  inner  truth  appears, 
and,  if  the  value  involved  is  vital,  the  presupposition  of  reality 


Valuation  and  Evaluation  395 

outlasts  the  cruder  factual  and  truth  judgments.  The  history 
of  religions,  of  Christianity  no  less  than  the  others,  shows  this 
progressive  clarification  of  reality-meanings,  the  historical 
culmination  in  the  case  of  Christianity  being  the  attempt  to 
identify  the  presupposition  of  reality  implicit  in  the  ideal  of 
perfection  with  logical  or  truth  objectivity  in  its  most  abstract 
form.  The  failure  of  this  attempt,  as  in  the  Ontological  Proof, 
marks  the  full  realisation  of  the  primacy  of  values.  Experience 
has  shown  the  remarkable  power  of  religious  beliefs  to  recuperate 
and  readjust  themselves,  and  from  this  vitality  we  may  probably 
infer  that,  until  the  values  of  men  themselves  change,  the  value 
judgments  of  religion  need  fear  nothing  from  the  appearance 
of  new  judgments  of  fact  and  truth. 


III.   THE  SUFFICIENT  REASON  OF  VALUATION— THE  GROUND 
OR  SANCTION  OF  VALUE 

i.  The  General  Problem  of  Sufficient  Reason  or  Sanction. 

The   situation    which    our   preceding    analysis   has    disclosed 
is  one  which  has,  in  many  quarters,  compelled  the  abandon 
ment   of  that   form   of  intellectualism  which   makes   objective 
value  derivative  from  preformed  and  already  organised  judg 
ments  of  fact  and  truth.     The  several  meanings  of  reality  and 
realisation  show  clearly  that  normative  objectivity,  and,  there 
fore,  the  predicate  of  reality  when  applied  in  the  only  sense 
proper  in  the  sphere  of  values,  cannot  be  completely  identified 
with  factual  and  truth  objectivity  in  the  narrower  sense,  either 
of   physical   and   psychical   existence   and   connections    among 
those  existents,  or  of  logical  validity.     With  the  realisation  of 
these   facts  has  come  the  various  forms  of  a  logism,  seeking 
for  concepts  and  criteria  of  validity  which  shall  embrace  and 
legitimate   the   various   reality-meanings   not   thus   reducible, 
and  finding  expression,  therefore,  in  a  reaching  out,  either  after 
an  independent  principle  of  sufficiency  for  value  judgments,  or 
toward  an  expansion  of  the  concepts  of  reason  and  truth  to 
include  grounds  not  recognised  in  the  older  logical  and  epistem- 
ological  theories.     This  general  problem— of  which  our  search 
for  the  meaning  and  grounds  of  normative  objectivity  is  but  a 
special  form,  may  then  be  properly  described  as  the  formula 
tion  of  a  theory  of  the  Sufficient  Reason  of  Valuation—or    if 


396  Valuation  :   its  Nature  and  Laws 

the  term  "  reason  "  seem  too  intellectualistic  in  its  connotation, 
Sufficient  Sanction.  It  includes  the  discovery  of  the  ultimate 
core  of  meaning  in  the  various  demands  for  reality,  and  the 
determination,  in  the  light  of  this  conception,  of  the  various 
types  of  experience  which  constitute  sufficient  fulfilment  of 
the  specific  demands.  We  are  thus  led  to  the  second  aspect 
of  our  general  problem. 

The  proposal  to  develop  an  independent  Sufficient  Reason 
of  valuation  seems,  at  first  sight,  to  give  away  the  whole  case. 
Because  of  the  intellectualistic  connotation  of  the  word  reason 
,  the  task  appears  predestined  to  failure,  the  use  of  the  term 
reason  or  ground  seeming  to  refer  wholly  to  grounding  in  pre 
ceding  judgments  of  fact  or  truth.  But  the  history  of  the 
term,  as  well  as  its  present  usage,  shows  that  it  has  a  much 
wider  connotation,  and  is  used  interchangeably  for  truth  and 
value.  Leibnitz  constantly  interchanges  the  two,  and  in 
applying  his  principle  of  sufficient  reason,  not  only  grounds 
judgments  of  value  in  more  ultimate  judgments  of  value, 
but  judgments  of  truth  and  fact  in  judgments  of  value.  As 
synonymous  with  his  principle  he  sometimes  uses  the  phrase 
"  principle  of  fitness  "  (convenance)  or  harmony.  We  give  the 
sufficient  reason  of  a  thing,  he  says,  when,  in  addition  to  its 
abstract  possibility,  we  show  its  compossibility  with  other 
things  ;  and  in  developing  that  compossibility  he  often  introduces 
a  concept  of  inclining  or  moral  compossibility — evidently  pure 
worth  conceptions.  Rationality  is  ultimately  identical  with 
continuity  of  experience  in  all  its  forms.  It  is  true,  as  I  have 
elsewhere  shown,1  that  in  the  succeeding  developments  of  the 
principle  the  tendency  has  been  again  to  restrict  it  to  judgments 
of  factual  and  truth  objectivity  of  the  types  defined,  but  this 
has  merely  had  the  effect  of  developing  a  counter-tendency  to 
the  various  forms  of  alogism  which  characterise  the  thought 
of  the  present. 

2.  The  Pragmatic  Criterion — Criticism. 

This  general  tendency  to  alogism  appears  especially  in 
connection  with  the  problem  of  illusions  and  disillusionment 
which  has  become  so  prominent  with  the  development  of  the 
causal  or  scientific  point  of  view.  Some  of  its  most  interesting 

1   The  History  of  the  Principle  of  Sufficient  Reason :  Its  Metaphysical  and  Logical 
Formulations.    Princeton  Contributions  to  Philosophy,  No.  i,  The  Princeton  Press,  1897. 


Valuation  and  Evaluation  397 

expressions  have  arisen  in  connection  with  just  those  problems 
which  have  emerged  in  our  study  of  the  phenomenology  of 
valuation— whether,  namely,  the  ideal  objects  with  their  im 
puted  values  are  merely  illusions,  or  whether  they  have  some 
relation  to  reality.  It  is  present,  for  instance,  in  all  the  works 
of  Guyau,  but  is  especially  emphasised  in  his  Esquise  d'une  moral, 
where  in  developing  a  purely  positive  morality  sans  sanction, 
either  theological  or  metaphysical,  he  finds  certain  value  judg 
ments  and  obligations  resting  upon  assumptions  which,  from  the 
external  point  of  view  of  the  existential  and  truth  judgments 
of  science,  are  wholly  unfounded,  and  therefore,  illusions. 
But  they  are  "fruitful"  illusions— fruitful  for  life,  and  must 
therefore  have  some  ultimate  reality  and  meaning.  Again, 
it  appears  to  Ehrenfels  that,  while  from  the  point  of  view 
of  actual  social  values  the  assumptions  upon  which  many 
subjective  worth  experiences  rest  seem  fallacious,  nevertheless, 
though  illusions,  they  still  have  some  instrumental  value.' 
Finally,  we  may  note  the  pragmatic  point  of  view  of  James  in 
his  estimate  of  the  subjective  and  personal  worths  of  religious 
experience.  He  deduces  their  truth  from  their  fruitfulness, 
and  from  their  value  for  life  is  disposed  to  argue  some  objective 
existent.  Moreover,  and  this  is  an  important  point,  he  dis 
tinctly  denies  the  jurisdiction  of  the  existential  and  truth  judg 
ments  of  science,  more  particularly  of  physiology  and  pathology, 
in  the  sphere  of  worth  judgment. 

In  all  these  cases  we  have,  broadly  speaking,  an  application 
of  the  pragmatic  criterion.  And  when  closely  examined,  this 
latter  further  seems  to  be  but  a  special  application,  with  greater 
intensity  of  purpose,  and  to  a  wider  range  of  phenomena  than 
formerly,  of  a  philosophical  formula  by  no  means  new,  and  one 
which  has  deserved  a  more  whole-hearted  application  than 
it  has  hitherto  received.  The  formula,  wherever  there  is  appear 
ance  there  is  reality,  or,  in  Herbart's  terms,  "  wie  viel  Schein,  so 
viel  Hindeutung  auf  Sein,"  has  been  accepted  in  some  form  or 
other  by  all  except  the  most  extreme  intellectualists,  but  in 
most  cases— and  notably  in  the  case  of  Herbart  himself— 
inveterate  rationalistic  prejudices  have  prevented  the  full 
development  of  its  implications.  The  turn  which  the  prag- 
matist  has  given  to  the  formula  seems  to  be  but  an  over-emphasis 
upon  one  type  of  indirect  conversion  of  the  implicit  presupposition 
of  reality,  namely  the  instrumental,  and  to  be  due  historically 
to  the  peculiar  emphasis  of  the  present  time  upon  the  utility 


39$  Valuation  :   its  Nature  and  Laws 

conceptions    of   biological   evolution,   with   which   in   fact   the 
pragmatic  conception  of  truth  is  closely  connected. 

With  that  phase  of  the  pragmatist's  contention  which  in 
sists  upon  a  broader  conception  of  reality  and  truth,  one  which 
will  include  and  legitimatise  certain  incontestable  presuppositions 
of   reality    implicit    in    values,   we   may    confess    ourselves    in 
sympathy.     We  have  seen  that,  while  values  presuppose  reality 
and  truth,  they  do  not  always  presuppose  such  conceptions  of 
truth  as  can  be  abstractly  denned  in  terms  of  "  correspondence 
of  image  with  object "  or  of  "  logical  consistency."     These,  to  be 
sure,  constitute  grounds  of  fulfilment  or  realisation  in  particular 
cases,  as  analysis  of  the  meanings  of  reality  shows.     When  the 
presupposition  of  reality  means  physical  or  psychical  existence 
or  ideal  consistency,  these  constitute  the  test  of  reality.     But 
the  presupposition  of  reality  means  more  than  this,  and  the 
application  of  such  specific  criteria  as  though  they  were  absolute 
negates  the  more  general  conception  already  formulated,  and 
gives  us  a  generous  crop  of  illusions.     With  the  criticisms  of 
the  formulations  of  "  truth  in  general "  and  their  direct  applica 
tion  as  criteria  of  realities,  we  are  therefore  in  accord,  but  it  is 
doubtful  whether  the  criterion  which  is  proposed  in  their  place, 
that  of  utility  or  of  indirect  conversion  through  instrumental 
judgments,  is  adequate  either  as  a  definition  of  the  meaning 
of  truth,  or  as  a  test  of  the  validity  of  the  various  implications 
of  reality.     All  values  have  a  reference  to  reality,  but  the  mean 
ing  of  that  reference  cannot  be  exhausted  in  the  concept  of 
utility. 

It  must,  nevertheless,  be  admitted  that  the  test  of  utility 
does  from  the  ground  of  many  judgments  of  value  which  have  an 
inner  presumption  of  reality,  but  are  not  directly  convertible  into 
judgments  of  existence.  The  concept  of  utility  has,  moreover, 
been  so  broadly  interpreted  as  to  include  all  types  of  instrumental 
relations,  biological,  social,  and  individual.  But  it  is  not  difficult 
to  see  that  this  concept  of  pragmatic  sanction  owes  its  value 
primarily  to  its  generality  and  that  it  suffers  from  the  same 
difficulties  as  the  other  criteria.  As  soon  as  the  abstract  con 
cepts  of  "  utility,"  or  "  fruitfulness  for  life,"  are  defined  and 
applied  to  particular  situations,  certain  equivocations  arise,  and 
with  them  certain  difficulties. 

Thus,  if  our  pragmatism  is  of  the  crude  type,  if  we  conceive 
"  life  "  in  an  external  way,  then  our  criterion  still  fails  to  include 
many  internal  meanings,  many  intrinsic  and  individual  values 


Valuation  and  Evaluation  399 

which  cannot  be  thus  related  to  the  external  biological  and  social 
existences.  Taken  in  this  sense,  the  criterion  is  still  external, 
in  that  it  really  does  not  transcend  the  crude  distinction  of 
physical  and  psychical  reality.  For  the  concept  of  instrumen 
tality,  the  relation  of  means  to  ends,  can  be  applied  only  to 
pre-established  existents.  An  idea  or  ideal  can  be  instrumental 
to  the  production  only  of  existences,  physical  goods,  or  psychical 
experiences,  and  to  be  thus  instrumental  it  must  itself  be  an 
existent. 

But  even  if  utility  is  taken  in  a  broader  sense,  if  our  prag 
matism  is  of  the  more  refined  type,  i.e.,  if  we  include  in  our  world 
of  pragmatic  truth  ideal  meanings  which  are  individual  and  in 
trinsic,  such  as  the  ethical  and  religious  "  practical  absolutes," 
we  can  do  so  only  by  resorting  to  a  shifting  use  of  the  terms 
utility  and  instrumental  value.  Thus,  when  Augustine  says, 
'  I  seek  thee  in  order  that  my  soul  may  live,"  he  attributes  in 
this  utterance  only  mediate  value  to  the  object  of  his  faith, 
but  when  in  other  utterances  he  speaks  of  God  as  the  highest 
or  only  good,  as  goodness  and  truth  itself,  the  object  of  his  faith 
appears  invested  with  immediate  value.  Between  the  two  the 
pragmatist  seems  to  find  no  difference,  when,  as  a  matter  of 
fact,  even  pragmatically  viewed,  there  is  all  the  difference  in 
the  world — a  difference  not  only  in  the  meaning  of  the 
presupposition  of  reality,  but  also  in  the  type  of  experience 
in  which  it  is  realised.  This  fallacy  of  equivocation  is  one 
to  which  the  pragmatist  is  constantly  prone.  He  uses  his 
concept  of  utility  primarily  to  apply  to  truths  in  the  case  of 
which  the  value  is  in  the  strict  sense  instrumental,  but  when 
he  finds  truths  and  values  of  a  purely  intrinsic  kind,  of  the 
higher  type  of  immediacy  where  the  value  is  immanental  and 
not  transgredient,  he  covers  the  situation  by  the  use  of  the 
general  phrase  "  fruitful  for  life."  This  appears  in  the  relative 
truth  which  James  has  been  recently  according  to  the  concept 
of  the  absolute.1  He  admits  its  function  as  practical,  as  the  means 
to  "  moral  holidays,"  and  again  as  the  presupposition  of  the 
immanental  values  of  aesthetics  and  religion,  but  does  not  realise, 
although  other  pragmatists  apparently  do,  that  the  whole 
instrumental  concept  is  here  in  danger.  For  to  realise  these 
experiences,  these  "  holidays,"  at  all,  the  very  condition  is  that 
you  do  not  make  them  conscious  ends,  and  still  less  means  to 
ends.  The  condition  of  their  functioning  at  all  is  that  they 

1  James,  Pragmatism,  Longmans,  Green  &  Co.,  1907,  Lecture  II,  pp.  73-79. 


400  Valuation :   its  Nature  and  Laws 

remain  implicit  assumptions — otherwise  the  process  is  self- 
defeating.  This  is  what  is  meant  at  bottom  by  the  criticism 
that,  as  a  theory,  pragmatism  itself  lacks  pragmatic  value.  We 
may  conclude,  then,  that  the  pragmatist's  attempts  to  charac 
terise  in  retrospective  terms  the  presupposition  of  reality  in  all 
its  meanings  is  no  more  ultimate  than  those  which  it  criticises, 
and  that  any  conception  of  the  primacy  of  values  which  shall 
be  satisfactory  must  not  confine  itself  to  the  instrumental  con 
ception. 

3.  The  Ultimate  Meaning  of  the  Presupposition  of  Reality  and  of 
its  Fulfilment — The  Demand  for  Continuity. 

What,  then,  is  the  ultimate  ground  or  sanction  of  value 
judgments  ?  The  answer  to  this  question  evidently  involves 
a  deeper  analysis  of  the  ultimate  presupposition  of  reality 
and  its  satisfaction,  one  that  goes  beyond  the  pragmatic  in 
terpretation,  valuable  as  it  is  as  a  protest  against  the  logical 
deduction  of  values,  and  as  an  attempt  to  widen  the  concept  of 
sufficient  reason.  It  is  equally  apparent  that  the  problem 
centres  about  those  intrinsic  values  the  objects  of  which  are 
ideals  not  reducible  either  directly  or  indirectly  to  judgments  of 
fact  or  truth,  and  not  related  instrumentally  to  values  the 
objects  of  which  have  been  already  acknowledged  as  true  or 
existent.  There  are,  as  our  study  has  abundantly  shown,  forms 
of  realisation  or  satisfaction  which  do  not  depend  for  their 
validity  upon  either  of  these  kinds  of  acknowledgment,  where, 
indeed,  realisation  is  possible  only  on  condition  that  the 
assumption  of  reality  merges  directly  into  realisation,  and  that 
we  do  not  seek  to  reduce  it  to  retrospective  judgments  of  fact, 
truth,  or  utility.  The  objectivity  claimed  for  the  ideal  values 
of  ethical,  aesthetic,  and  religious  experience  cannot  be  defined 
in  terms  of  any  of  these  conceptions  of  direct  or  indirect  con 
version  into  existential  and  truth  judgments.  To  determine 
just  what  constitutes  their  sufficient  reason  we  must  know  what 
the  demand  for  reality  ultimately  means. 

A  definition  of  this  demand,  as  a  postulate  of  reflective 
evaluation,  requires  a  comprehensive  analysis  of  the  common 
element  in  all  the  various  forms  of  presumption  and  assumption 
as  they  appear  on  the  different  levels  of  unreflective  experience. 
Here  it  is  no  longer  a  question  of  the  genesis  of  specific  forms  of 
the  common  presupposition,  but  of  the  intelligibility  of  the 


Valuation  and  Evahiation  401 

common  element  or  the  logical  core  which  gives  it  meaning,  and 
which    survives    its    transformation    into    various    forms.     If, 
then,   we  review  our  analysis  of  the   axiological  predicates— 
this  time,  however,  with  a  view  to  their  intension  rather  than 
to   their  extension,— we  find  but  one  element  which  is  clearly 
common  to  them  all,  namely,  the  postulate  that  our  experiences 
of  feeling  and  will,  as  subjective  and  individual,  are  in  some  way 
identical    or   continuous    with    a   reality    that    transcends    our 
momentary   experience.      They  have  a  reference,  either  trans- 
gredient  or  immanental,  beyond  themselves.     Sometimes   this 
postulate  means,  as  our  analysis  has  shown,  that  the  object, 
as  desired  and  enjoyed  by  us,  exists  in  the  physical  sense ;  and 
the  value  is  said  to  be  objective  and  valid  when  the  feeling  of 
value  continues  after  presumption  has  passed  into  existential 
judgment.     Valid  also  is  that  subjective  feeling  of  participation 
when  the  object  of  our  subjective  feeling  is  also  the  object  of 
other  wills  than  our  own,  when  our  feeling  and  will  is  continuous 
with  an  over-individual  experience.     Again  a  subjective  feeling 
of  value  is  valid  when  it  is  identical  or  continuous  with  dis 
positions  or  forms  of  will  in  ourselves  which  have  already  attained 
an    objective    and    over-individual    reference.     These    direct, 
together  with  the  corresponding  indirect,  forms  of  conversion 
through  judgments  of  truth,  constitute  the  usual  meanings  of 
realisation.     But  finally— and  this  is  the  most  significant  point 
—in  the  case  of  intrinsic  ideal  values  the  postulate  means  only 
inner  identity  and  continuity  of  the  will  with  its  objects  or  with 
itself,    through    successive    empirical    moments    of    realisation. 
This    internal    identity    and   continuity,    as    over   against   the 
discontinuity  of  momentary  and  isolated  desires  and  objects, 
creates    an    objectivity    within    the   subject's    experience,    and 
constitutes  the  last  meaning  of  objectivity  as  reality. 

The  character  of  this  ultimate  form  of  the  postulate  gives 
the  key  to  the  meaning  of  the  postulate  of  identity  in  all  its 
forms.  Here  the  transcendental  reality  presupposed  is  clearly 
will,  and  the  axiological  predicate  of  inner  truth  means  identity 
of  will  with  will.  The  other  expressions  of  the  postulate  are 
but  more  indirect  and  disguised  forms  of  the  same  claim  of 
identity  of  subjective  with  more  objectified  forms  of  will.  They 
are  specialised  demands  which  have  developed  as  secondary 
meanings  ;  and,  while  the  desire  for  existence  as  such,  and  for 
abstract  truth,  may  appear  apart  from  other  desires,  normally 
factual  and  truth  objectivity  are  demanded  only  as  instrumental 


2   D 


402  Valuation :   its  Natiire  and  Laws 

to  that  inner  reality  and  truth  which  arises  from  identity  of 
one  act  of  will  with  another,  and  all  secondary  distinctions  are 
made  with  a  view  to  the  maintenance  of  that  continuity. 

The  presupposition  of  reality  in  all  valuation  is,  then,  the 
identity  or  continuity  of  subjective  volition  with  forms  of  will 
which  transcend  the  individual  and  momentary  experience — 
not  merely  identity  of  subjective  "  image "  with  objective 
"  thing,"  of  subjective  will  with  over-individual  in  the  sense 
of  social  will,  nor  yet  of  subjective  will  with  itself  as  objectified 
as  a  disposition  of  a  person — but,  ultimately  identity  of  subjective 
will  with  a  met- empirical  will  not  completely  expressed  in  any  of 
these  forms.  Whether  that  identity  is  acknowledged,  that 
continuity  maintained,  by  explicit  existential  judgment  after 
arrest,  i.e.,  after  distinction  between  outer  and  inner,  or  by 
explicit  postulation  or  acknowledgment  of  persistent  assumptions 
or  ideals,  or,  finally,  by  the  mere  continuance  of  an  implicit 
assumption, — in  every  case  some  form  of  reality  or  being  is 
acknowledged  because  of  identity  of  subjective  with  objective 
will. 

The  realisation  of  this  postulate  of  identity  in  any  of  the 
specific  forms  of  empirical  continuity  gives  objective  value. 
Complete  identity  would  mean  absolute  value ;  and,  could  we 
formulate  this  met-empirical  will  in  such  a  way  as  to  deduce 
from  it  its  actual  content  or  objects,  a  system  of  absolute  values 
or  of  unconditioned  satisfactions  could  be  developed.  For 
objects  which  could  afford  the  basis  of  such  complete  identity 
would  be  universal  and  eternal  values,  unmodifiable  by  any 
empirical  conditions,  objects  of  an  over-individual  will  or  long 
ing  unaffected  by  the  desires  and  feelings  of  the  individual. 

The  attempts  to  define  the  met-empirical  will  in  terms  of 
realisation  of  self,  or  realisation  of  an  impersonal  over-individual 
will  interpreted  socially,  have  been  criticised  in  the  preceding 
chapter.  The  wholly  acquired  and  intra-experiential  character 
of  these  distinctions  and  meanings  makes  it  inconceivable  that 
either  should  be  completely  identical  with  the  total  meaning 
of  the  over-individual  will;  and  the  analysis  of  actual  value 
judgments  in  these  spheres  showed  their  relative  nature.  It 
is  still  conceivable,  however,  that,  if  the  meaning  and  content 
of  volition  should  be  abstractly  enough  defined,  it  might  be 
found  ultimate  enough  to  form  an  absolutely  necessary  pre 
supposition  of  all  actual  volition,  and  at  the  same  time  broad 
enough  to  include  these,  as  well  as  all  other,  forms  of  realisation — 


Valuation  and  Evaluation  403 

and  thus  to  be  an  incontestable  value.  The  many  attempts 
thus  to  characterise  the  unconditioned  act  of  will  presupposed 
in  all  relative  and  subjective  volitions — as  "  the  will  to  love," 
"  the  will  to  power,"  etc.,  can  logically  lead  only  to  an  abstraction 
in  which  all  elements  of  individual  desire  and  emotion  are 
eliminated,  and  finally  to  the  emptiest  of  logical  abstractions 
—the  will  to  pure  being.  Such  is  the  recent  attempt  to  ground 
all  values  in  an  over-individual  longing,  an  over-individual  will 
to  the  self-maintenance  of  the  world,  or,  more  briefly,  "  Wille 
zur  Welt."  Suggestive  as  this  characterisation  of  the  ultimate 
presupposition  of  reality  undeniably  is — as  a  logical  characterisa 
tion,  it  is  too  thin  an  abstraction  upon  which  to  build  a  world 
of  actual  values.  For,  either  it  includes  in  it  the  concrete 
content  of  an  already  defined  and  realised  world,  or  else  it  means 
nothing  more  than  mere  persistence  or  continuity  of  will.  In 
the  first  case,  the  relativity  of  all  ideal  construction  makes  it 
impossible  to  show  the  absolute  identity  of  actual  ideals  and 
their  realisations  with  such  a  will ;  the  postulate  of  eternity  of 
value  does  not  include  the  eternity  of  specific  objects  of  value. 
In  the  second  case,  the  very  abstractness  of  the  meaning  of  will 
as  thus  defined  makes  it  so  empty  a  concept  that  it  is  useless  as 
a  criterion  of  evaluation  until  it  is  reduced  to  the  specific  forms 
of  the  demand  for  continuity.  Viewed  in  either  way,  it  is  useless 
as  a  point  of  departure  for  developing  a  system  of  absolute 
values. 

From  such  reflections  it  seems  necessary  to  conclude  that 
the  presupposition  of  reality,  realised  in  various  forms  of  actual 
worth  experience,  cannot  be  defined  in  such  a  way  as  to  deduce 
from  it  absolute  values.  It  can  be  characterised  merely  as  a 
postulate  of  the  continuity  of  value,  the  persistence  of  will  and 
its  satisfaction  beyond  any  empirical  forms  which  it  may  assume. 
The  logical,  or  better  axiological,  postulate  of  identity  of  the 
empirical  with  the  met-empirical  will  becomes  for  practical 
purposes  the  demand  for  continuity  of  empirical  desire  and 
feeling,  for  continuity  of  subjective  with  already  objectified  forms 
of  will.  The  real  problem  is,  then,  to  find  the  empirical  forms 
in  which  the  demand  for  continuity  appears,  and  to  determine 
the  extent  to  which  the  empirically  derived  objects  fulfil  these 
demands. 

*  Hugo  Miinsterberg,    Philosophic  der   Werte :    Grundzuge  einer   Weltanschauung 
Leipzig  :  Verlag  Johann  Ambrosius  Earth,  1908. 


404  Valuation :   its  Nature  and  Laws 


4.    The  Postulate  of  Continuity:    Acquirement  and  Conservation 
of  Value — Practical  Absolutes. 

The  two  forms  in  which  this  demand  appears  in  immediate 
experience  are  the  transgredient  and  immanental  references 
of  feelings  of  value.  When,  however,  these  references  become 
the  explicit  demands  of  reflective  evaluation,  they  appear  as  the 
postulates  of  acquirement  and  conservation  of  meaning  or  value. 
In  order  that  the  function  of  valuation  may  continue,  that 
volition  may  persist,  new  values  must  be  acquired,  and  those 
already  acquired  must  persist  or  be  conserved  in  new  objects 
on  new  levels.  Every  specific  form  of  the  demand  for  continuity 
may,  therefore,  be  comprehended  in  these  two  aspects  of  the 
general  postulate  of  valuation. 

It  is,  perhaps,  idle  to  raise  the  question  as  to  which  of  the 
two  aspects  is  ultimate.  In  his  admirable  discussion  of  religious 
values,  Hoffding  *  makes  the  concept  of  conservation  the  more 
ultimate,  deducing  the  demand  for  acquirement  from  conserva 
tion  on  the  ground  that  the  condition  of  the  persistence  of  value 
is  the  creation  of  new  values.  But  further  reflection  seems  to 
require  a  reversal  of  this  relation.  The  inmost  principle  of 
sufficient  reason  of  volition,  and  the  ultimate  criterion  of  norma 
tive  objectivity  alike,  are  to  be  found  in  the  fact  that  an  act 
of  will  affords  the  basis  for  ever  new  forms  of  will,  and  that 
a  value  is  but  the  starting-point  for  new  values.  But  in  this 
principle  is  included  the  further  postulate  that  any  volition 
already  fulfilled,  any  value  experienced,  is  conserved  in  some 
form  as  the  platform  for  new  volitions  and  values,  that  any 
essential  value  persists  in  new  forms  of  reality.  The  postulate 
of  conservation  is  a  retrospective  formula  having  reference 
to  those  implicit  assumptions  which  constitute  the  indispensable 
conditions  of  continuity  of  valuation. 

In  the  recognition  of  this  fact — that  the  postulate  of  con 
servation  is  included  in  the  ultimate  presupposition  of  reality, 
we  have  finally  the  basis  for  an  understanding  of  the  function 
of  the  axiological  predicates  of  existence  and  truth.  The  judg 
ment  of  value,  while  an  appreciation,  is  one  which  includes 
reality  ;  while  assertorial  in  form,  it  has  a  hypothetical  reference 
to  reality.  The  predicates  of  existence  and  truth  are,  when 

1  Hoffding,    The   Philosophy  of  Religion,    translation   from    the   German    edition. 
Macmillan  &  Co.,  1906.     pp.  215  ff. 


Valuation  and  Evaluation  405 

used  axiologically,  the  explicit  meanings  developed  in  the  interest 
of  conservation  of  the  implicit  presupposition.  As  prospective 
ideals,  values  require  merely  that  inner  truth  which  is  identical 
with  value.  As  norms  conserving  the  values  already  acquired, 
and  forming  the  basis  of  new  values,  they  must  have  that  outer 
existence  and  truth  which  makes  them  objective  and  over- 
individual.  The  demand  for  outer  existence  and  truth  in  its 
various  forms  is  but  an  expression  of  the  demand  for  persistence 
and  conservation,  and  the  predicates  in  which  this  demand  is 
acknowledged  are  the  signs  of  such  conservation. 

It  is  also  clear  that  any  form  of  existence  or  truth  may, 
under  certain  circumstances,  have  absolute  value.  Any  of  the 
objects  of  actual  desire,  any  of  the  ideal  constructions  developed 
in  the  empirical  processes  of  valuation,  may  become  practical 
absolutes  in  that  their  persistence  is  demanded  unconditionally. 
The  will  to  live,  the  will  to  be  a  person,  and  the  will  to  participate 
may,  as  we  have  seen,  under  certain  circumstances,  be  un 
conditional,  and  the  realisation  of  certain  objects  of  will,  as 
for  instance,  the  minimum  of  existence,  of  characterisation,  or  of 
participation,  may  have  absolute  value.  The  feeling  of  absolute 
obligation  in  its  several  forms  appears  in  connection  with  those 
objects  which  are  without  capacity  of  substitution,  and  which 
are  therefore  assumed  to  be  the  indispensable  conditions  of 
continuity  of  valuation.  In  general  it  may  be  said  that  the 
feeling  of  absolute  value  of  the  existence  or  truth  of  certain 
concrete  objects  thus  arises  because,  under  the  empirical  con 
ditions  of  their  origin,  already  described,  they  are  the  objects 
in  which  the  met-empirical  postulate  of  identity  finds  con 
crete  expression  in  an  individual  demand.  While  there  are 
no  absolute  values  in  the  theoretical  sense  that  the  object 
of  the  individual  empirical  will  can  be  shown  to  be  absolutely 
identical  with  the  object  of  the  met-empirical  will,  the 
empirically  derived  presupposition  identical  with  the  logical 
presupposition,  there  are,  nevertheless,  as  we  have  seen,  practical 
absolutes,  objects  which  in  specific  situations  are  the  indis 
pensable  conditions  of  realisation  of  the  postulate  of  continuity. 

5.  Axiological  Sufficiency — The  Well-Founded  Value. 

The  presupposition  of  reality,  the  claim  to  objectivity,  means 
then,  ultimately,  the  fulfilment  of  the  postulate  of  continuity 
of  value  in  its  two  forms  of  acquirement  and  conservation. 


406  Valuation :   its  Nature  and  Laws 

The  significance  of  this  conclusion  is  far-reaching.  If  this  is 
what  the  postulate  of  valuation  ultimately  means,  the  sufficient 
reason  of  any  assumption  of  reality,  any  subjective  experience 
of  feeling  or  will,  is  to  be  found  in  just  this  continuity,  in  the 
fact  that  it  maintains  itself  in  succeeding  experiences  of  feeling 
and  will.  Unhindered  activity  or  continuity  is  the  source  of 
reality  and  value.  Judgments  of  existence  or  of  logical  con 
sistency  are  merely  special  forms  of  registering  the  fact  that 
specific  presuppositions  have  maintained,  themselves.  They  are 
but  secondary  modifications  of  the  primary  feeling  of  reality. 
An  incontestable  assumption,  postulate,  or  belief  registers  the 
same  fact. 

This  is  the  axiological  meaning  of  certain  facts  brought 
out  in  our  psychological  analysis.  We  have  seen  (pp.  42-49) 
that  it  is  not  the  entrance  of  a  new  element,  the  existential 
judgment,  which  makes  the  mere  idea,  until  now  unreal,  ob 
jective  and  real;  but  it  is  rather  the  process  of  abstraction— by 
which  an  "objective,"  with  primitive  presupposition  of  reality, 
is  turned  into  a  mere  object,  that  creates  the  feeling  of  unreality. 
When  this  abstraction  has  taken  place,  when  the  presumption 
of  reality  is  disturbed,  it  must,  to  be  sure,  be  restored  in  some 
explicit  acknowledgment  of  the  presupposition  of  reality,  but 
unless  disturbed  it  is  valid.  No  less  valid,  it  may  be  added, 
are  any  secondary  and  derived  assumptions  and  postulates 
which  form  the  basis  of  continuity  of  actual  experience  of  value. 

This  being  the  meaning  of  the  presupposition  of  reality  and 
of  its  fulfilment,  it  is  clear  in  what  direction  the  criterion  of 
its  realisation  must  be  found,  fit  is  apparent  that  we  must 
look,  not  for  absolute  grounds,  but  for  sufficient  sanctions  of 
value  judgments,  not  for  absolute  norms,  but  for  well-founded 
ideals  of  value.  The  postulate  of  identity  of  the  empirical  with 
the  met-empirical  will  is  realised,  not  in  absolute,  but  in  sufficient 
identities — i.e.,  sufficient  for  the  continuity  of  value.'  Here, 
then,  in  this  question  of  the  nature  of  axiological  sufficiency 
lies  our  real  problem.  For,  if  it  is  true  that  we  cannot  find  points 
of  absolute  logical  identity  between  the  objects  of  the  empirical 
and  met-empirical  will,  it  is  no  less  true,  as  we  have  already 
seen,  that  the  concept  of  mere  continuity  is  useless  until  made 
more  concrete  and  practical.  A  closer  scrutiny  of  the  concept 
of  sufficiency  is  demanded,  and  here  again  the  thinking  of  Leibnitz 
is  suggestive. 

For  him,  as  we  have  seen,  rationality  and  continuity  are 


Valuation  and  Evaluation  407 

identical.     But  complete  continuity  of  finite  objects  with  the 
ultimate    reality,    and    therefore    absolute    grounding,    would 
require,    as    he    says,    "  infinite    analysis."     Nevertheless,    in 
order  that   a  phenomenon  may  be   "  well-founded,"   in  order 
that  practical  distinctions  may  be  made  between  the  subjective 
and  the  objective,  such  infinite  analysis  is  not  necessary.     It  is 
required  merely  that  the  analysis  shall  be  sufficient,   that  it 
shall  satisfy  a  specific  form  of  the  demand  for  continuity,  i.e., 
sufficient  in  the  sense  in  which  we  say  of  a  certain  magnitude  that 
it  is  sufficient  when  it  satisfies  a  given  equation.1   In  like  manner, 
recalling  the  steps  of  our  argument  up  to  this  point,  we  may  say 
that  a  value  is  well-founded  when  it  fulfils  a  specific  presupposition 
of  reality.     In  the  case  of  a  value  judgment,  the  "  equation"  to 
be  satisfied  is  formed  by  the  analysis  of  the  specific  presupposition 
of  reality  or  demand  for  objectivity  implicit  in  a  value  judgment, 
and  its  solution  consists  in  determining  the  specific  meaning  of 
the  predicate  of  reality  which  satisfies  that  demand.     The  entire 
question  is  one  of  relevancy.     With  regard  to  the  world  of  cog 
nitive   judgment,   Leibnitz   says :     "  Although   this   entire   life 
were  said  to  be  nothing  but  a  dream,  and  the  visible  world 
nothing  but  a  phantasm,  I  should  call  this  dream  or  phantasm 
real  enough  if  we  are  never  deceived  by  it,  when  we  use  our 
reason    rightly."  2     In    like    manner,    paraphrasing    Leibnitz's 
thought  for  the  sphere  of  practical  value  judgment,  we  may  say  : 
a  value  is  real  and  objective  enough  if  it  maintains  itself,  if 
the  ideal  continues  when  we  reason  and  will  rightly,  i.e.,  if  we 
do  not  take  the  presumption  or  assumption  of  reality  to  mean 
that  which,  in  the  light  of  its  place  in  the  system  of  our  experi 
ences,  it  does  not,  and  indeed  cannot  mean. 

The  applications  of  this  principle  to  the  practical  reasonings, 
anticipations,  and  postulates  of  our  individual  and  social,  of 
our  economic,  ethical,  religious,  and  aesthetic  life,  are  compre 
hensive  and  varied,  and  to  some  of  these  applications  we  shall 
direct  attention  in  the  sequel.  Here  we  may  note,  merely 
in  the  light  of  what  has  preceded,  that  it  introduces  into 
our  application  of  the  criterion  of  continuity  that  element  of 
control  which  gives  it  practical  axiological  value.  For  it  tells 
us  this  much  negatively  at  least — that  the  objectivity  or  reality 
implicit  in  a  value,  although  well-founded,  need  not  necessarily 

1  R.  Latta,  Leibnitz:  The  Monadology,  etc.     Oxford:  The  Clarendon  Press,  1898, 
p.  236,  note. 

2  De  Modo   distingucndi  phenomena  realia  ab  imaginariis.     Ed.  J.   E.   Erdmann, 
Berlin,   1840,  4442. 


408  Valuation :   its  Nat^tre  and  Laws 

mean  identity  with  factual  or  truth  objectivity  of  any  of  the 
types  described.     Reality  of  ideals  does  not  necessarily  mean 
their  translation  into  terms  of  social  objectivity  or  existence  • 
nor  can  we  infer  that  specific  appreciative  constructions  or  ideals 
will  continue  in  precisely  the  same  form.     We  have  found  that 
except  in  the  case  of  the  most  ultimate  personal  worths    these 
ideals  have  capacity  of  substitution.     As  long  as  new  experi 
ences  arise,  so  long  there  will  be  new  valuations.     A  highest 
value,  other  than  the  practical  absolutes  of  the  personal  sphere 
is  not  demonstrable.     Value  can  be  preserved  only  by  trans- 
Drmations,  and  all  that  we  can  say  is  that,  since  value  persists 
essential  value  in  any  ideal  will  be  acknowledged  in  new 
constructions  and  new  existential  judgments. 


IV.  APPLICATION  OF  THE  PRINCIPLE  OF  SUFFICIENT  REASON 
OR  SANCTION  OF  VALUATION— Irs  RELATION  TO  CONCRETE 
NORMS  AND  ASSUMPTIONS 

i.  The  General  Problem. 

The   criterion   of  normative   objectivity,   of   reality   in   the 
only  sense  in  which  it  is  applicable  to  values  when  conceived 
omprehensively,    has    now    been    developed.     Its    implication 
with   respect   to   an   ultimate   postulate   of   worth   continuity, 
and  its  further  implication  that  sufficiency  is  always  relative 
a  specific  presupposition  of  reality,   includes,   I   think,   all 
hat   is   significant   in   the   pragmatic   criterion   and   furnishes 
the  key  to  the  ultimate  grounds  of  valuation.     In  the  concept 
of  sufficient  as   distinguished    from    absolute  ground  of  value 
and  m  the  definition  of  sufficiency  as  relative  to  specific  con 
tinuities  and  specific  presuppositions,  we  have,  moreover,  the 
point  of  connection  between  the  axiological  and  phenomenological 
omts  of  view— between  the  specific  presuppositions,  presump- 
ons  and  assumptions  as  genetically  derived,  and  the  presupposi 
tion  of  reality  as  logically  interpreted.     Our  task  is,  therefore, 
to  deduce  logically  a  system  of  values  from  one  absolute 
contestable  value,  but  rather  to  interpret  the  actual  demands 
d  presuppositions  of  valuations  axiologically,  i.e.,  to  show  the 
relation  of  the  actual  presuppositions  of  concrete  values  to  the 
ultimate  logical  presupposition  of  valuation,  the  degree  to  which 
continuity  of  valuation  is  realised  in  specific  ideals. 


Valuation  and  Evaluation  409 

Of  the  possible  meanings  of  the  claim  to  reality  or  normative 
objectivity,  only  that  of  inner  reality  and  truth  requires  special 
consideration.  In  the  other  cases  normative  objectivity  is 
identical  with  factual  and  truth  objectivity,  and  the  realisation 
of  the  presupposition  of  reality  requires  acknowledgment  in 
specific  existential  and  truth  judgments.  Only  by  such  con 
version  of  presumption  and  assumption  into  judgment  is  con 
tinuity  of  valuation  possible,  is  the  value  as  a  subjective  ex 
perience  well-founded.  The  test  of  the  well-founded  value  in 
this  case  presents  no  special  problem.  The  claim  to  reality 
is  specific  in  its  meaning,  and  the  illusions  which  follow  its  mis 
interpretation  are  manifest  to  all  who  tamper  unwittingly  with 
the  physical  and  economic  worlds.  But  with  the  claim  of 
intrinsic  ideals  to  inner  truth  and  reality  the  case  is  different. 
Here  validity  does  not  involve  complete  conversion  of  assump 
tion  into  already  determined  fact  and  truth ;  the  faiths  embodied 
in  such  ideals  need  not  be  completely  actualised  in  the  specific 
spheres  of  physical  and  social  reality.  What,  then,  are  the 
specific  tests  which  such  claims  must  fulfil  in  order  to  be 
accepted  as  well-founded  ? 

(a)  Inner  Truth  has  a  Reference  to  Inner  Existence  or  Psychical 
Matter  of  Fact. — The  Nature  of  this  Reference. 

The  fact  that  ideals  transcend  experience  in  the  sense  that 
they  do  not  demand  complete  conversion  into  factual  and 
truth  judgments,  i.e.,  continuity  at  every  point  with  the  objective 
world  of  science,  has  been  taken  to  mean,  as  we  have  seen,  that 
here  the  presupposition  of  reality  is  to  be  interpreted  as  an 
assumption  of  universality  and  eternity  of  the  values  involved. 
If  this  is  its  meaning,  then,  of  course,  there  are  no  empirical 
tests  of  sufficiency,  any  reference  to  actual  empirically  con 
ditioned  feeling  and  will  being  irrelevant.  But  this  interpretation 
of  the  presupposition  of  reality  we  have  found  to  be  a  logical 
distortion  of  its  meaning.  While  the  ultimate  postulate  of 
valuation  is  an  assumption  of  the  eternity  of  value,  it  does  not 
include  the  eternity  of  specific  objects  and  concepts  of  value; 
while  it  assumes  identity  with  an  over-individual  will,  it  does 
not  assume  that  identity  without  a  difference  comprehended 
in  the  concept  of  logical  universality.  Again,  while  ideals 
transcend  experience  in  the  sense  defined,  they  nevertheless 
have  their  roots  in  certain  regions  of  experience  and  fact. 


410  Valuation:   its  Nature  and  Laws 

Founded  as  they  are  upon  psychical  matter  of  fact,  they  contain 
an  implicit  reference  to  retrospectively  denned  reality  which 
must  be  made  explicit  and  subjected  to  test.  Can  the  nature 
and  measure  of  this  reference  be  denned,  and  can  this  definition 
be  applied  reflectively  to  the  evaluation  of  ideals  and  the  judg 
ments  based  upon  them-in  other  words,  to  the  differentiation 
of  valid  ideals  from  "  pathetic  fallacies  "  ? 

The  nature  of  this  reference  is  not  difficult  to  define  in  general 
terms.     Whenever  the  claim  to  objectivity  is  made  in  the  case 
ideal  values  and  their  norms,  this  claim  includes  the  belief 
that  they  are  realisable  in  some  fashion.     We  live  by  ideals 
it  is  true,  but  only  because  they  are  at  the  same  time  realities' 
An  ideal,  as  soon  as  it  becomes  practical,  is  an  anticipation  of 
•eality.     The  claim  to  universality  and  eternity  has  no  meaning 
until  in  that  claim  we  admit  the  temporal  anticipation  that 
with  the  actual  empirical  extension  or  persistence  of  the  ideal' 
there  will  be  a  corresponding  actual  fulfilment    in    empirical 
realisations,   that  the  postulate  of  continuity  will  be  realised 
in  its  two  aspects  of  acquirement  and  conservation.     But  antici 
pations  are  meaningless  and  groundless  unless  based  upon  some 
reference  to  fact,  and  in  this  case  the  "  matter  of  fact  "  is  what 
we  call  broadly  "human  nature,"  human  feeling  and  volition  in 
its  various  aspects  and  forms. 

This  very  general  characterisation  of  the  reference  to  matter 
of  fact  implied  in  all  ideals  can,  moreover,  be  made  more  definite 
tor  the  purposes  of  axiology.     Using  the  form  of  words  employed 
the  Introduction,  and  now  charged  with  a  fuller  meaning,  we 
may  say  :  every  assertion  of  value  involves  ipso  facto  an  assertion 
of  its  conformity  with  the  laws  of  feeling  and  will.     The  abstract 
formulations  of  the  normative  sciences,  their  ideals  and  norms 
cannot  be  anything  else  than  the  development,  in  other  terms 
and  for  other  purposes,  of  what,  from  another  point  of  view,  we 
call    psychological   laws.     As    now    understood,    the    empirical 
laws  of  valuation  have  axiological  significance  ;    it  is  through 
them  that  the  specific  empirical  demands  are  to  be  interpreted. 
For  the  processes  of  feeling  and  will,  in  which  ideals  are  con 
structed,  permanent  dispositions  are  formed,  and  the  presupposi 
tions  of  value  judgments  are  established,  are  real  processes  of  the 
real  will.   They  determine  the  specific  conditions  of  the  realisation 
of  the  met-empirical  will  presupposed  in  all  valuation  ;    they 
create  spheres  of  actual  and  possible  experiences  which  in  turn 
condition  the  inner  reality  of  ideals.     The  sufficient  reason  or 


Valuation  and  Evaluation  4 1 1 

sanction  of  ideals  must  therefore  include  the  question  whether 
they  are  axiologically  possible,  and  in  what  sense  and  within 
what  limits  they  are  possible. 

This  conception  of  the  relation  of  validity  to  psychical 
matter  of  fact  is  inevitable  for  any  one  who  holds  that  every 
actual  experience  of  value,  while  presupposing  objective  validity, 
is,  nevertheless,  in  one  aspect  a  subjectively  conditioned  feeling, 
that  every  judgment  of  value,  however  over-individual  its 
reference,  must  express  the  felt  realisations  of  some  subject. 
Nor  does  such  a  view  commit  its  holder  to  that  form  of  Psy- 
chologismus  which  identifies  continuity  of  feeling  as  a  mere 
psychical  fact  with  continuity  of  value  as  an  axiological  principle. 
It  merely  sets  him  in  opposition  to  the  view  which  holds  that 
realisation  and  satisfaction  are  unrelated  to  empirical  feeling. 
For  us  the  disjunction  between  logical  validity  and  psychical 
fact  is  not  complete  ;  the  genetic  conception,  with  its  implication 
of  the  reality  of  the  developing  processes  of  feeling  and  will — an 
implication  which  we  have  seen  underlies  all  appreciative  de 
scription  and  the  genetic  method  to  which  it  gives  rise — consti 
tutes  a  middle  ground.  Upon  this  middle  ground  our  solution 
of  the  problem  must  be  based. 

(b)  Axiological  Sufficiency  and  Possibility. 

The  sufficient  reason  of  a  judgment  which  contains  any 
implication  or  presupposition  of  reality  must,  as  Leibnitz  long 
ago  pointed  out,  include  a  reference  to  possibility.  In  addition 
to  possibility  every  such  judgment  must  have  a  further  mark, 
which  Leibnitz  describes  as  the  special  characteristic  of  suf 
ficiency — "  compossibility,"  or  compatibility  of  the  given  judg 
ment  with  other  established  judgments.  The  region  of  the 
possible  is  negative,  and  is  determined  by  the  application  of 
the  principle  of  non-contradiction  ;  that  of  compossibility  is 
more  positive,  and  is  determined  by  the  harmony  of  the  given 
concept,  judgment,  or  postulate  with  the  entire  system  of 
experience  and  the  ultimate  postulate  of  continuity  which 
underlies  it. 

That  sufficiency  must  include  possibility  and  compossibility 
is  clear,  but,  in  order  that  we  may  apply  these  conceptions  to 
the  region  of  worth  experience,  we  must  see  just  what  meaning 
they  have  in  connection  with  valuation.  That  there  may  be 
contradiction  between  facts  and  values  which  make  values. 


412  Valuation:   its  Nature  and  Laws 

at  least  in  the  form  originally  conceived  or  presented,  impossible, 
that  there  may  be  disharmonies  or  incompatibilities  between 
values  and  values,  between  ideals  and  ideals,  which  require 
the  elimination  or  reconstruction  of  ideals,  are  facts  which 
are  patent  to  all.  But  these  contradictions  and  disharmonies 
cannot  be  reduced  to  logical  terms.  As  it  is  impossible  to  say 
that  a  logically  necessary  conception  is  ipso  facto  a  felt  value, 
so  it  is  impossible  to  say  that  a  contradiction  in  feeling  and  will 
is  necessarily  reducible  to  abstract  logic.  What,  then,  is  the 
meaning  of  the  possible  and  compossible  in  the  sphere  of  valua 
tion,  and  what  is  its  axiological  interpretation  ? 

The  whole  question  is  evidently  one  of  relevancy.     What 
realms  of  fact  and  what  laws  are  relevant  in  the  determination 
of    axiological    possibility  ?     This    question    we    have    already 
considered  in  a  general  way  in  the  concluding  paragraphs  of 
Chapter  VI.     It  is  a  pertinent  question,  we  found,  to  ask  whether 
an  ideal  is  possible — whether,  for  instance,  to  consider  again 
the  illustration  there  used,  it  is  rational  for  a  man  to  desire 
indefinite  increase  of  a  valued  disposition  on  the  assumption 
that  with  this  increase  there  would  be  increase  of  value.     It  is 
not  pertinent,   however,   to   answer   the   question  by  showing 
that  the  assumption  is  in  contradiction  with  inferences  from 
the  principle  of  the  conservation  of  energy.     The  only  question 
that   is   relevant   is    whether    such    an    assumption    conforms 
to  inner  psychical  matter  of  fact,  to  the  laws  of  feeling  and  will. 
Again,  it  is  pertinent  to  ask  whether  certain  ideals  of  social 
good  are  possible,  whether,  for  instance,  the  will  to  universalise 
altruistic   dispositions,   or   to   make   the   maxim   of  one's  con 
duct  an   eternal   law,   presupposes   assumptions   which   are   or 
are  not  in  harmony  with  the  fundamental  facts  and  laws  of 
social  value.     The  assumption  in  this  case  is  that  amount,  i.e., 
frequency  of  repetition,  is   accompanied  by  corresponding  in 
crease  of  instrumental  value,  an  assumption  which  is  not  in 
harmony  with  the  laws  of  participation  value.     But  here  also 
there  is  only  one  type  of  existential  judgment  which  is  relevant 
—namely  judgment  concerning   the  psychical  dispositions  de 
termined  by  the  working  of  the  laws  of  sympathetic  participa 
tion.     In  general  it  may  be  said  that,  while  there  are  specific 
regions  of  inner  psychical  fact  which  must  enter  into  the  retro 
spective  evaluation  of  any  such  ideal  or  assumption,  there  are 
entire  regions  of  outer  fact  and  truth  which  are  totally  irrelevant, 
namely  the  causal  and  mechanical  generalisations  of  physical 


Valuation  and  Evaluation  413 

science.  They  are  without  competency  in  this  connection, 
because  they  are  formed  for  an  entirely  different  purpose,  and 
by  abstracting  wholly  from  the  experience  with  which  we  are 
here  concerned. 

The  question  of  compossibility  or  compatibility  of  value 
judgments  involves  something  more  than  possibility.  Here  the 
question  is  one  of  compatibility  of  the  presuppositions  of  one 
value  judgment  with  those  of  another ;  and  it  arises  in  all  those 
cases  of  judgment  and  volition  which  involve  transition  from 
one  norm  or  ideal  to  another — either  continuity  between  different 
types  or  levels  of  valuation,  or  the  reduction  of  norms  and 
standards  to  a  single  ideal.  The  working  of  the  fundamental 
laws  of  valuation  and  worth  construction,  the  value  movements 
from  lower  to  higher  levels,  create  those  appreciative  distinctions 
or  qualitative  differences  in  the  meanings  of  our  feelings,  which, 
while  not  absolute  and  a  priori,  nevertheless  are  practically 
absolute  in  the  sense  that  they  fix  the  limits  to  the  activity  of 
specified  ideals.  It  is  chiefly  in  the  attempts  to  universalise 
specific  ideals,  as,  for  instance,  the  ideals  of  pleasure,  perfection, 
and  self-sacrifice,  that  incompatibilities  appear. 

The  general  relation  of  possibility  to  compossibility  is  now 
clear.  In  the  former,  the  creation  of  psychical  disposition 
determines  the  possibility  or  impossibility  of  certain  feelings 
and  volitions.  In  the  latter,  the  development  of  dispositions, 
ideals,  and  levels  creates  certain  conscious  meanings  and  dis 
tinctions,  certain  universes  of  appreciative  discourse  which 
mutually  limit  and  determine  each  other.  In  the  main  it  may 
be  said  that  the  special  studies  of  the  different  types  of  value 
feelings  and  judgments — their  conditions,  laws,  and  limits — 
were  studies  in  axiological  possibility,  while  those  concerned 
more  specifically  with  the  problems  of  synthetic  preference  and 
valuation  dealt  with  problems  of  compossibility. 

Having  thus  distinguished  the  two  concepts  of  possibility 
and  compossibility,  it  is  possible  to  unite  them  again  in  a  more 
comprehensive  concept.  As  the  well-founded  ideal  value  is 
that  which  makes  possible  continuity  of  valuation,  so  the  ideal 
which  is  not  well  founded  is  that  which  develops  discontinuity. 
Where  impossibility  and  incompatibility  appear  the  process 
is  "  self-defeating."  In  this  concept  of  the  self-defeating  pro 
cess  we  have  a  negative  test  of  validity. 

The  full  significance  of  this  concept  is  apparent  only  when 
it  is  recognised  that  it  is  volitional  in  meaning  and  transcends 


4^4  Valuation:   its  Nature  and  Laws 

the  sphere  of  merely  logical  contradiction.     That  it  is  capable 
•i  almost  purely  logical  statement  is  evidenced  by  the  extensive 
use  of  the  principle  of  "  endless  regress  "  and  of  self-defeating 
activities  of  thought,  as  seen  especially  in  the  antinomies.     But 
even  here,  as  Bosanquet 1  has  pointed  out,  the  real  meaning 
ultimately    affective-volitional.     The    wearisome    vanity    of 
these  empty  repetitions  without  hope  of  unity  or  finality,  with 
out  repose  of  realisation  and  satisfaction  of  the  will,  is  what 
really  leads  to  value  movement  to  new  types  of  activity  and  ideal 
construction.     That  this  is  Kant's  ultimate  meaning  appears 
think,  from  the  close  connection  of  his  doctrine  of  antinomies 
with  his  practical  philosophy.     The  transition  from  the  agnosti 
cism  of  the  theoretical  to  the  faith  of  the  practical  reason  is  a 
value  movement  on  a  large  scale.     Be  this  as  it  may,  that  the 
•rincrple   is   essentially   one   of   volition   and   valuation   rather 
than  of  thought  and  logic,  appears  from  the  fact  that,  while 
s  we  shall  see  later,  all  logical  applications  of   the   principle 
are  referable  to  volitional   categories,   not   all   applications   to 
valuation  are  logical.     This  is  especially  apparent  in  the  so- 
called  hedonistic  paradox,  where  the  pursuit  of  pleasure  is  said 
to   be   self-defeating— not   because   of   a   logical   contradiction 
but  of  an  inherent  character  of  the  practical  will.     The  self- 
defeating  process  is  accordingly  one  which  arises  when  we  will 
an  idea  in  connections  in  which  the  ideal  is  untenable,  wherever 
there  is  demand  for  a  type  of  realisation  which  is  inherently 
impossible.     Correspondingly,    all    fallacies    of    valuation    arise 
in  the  acknowledgment  of  the  primary  presupposition  of  reality 
in  spheres  of  existence  and  truth  not  comprehended  or  implied 
in  this  presupposition. 

2.  Application  of  these  Principles  to  Specific  Problems. 

In  this  conception  of  axiological  possibility  and  compossibility 
we  have  the  sufficient  sanction  of  ideal  values,  the  sufficient 
reason  or  criterion  of  their  inner  truth.  The  application  of 
this  criterion  in  connection  with  the  specific  problems  which 
have  emerged  in  the  course  of  our  investigation,  and  which 
have  been  emphasised  at  the  beginning  of  this  chapter,  would 
involve  practically  the  rehearsal  of  the  entire  content  of  the 
preceding  discussion.  Though  tedious,  it  would  not  be  difficult 
to  show  that  the  entire  question  of  what  are  realisable  and 

1  Logic,  Vol.  I,  p.  173. 


Valuation  and  Evaluation  415 

well-founded  ideals,  in  the  various  forms  in  which  it  presents 
itself  in  the  economic  and  moral,  individual  and  social  life, 
must  ultimately  find  its  solution  by  reference  to  this  sort 
of  truth  and  rationality.  The  disputes  between  the  individualist 
and  the  socialist,  for  instance,  as  to  what  are  possible  and  im 
possible  motives,  practicable  or  impracticable  ideals,  imply 
certain  assumptions,  on  the  part  of  both,  which  can  be  tested  only 
by  the  application  of  such  axiological  criteria.  Thus  the  question 
of  the  desirability  of  certain  forms  of  distribution  of  economic 
goods,  of  the  extension  of  certain  moral  dispositions — their 
effect  upon  values — raises  the  further  question  of  the  relation 
of  individual  to  social  values,  whether,  more  specifically,  certain 
values  are  not  conditioned  by  their  being  exceptional  and  isolated 
by  personal  and  group  contrasts,  and  whether,  even  if  their 
extension  were  economically  and  psychologically  possible, 
such  a  process  would  not  be  axiologically  self-defeating  in  that 
they  would  then  be  incompatible  with  other  values. 

With  equal  truth  it  may  probably  be  said  that  whatever 
fallacies  appear  in  such  reasonings  about  and  anticipations  of 
values,  are  in  the  main  a  specific  axiological  form  of  the  material 
fallacies  of  composition  and  accident,  fallacies  which  arise  from 
what  may,  perhaps,  be  called  false  quantities  in  volition.     Thus 
the  assumption  that  a  motive  which  has  its  origin  in  a  limited 
participation  in  class  consciousness,  and  which  has  attained  its 
intensity  and  extension  within  that  class  by  the  fact  of  contrast 
and  opposition,  would,  if  universalised,  exist  in  the  same  clear 
ness  and  intensity,  is  one  which  requires  examination,  and  may 
possibly  contain  a  fallacy  which  vitiates  the  entire  argument  in 
which  it  appears.     When  the  socialist  argues  that  the  motives 
which  have  been  characteristic  of  special  classes,  such  as  the 
military,   artistic,   or   philanthropic,  could   be   extended   to  all 
industrial   activity,   he    may   be   right — there   may   be   certain 
universal  elements  unmodifiable  by  specific  function,  but  any 
prediction  of  the  form  they  would  assume,  or  of  the  results  of 
such  universalisation,  is  liable  to  errors  of  this  type.     Equally 
doubtful  is  any  argument  which  proceeds  upon  the  assumption 
that  the  ideals  which  actuate  men  in  exceptional  situations,  and 
which  are  perfectly  legitimate  expectations  in  the  sphere  of  per 
sonal  obligation  and  imputation,  are  susceptible  of  repeated  appli 
cation  in  normal  circumstances  or  of  universal  extension.    On  the 
other  hand,  we  have  the  converse  fallacy  when  we  apply  the 
laws  of  social  value  judgment  of  men  en  masse  to  individual 


4i 6  Valuation:   its  Nature  and  Laws 

situations  and  persons — when  we  seek  to  argue  from  economic 
and  sociological  generalities,  made  for  certain  purposes,  to  the 
nature  and  capacities  of  individual  originators  of  values. 

(a)  Again  the  Monistic  'theories  of  Ultimate  Value. 

But  with  such  special  problems  we  are  not  here  concerned. 
Our  object  is  rather  to  show  in  a  general  way  the  principles 
which  underlie  all  reflective  evaluation  of  ideals  and  values. 
Nevertheless,  while  we  cannot  take  up  this  criticism  in  detail, 
we  may  in  conclusion  profitably  return  to  a  reconsideration, 
from  our  present  point  of  view,  of  the  question  of  ultimate  ideals 
and  norms  as  presented  in  the  preceding  chapter.  To  such 
assumptions  of  ultimate  ends,  and  their  corresponding  norms, 
all.  the  special  assumptions  which  we  have  just  been  con 
sidering  in  turn  go  back,  and  the  question  of  their  validity 
or  workableness  is  closely  connected  with  the  more  ultimate 
question  of  the  validity  of  the  ideals  on  which  they  rest. 

In  all  reasonings  and  anticipations  of  the  type  described, 
some  single  ideal,  such  as  happiness,  self-realisation,  or  realisa 
tion  of  over-individual  social  good,  is  usually  presupposed, 
either  explicitly  or  implicitly.  These  ideals  are  implicitly 
assumed  as  norms  of  certain  types  of  judgments  of  obligation 
and  imputation,  but,  in  the  several  monistic  theories  discussed, 
each  has  also  been  taken  separately  as  the  absolute  ideal  of 
valuation,  as  the  single  explicit  end  of  all  volition.  Examina 
tion  of  our  actual  feelings  and  judgments  has  shown  us  that 
they  do  not  afford  ground  for  such  monistic  theories.  The 
method  of  testing  these  ideals  and  norms  employed  in  the  pre 
ceding  chapter  showed  that  they  break  down  at  the  limits. 
Now,  with  our  present  conceptions  of  possibility  and  compossi- 
bility,  we  may  see  why  these  single  ideals,  as  single  ideals,  are 
not  well-founded. 

In  general  it  may  be  said  that,  with  reference  to  certain 
situations,  in  connection  with  certain  processes  and  levels  of 
valuation,  each  of  these  ideals  is  well-founded,  in  that  the  funded 
meaning  finds  realisation,  in  that  the  postulate  of  continuity,  as 
we  have  defined  it,  finds  fulfilment.  These  ideals  or  assumptions 
prove  illusory  and  fallacious,  however,  when,  abstracted  from 
the  specific  processes  in  which  they  are  developed  and  which 
they  in  turn  control,  they  are  universalised  and  conceived  to 
hold  absolutely  and  unconditionally.  It  is  then  that  im- 


Valuation  and  Evaluation  417 

possibilities  and  incompatibilities  appear ;  and  they  appear 
just  because  the  very  act  of  taking  them  unconditionally  in 
volves  a  misinterpretation  of  the  presupposition  of  reality. 

These  general  principles  have  been  applied,  either  con 
sciously  or  unconsciously,  in  the  many  criticisms  of  Hedonism 
which  have  sought  to  show  its  self-defeating  character  and  to 
display  its  practical  fallacies,  but  they  have  significance  for 
us  only  in  their  general  logical  bearings,  and  these  can  be  shown 
in  a  few  words.  The  ideas  of  pleasure  or  happiness  are,  as 
we  have  seen,  ideal  constructions.  As  objects,  as  passive 
states,  they  are  abstractions,  the  product  of  a  process  of  abstrac 
tion  exercised  upon  our  "  condition "  worths,  including  the 
primary  condition  worths  together  with  the  complementary 
values,  ethical  and  aesthetic,  which  arise  on  that  level.  Now, 
within  certain  limits,  this  abstraction  has  a  basis  in  reality, 
i.e.,  the  desire  for  pleasure  is  well-founded.  When  we  believe 
pleasure  to  exist,  and  assume  that  it  may  be  attained,  our  belief 
and  assumption  are  justified  by  the  fact  that  up  to  a  certain  point 
the  assumptions  are  realised,  not  indeed  in  "  pleasures,"  but 
in  actual  worth  feelings.  This  "  point  "  is  determined  by  the 
nature  of  the  processes  and  laws  in  which  the  ideal  is  implicit 
or  embedded.  As  a  convenient  representation  of  objects  of 
condition  worth,  and  within  the  limits  of  the  laws  governing 
such  values,  it  is  a  well-founded  object  of  conation.  But  when 
applied  beyond  that  sphere — to  represent  personal  worths,  for 
instance,  its  capacity  as  an  object  of  continuous  valuation  dis 
appears,  as  may  be  seen  in  the  fact  that  it  is  not  a  substitute 
for  personal  worth.  No  less  fallacious  is  the  attempt  to  sub 
stitute  a  sum  of  pleasures  for  over-individual,  social  good;  for 
the  fulfilment  of  the  demand  for  this  good  does  not  include 
necessarily,  i.e.,  in  so  far  as  the  test  of  experience  can  show, 
the  realisation  of  the  demand  for  pleasure. 

The  self-defeating  or  paradoxical  character  of  Hedonism 
as  a  monistic  ideal  appears,  therefore,  at  two  specific  points. 
If  I  assume  pleasure  to  be  an  object  of  absolute  value,  in  the 
sense  of  a  perpetually  tenable  end,  my  experiences  will  probably 
prove  such  an  assumption  unfounded  and  illusory,  for  the  reason 
that  it  is  in  contradiction  with  the  laws  of  condition  worths 
in  connection  with  which  the  idea  alone  has  meaning.  The 
assumption  is  axiologically  impossible,  for  it  is  contrary  to  the 
laws  of  valuation.  On  the  other  hand,  if  I  proceed  upon  the 
assumption  that  pleasure  may  be  taken  to  represent  indis- 


2   E 


4i 8  Valuation:   its  Nature  and  Laws 

criminately  all  values  on  whatever  level,  I  am  equally  subjected 
to  practical  illusions  ;  my  assumption  turns  out  to  be  incom 
patible  with  those  fundamental  appreciative  distinctions  which 
have  been  genetically  developed,  and  is,  therefore,  axiologically 
incompossible.  In  both  cases  the  presupposition  of  reality  in 
the  ideal  of  pleasure  has  been  misinterpreted,  reality  has  been 
claimed  where  the  claim  cannot  be  realised. 

No  less  paradoxical  and  self-defeating  is  the  universalising 
of  the  ideal  of  over-individual  and  impersonal  good  with  its 
corresponding  norm,  for  here,  again,  such  a  process  involves 
a  misinterpretation  of  the  presupposition  of  reality.  The 
classical  criticisms  of  this  ideal,  as  for  instance  Schiller's  criticism 
of  the  Kantian  conception  of  obedience  to  an  absolute  law 
uncontaminated  by  individual  feeling,  and  Spencer's  criticism 
of  absolute  altruism,  emphasise  the  unreality  of  the  ideal,  its 
incompatibility  with  other  fundamental  values.  Wherein  does 
this  unreality  consist  ? 

Our  study  of  the  nature  and  origin  of  the  ideal  of  over- 
individual  and  impersonal  good  has  shown  it  to  be  also  an  abstrac 
tion  standing  for  certain  concrete  social  ideals  which  emerge  in 
the  processes  of  social  participation.  As  such  it  has  a  certain 
reality.  But  in  this  case  the  presupposition  of  reality,  the 
demand  for  realisation  implicit  in  the  ideal,  can  mean  only 
one  thing — namely,  that  type  of  outer  existence  described  as 
social,  which  consists  in  the  fact  that  the  ideal  as  subjective 
experience  finds  its  fulfilment  or  realisation  in  a  corresponding 
over-individual  and  social  demand.  If  it  cannot  be  thus  ful 
filled,  the  ideal  is  empty  and  lacks  reality,  and  the  judgments 
of  obligation  and  imputation  founded  upon  it  are  fallacious. 
Normative  objectivity  implies  factual  objectivity,  i.e.,  an  ideal 
in  order  to  have  reality  must  have  actual  participation  value. 
Such  ideals  are  well-founded,  therefore,  just  in  so  far  as  they 
are  in  conformity  with  the  laws  of  participation  value  which 
determine  both  the  nature  and  the  extent  of  the  demand  which 
gives  them  their  reality. 

When,  however,  the  subject  acts  and  judges  as  though 
the  presupposition  of  reality  or  claim  to  objectivity  means, 
not  this  factual  objectivity,  but  that  purely  logical  objectivity 
characterised  as  universality  and  eternity,  fallacies  both  of 
obligation  and  imputation  arise.  Ideals  and  their  norms  are 
abstracted  from  the  empirical  processes  of  emotional  participa 
tion  of  persons  and  groups,  and  instead  of  being  treated  as 


Valuation  and  Evaluation  419 

founded  upon  actual  desire  and  its  satisfaction,  they  are  held  to 
be  determined  by  a  wholly  over-individual  and  impersonal  will. 
The  self-defeating  character  of  such  an  assumption  appears  in 
the  fact,  as  shown  in  detail  in  preceding  chapters,1  that  the  ideal 
either  becomes  an  unreal  abstraction  or  passes  over  into  a  wholly 
individual  and  personal  value. 

We  are  finally  brought,  then,  to  the  question  of  the  reality 
of  the  personal  ideal  and  of  personal  values.  Here  the  claim  to 
absoluteness  takes  the  form  of  the  ideal  of  perfection  ;  "  per 
fection  is  eternal."  A  certain  paradoxical  and  self-defeating 
character  has  been  charged  to  this  ideal  also,  and  without  doubt 
it  may  give  rise  to  fallacies  both  of  judgment  and  action,  which 
must  engage  our  attention.  Our  first  task,  however,  is  to  deter 
mine  to  what  extent  it  is  well-founded,  what  is  the  true  inter 
pretation  of  the  presupposition  of  reality. 

Like  the  ideals  already  considered,  it  is  an  ideal  construction 
developed  in  certain  empirical  processes.  In  these  processes 
funded  meaning  is  acquired,  and  within  certain  limits  this 
funded  meaning  is  realised  in  certain  actual  feelings,  ethical, 
aesthetic,  and  religious.  The  forms  in  which  the  ideal  of  personal 
value  appears  are,  as  our  study  has  shown,  the  ideal  of  absolute 
sacrifice  at  the  characterisation  minimum  and  of  complete 
inner  harmony.  That  these  ideals  have  a  certain  basis  in  reality 
is  indicated  by  the  fact  that  they  form  the  presupposition  of 
the  very  real  experiences  of  personal  obligation  and  of  aesthetic 
realisations  of  intrinsic  values.  But  the  condition  of  the  realisa 
tion  of  these  values  is  the  isolation  of  the  personality.  The  ideal 
of  perfection  is  the  organising  principle  of  the  personal  series, 
but  it  is  realised  only  in  so  far  as  it  remains  implicit  in  that 
process.  The  presupposition  of  reality  or  the  claim  to  objectivity 
in  this  case  means  just  that  inner  continuity  and  reality  of  the 
ideal  which  we  have  found  to  be  the  ultimate  meaning  of  norma 
tive  objectivity.  In  so  far  as  the  ideal  remains  thus  an  implicit 
organising  principle  of  experience,  it  is  an  assumption  which 
includes  its  own  reality,  fulfilling  the  demands  of  possibility 
and  compossibility  alike,  as  our  study  of  the  laws  of  valuation 
in  this  sphere  has  shown.  The  personal  obligations  felt,  and 
the  aesthetic  and  religious  values  imputed  on  the  basis  of  the 
belief  in  perfection,  are  not  pathetic  fallacies,  but  the  highest 
realities.  But  let  this  meaning  of  reality  be  misinterpreted, 
let  us  seek  to  convert  this  inner  reality  into  outer  reality  through 

1  Chapters  xn  and  xm. 


420  Valuation :   its  Nature  and  Laws 

existential  and  instrumental  judgments,  and  illusions  and  un 
realities  appear. 

Of  these  false  applications  of  the  ideal  of  personal  perfection, 
the  one  most  commonly  emphasised  is  that  which  appears  when 
we  make  the  implicit  presupposition  or  norm  of  personal  obliga 
tions  and  imputations  an  explicit  end,  single  and  conscious, 
to  which  other  ends  and  values  are  sacrificed.  In  this  case 
the  process  is  self-defeating,  and  pathetic  sentimentalities  and 
unrealities  appear.  Self-forgetfulness  is  no  less  the  condition 
of  perfection  than  of  happiness,  and  those  single  experiences 
which  come  by  the  grace  of  the  gods  are  not  to  be  repeated  at 
will.  The  quasi-aesthetic  moment  of  isolation,  deeply  as  it  is 
rooted  in  reality,  is  not  to  be  petrified  into  a  permanent  attitude 
without  loss  of  value  and  reality.  The  fallacy  of  the  situation 
is  again  misinterpretation  of  the  presupposition  of  reality. 
In  this  case  the  subject  of  the  illusion  assumes  that  realisation 
of  the  ideal  means  its  actualisation  in  certain  psychical  states. 
Assuming  that  reality  means  psychical  existence,  an  intrinsic 
ideal  is  turned  into  a  means  to  ends  ;  it  becomes  instrumental 
to  certain  psychical  experiences,  and  the  inevitable  consequence 
is  the  "backward  value  movement"  to  the  hedonic  accompani 
ments  of  the  ideal.  No  less  fallacious  is  that  other  application 
of  the  ideal  in  which  the  assumption  of  its  reality  is  interpreted 
as  demand  for  existence  in  the  social  sense.  As  social  ideals 
and  norms  become  empty  and  unreal  when  abstracted  from 
the  empirical  processes  of  participation  in  which  they  are  con 
structed  and  which  they  control,  so  here  expectations  generated 
in  the  ideal  construction  of  the  personality  do  not  necessarily  find 
fulfilment  in  social  demand. 


(b)  Inferences  from  the  Application  of  Axiological  Principles 
to  Special  Problems. 

From  this  axiological  criticism  of  ultimate  assumptions 
two  conclusions  may  be  drawn.  In  the  first  place  the  working 
out  of  our  principle — that  the  sufficient  sanction  of  an  ideal 
with  its  assumption  of  reality  is  always  relative  to  the  specific 
meaning  of  reality  implied,  has  shown  us  that,  while  it  is  the 
nature  of  ideals  to  transcend  experience,  in  that  they  are  not 
completely  convertible  into  factual  and  truth  judgments,  never 
theless  they  are  real,  and  control  experience  only  in  so  far  as 


Valuation  and  Evaluation  421 

they  are  well-founded  anticipations  of  experience.  Some  refer 
ence  to  experience  already  organised  is  necessary. 

But  a  second  conclusion  of  even  greater  importance  may 
be  drawn — namely,  that  the  general  principle  of  indifference 
of  judgments  of  value  to  judgments  of  truth  and  fact  at  certain 
points  is  substantiated  by  an  examination  of  the  grounds  of 
different  judgments.  This  relative  independence  is  seen  first 
of  all  in  the  indifference  of  judgments  of  value,  where  the  pre 
supposition  of  reality  means  one  kind  of  fact  or  truth,  to  other 
judgments  of  value  implying  another  kind  of  fact  or  truth, 
where,  to  be  more  explicit,  distinctions  of  inner  and  outer  truth 
are  drawn.  Thus  a  certain  indifference  of  personal  and  social 
values  has  shown  itself  at  various  points,  and  has  become  so 
constant  as  to  be  almost  of  the  nature  of  a  fundamental 
principle,  into  the  meaning  of  which  we  may  well  inquire  more 
fully. 

If  we  view  social  values  and  their  mutations  from  the  ex 
ternal  point  of  view  of  mere  matter  of  fact,  we  must,  as  we  have 
seen,1  judge  certain  values  to  be  normal,  others  to  be  aspiring, 
and  still  others  to  be  outlived.  But  to  the  individual  who 
casts  in  his  lot  with  any  of  these  values,  who  realises  himself 
in  identifying  himself  with  them,  our  judgments  of  fact  are 
relatively  indifferent.  The  reactionary  judges  or  assumes 
what  you,  as  an  impartial  spectator,  have  inferred  to  be  an 
outlived  value,  to  be  of  absolute  worth,  and  if  he  realises,  in 
sacrifice  of  self  for  it,  the  highest  personal  worth,  that  is  his 
test,  and  your  existential  judgments  are  irrelevant.  The  same 
may  be  said  of  a  reformer  who  throws  himself  into  an  aspiring 
worth,  and  even  of  the  normal  man  who  finds  harmony  and 
peace  in  living  the  life  of  his  day. 

On  the  other  hand,  it  is  equally  true  that  much  that  is 
of  significance  from  the  personal  point  of  view  is  irrelevant 
for  objective  social  values  as  a  system  of  forces.  The  fate  of 
the  personal  worths  is  a  matter  of  indifference.  They  seem  to 
play  the  role  of  epiphenomena.  The  working  of  the  laws  of 
personal  worth  construction  may  produce  the  values  of  inner 
peace  and  tragical  elevation  in  connection  with  narrow  group 
worths  and  outlived  values  which,  from  the  point  of  view  of 
the  factual  judgments  of  the  objective  system,  are  merely 
individual  in  their  meaning,  and  luxuries  which  have  no  ap 
preciable  instrumental  value,  or  which,  if  they  have  any  value 

1  Chap.  XI,  pp.  333  ff. 


422  Valuation:   its  Nature  and  Laws 

at  all  through  social  imitation,  are  practically  negligible.  Often 
they  appear  as  pathetic  fallacies,  resting  upon  judgments  and 
assumptions  which  do  not  conform  to  the  system  of  values 
judged  as  a  system  of  fact.  Our  valuation  of  extremes  of 
altruism  is  often  possible  only  by  an  abstraction  of  the  person 
from  his  relations  to  society.1 

But    the    relative    indifference    discoverable    at    this    point 
becomes  still  more  marked  when  we  leave  the  sphere  of  ethical 
and  social  judgments  and  consider  judgments  of  value  in  their 
relation  to  the  more  neutral  or  "  wertfrei"  judgments  of  science. 
If  there  is  indifference  at  those  points  where  judgments  of  value 
and  fact  seem  most  closely  related,  it  is  needless  to  say  that 
the  indifference  will  become  much  more  pronounced  where  the 
scientific    constructions    and    formulas    are    still    more    remote 
from  value  judgments,  and  abstract  still  more  from  appreciation 
in  their  descriptions.     The  indifference  of  judgments  of  value 
to   the   constructions    of   science   becomes   progressively   more 
marked  as  we  pass  from  the  centre  to  the  periphery,  from  social 
and   economic   to   physiological,  biological,  and  physical   con 
structions.     It  is  only  when  the  abstract  conceptions  of  science 
are  really  not  abstract,  when  they  still  contain  an  appreciative 
connotation,    that    conflicts    arise.     Only    when    science    uses 
symbols  which  have  a  worth  connotation,  when  she  talks  of 
the  abysses  of  space,  of  fate,  of  the  reign  of  law,  of  the  struggle 
for  existence,  of  survival  of  the  fittest,  etc.,  does  she  come  into 
contradiction    with    values.     And    these    contradictions    arise 
precisely  because  when  she  thus   talks   another  language  she 
becomes  progressively  more  and  more  symbolic  and  equivocal. 
But  this  indifference  is  always,  it  must  be  remembered,  relative 
and  genetic,  not  absolute. 

V.  CONCLUSION 
i.  Existence — Truth — Value. 

The  development  of  a  point  of  view  in  evaluation  distinctive 
enough  to  merit  the  special  term  axiological,  or,  in  other  words, 

1  For  a  fuller  discussion  of  the  point,  see  the  writer's  two  papers,  The  Individual 
and  the  Social  Value  Series,  Philosophical  Review,  Vol.  XI,  Nos.  2  and  3.  Also 
Ehrenfels:  System  der  Werttheoric,  Vol.  II,  p.  153.  An  impartial  observation  of  the 
empirical  data,  Ehrenfels  confesses,  shows  us  that  the  concepts  "  socially  valuable,"  and 
"individual-ethical"  are  only  partially  and  occasionally  identical,  that,  as  a  matter 
of  fact,  "there  are  certain  dispositions  and  actions  which  come  under  the  concept  of  the 
individual-ethical  which  from  the  standpoint  of  social  morality  must  be  designated 
indifferent." 


Valuation  and  Evaluation  423 

of  a  special  principle  of  Sufficient  Reason  or  Sanction  of  valuation 
and  value  judgments,  seems  to  have  found  its  justification  in 
the  concrete  problems  which  it  both  reveals  and  solves.  The 
fact  that  judgments  of  value  have  as  their  "  objective  "  a  reality 
not  fully  exhausted  by  factual  and  logical  objectivity  is  made 
clear  by  the  merely  partial  and  relative  identities  of  normative 
with  factual  and  truth  objectivity,  and  by  the  principles  of 
practical  estimation  of  values  which  the  recognition  of  this  fact 
brings  about. 

But  while  we  have  in  these  axiological  principles  the  grounds 
for  the  practical  distinction  between  subjectively  and  objectively 
conditioned  values,  between  practical  reality  and  illusion, 
there  is  still  unquestionably  something  to  be  desired  from  the 
more  theoretical  point  of  view.  The  concepts  of  existence 
and  reality,  of  truth  and  value,  are  left  sufficiently  un 
related  to  give  some  ground  for  the  criticism  that  the  full  im 
plication  of  our  discussion  of  the  relations  of  these  concepts  is 
still  to  be  developed.  Somewhere,  it  will  be  said,  there  must 
be  a  point  of  ultimate  anchorage,  a  point  of  complete  identity 
between  reality — the  "  objective  "  of  judgments  of  value  and  fact 
alike,  and  one  or  the  other  of  these  subordinate  concepts,  an 
identity  so  complete  and  definite — of  reality  with  value  or 
reality  with  truth — that  in  the  one  case  all  facts  and  truths 
may  be  seen  to  be  forms  of  value,  or,  in  the  other,  all  values 
aspects  of  truth.  Ultimately  we  must  face  this  last  question  : 
Does  all  truth  and  fact  rest  upon  an  incontestable  value,  an 
absolutely  tenable  attitude  of  will,  or  do  values  rest  ultimately 
upon  an  undeniable  truth  the  opposite  of  which  is  unthinkable  ? 
Is  the  ultimate  statement  of  the  situation  to  be  in  intellectualistic 
or  voluntaristic  terms  ? 

That  some  answer  to  this  question  is  implied  in,  and,  in  a 
sense  perhaps  demanded  by,  our  preceding  discussions,  can 
scarcely  be  denied.  And  there  can  be  no  doubt  that  the  whole 
drift  of  our  discussion  has  been  in  the  direction  of  a  concept 
of  the  primacy  of  value  and  of  the  value  judgment,  in  the  sense* 
that  the  "  objective  "  or  intent  of  the  predicates  of  reality  is 
always  value,  and  that  the  existential  and  truth  judgments 
are  but  special  forms  of  valuation.  Let  us,  then,  in  conclusion, 
seek  to  find  the  logical  ground  of  this  inference,  to  take  this 
last  logical  step. 


424  Valuation  :   its  Nature  and  Laws 

2.  The  Meaning  of  the  Concepts  of  Indifference  and  Relevancy. 
Evidently   the   point   in   our   discussion   where   the   logical 
implications  were  not  completely  drawn  is  to  be  found  in  our 
conceptions  of  relative  indifference  of  value  and  truth,  and  of 
merely  partial  identity  of  normative  with   factual  and  truth 
objectivity.     To   some  such  position   we   were   driven  by  our 
analysis,   and  with   the   admission  of  the  fact  we  apparently 
accepted  the  old  dualism  between  appreciation  and  description 
in  a  new  though  none  the  less  serious  form.     Nor  would  such 
an   inference   be   entirely   without   foundation,   although   there 
is,  it  should  be  observed,  a  significant  difference  between  such 
a   principle   of    indifference    formulated    at    the    close    of    our 
investigations,   and   the    antithesis    between    appreciation    and 
description  which  from    the    beginning  makes  impossible  such 
description  of  worth  experience.     But  as  a  matter  of  fact  this 
is  not  an  adequate  view  of  the  situation,  for  the  principle,  as 
we  have  formulated  it,  has  two  aspects,  a  positive  as  well  as 
a  negative.     Not  only  does  it  warn  us  against   deducing  the 
reality  of  values  from  the  outer  truth  or  existence  of  the  objects 
valued,  but  it  also  tells  us  that  value  itself  involves  an  inner 
truth  and  reality  not  describable  in  these  terms. 

The  negative  significance  of  this  principle  has  been  fully 
developed.  It  tells  us  explicitly  that  the  complete  intent  or 
meaning  of  the  presupposition  of  reality  cannot  be  exhausted 
in  any  abstract  definitions  of  existence  and  truth,  and  that, 
therefore,  the  predicates  of  reality  as  used  in  valuation  cannot  be 
wholly  tested  by  such  norms.  Its  value  is  methodological  and  as 
fundamental  for  axiological  method,  I  believe,  as  the  principle 
of  psycho-physical  parallelism  for  psychology.  As  the  latter 
tells  us  that  the  psychical  can  never  be  reduced  to  the  physical, 
nor  the  physical  to  the  psychical— so  long,  at  least,  as  we  remain 
on  the  plane  of  scientific  description,  so  the  axiological  principle 
of  indifference  warns  us  not  to  seek  to  reduce  ah1  values  to 
factual  and  truth  objectivity,  all  worth  experiences  to  mere 
effects  of  social  processes  or  means  to  social  ends,  i.e.,  again,  so 
long  as  we  remain  on  the  axiological  plane. 

But  the  positive  implications  are  even  more  significant. 
The  very  reason  for  this  injunction  not  to  identify  reality  with 
existence  and  truth  is  that  the  intent  or  meaning  of  the  pre 
supposition  of  reality  transcends  any  retrospective  definitions 
of  existence  and  truth.  Analysis  of  the  predicates  of  reality 


Valuation  and  Evaluation  425 

in  their  extension  and  intension,  has  shown  us,  not  only  that 
the  meaning  of  these  predicates  is  varied,  but  also  that  acknowl 
edgment  of  the  presupposition  of  reality  in  existential  and  truth 
judgments  is  always  relative  to  the  specific  intent  of  the  pre 
supposition.  To  put  this  into  more  general  form  for  our  present 
purposes,  we  may  say — the  criterion  of  knowledge  as  such  is 
always  relative,  and  arises,  not  out  of  the  relation  of  a  special 
intent  to  an  absolute  intent,  but  from  relations  of  special  contents 
to  special  intents.  Error  and  fallacy  consist  in  discrepancy 
between  special  content  and  intent.  Attempts  to  generalise 
these  proximate  intents,  to  formulate  concepts  of  "  truth  in 
general  "  in  such  a  way  as  to  make  it  all  inclusive  and  to  give 
it  intrinsic  meaning,  serve  only  to  emphasise  their  relativity. 
When  truth  is  denned  as  "  correspondence  of  idea  with  reality," 
as  "  contradictionless  experience,"  or  even  pragmatically,  as 
"  instrumentality  of  ideas  for  life,"  such  formulas  prove,  as  we 
have  seen,  to  be  either  too  narrow  to  fulfil  the  intent  of  truth, 
or,  when  they  are  sufficiently  broadened  to  satisfy  our  sense 
of  rationality,  so  little  self-sufficient  as  to  force  us  beyond  con 
cepts  of  existence  and  truth  to  modes  of  immediate  experience 
or  of  intrinsic  appreciation  which  they  presuppose. 

We  are  thus  led  to  the  conclusion  that  judgments  of  existence 
and  truth,  far  from  being  exhaustive  of  the  intent  of  reality, 
themselves  have  axiological  meaning  only  in  so  far  as  they  serve, 
by  acknowledgment  of  fulfilment  of  a  special  intent,  to  lead  to 
that  identity  of  the  empirical  with  the  met-empirical  will  which 
constitutes  the  essence  of  value.  In  other  words,  they  are  not 
predicates  of  total  systems  of  experience.  Judgments  of  exist 
ence  and  truth  apply  only  to  relations  among  our  impressions 
and  ideas.  It  has  been  well  said  that  the  totality  of  experience 
is  as  little  describable  as  existent  or  true  as  the  totality  of 
matter  is  describable  as  heavy. 

But  if  we  admit  that  existence  and  truth  cannot  be  exhaust 
ively  interpreted  except  with  reference  to  concepts  or  postulates 
of  valuation,  and  therefore,  that  value  may  logically  be  prior  U 
existence  and  truth,  it  still  remains  an  open  question  whether  the 
identity  of  value  with  reality  can  be  so  characterised  as  to  make 
the  relations  of  value  to  existence  and  truth  intelligible.  It  may 
be  that  totalities  of  experience  can  be  only  appreciated,  that 
only  values  can  be  the  predicates  of  such  total  experiences. 
As  we  found  it  impossible  to  characterise  a  total  experience 
as  absolutely  true,  it  may  be  equally  difficult  to  say  what  would 


426  Valuation  :   its  Nature  ana  Laws 

make  a  total  experience  absolutely  valuable.  Now,  for  value 
and  reality  to  be  identical  means  that  an  experience,  in  order  to 
be  real,  must  be  satisfying.  Could  we  say  what  the  object  of 
absolute  satisfaction  is,  there  would  be  no  longer  any  points  of 
indifference  ;  we  could  deduce  the  nature  of  truth  and  existence 
from  it.  Such  definition  is,  as  we  have  seen,  impossible,  but  it  is 
not  inconceivable  that  we  may  be  able  so  to  describe  ap 
preciatively  our  ultimate  values  as  to  show  that  they  transcend 
truth  and  existence  and  include  them.  It  may  be  possible 
for  "  the  inspirations  of  reason,  appreciative  of  values,"  to  use 
Lotze's  fine  phrase,  so  to  apprehend  the  ultimate  nature  of  the 
will  and  its  satisfaction  as  to  enable  us  to  adumbrate  that  in 
contestable  value  in  the  light  of  which  we  can  pass  from  the 
object  of  thought  to  its  reality. 

Lotze's  own  well-known  attempt  so  to  apprehend  ultimate 
value  leads  him  to  find  this  final  satisfaction  in  the  feeling  or 
consciousness  of  harmony,  including  all  those  forms  of  activity 
in  repose,  of  unity  and  continuity,  describable  as  love,  beauty, 
perfection.  In  these  experiences — the  highest  inspirations  of 
reason — reason  finds  its  own  inmost  essence  ;  in  them  is  realised — 
in  practically  absolute  moments,  the  assumption  of  identity 
of  the  empirical  with  the  met-empirical  will.  It  would  be 
"  intolerable  "  that  these  ideals,  formed  in  the  activities  of  ideal 
construction,  should  have  no  existence,  power,  or  validity  in 
the  world  of  reality.  They  are  true  he  holds  :  we  can  feel  them 
when  we  cannot  think  them.  From  these  ultimate  appreciations 
he  then  deduces  the  relative  validity  and  value  of  those  special 
forms  of  continuity,  between  inner  and  outer  existence,  inner 
and  outer  truth,  which  are  the  indispensable  conditions  of  this 
realisation. 

Now  it  may  well  be  granted  that  these  inspirations  of  reason 
far  outrun  its  reasoned  convictions,  and  that  this  insight  into 
the  identity  of  value  and  reality  cannot  make  itself  wholly 
intelligible.  Our  handling  of  the  speech  of  valuation  is  still 
•inept,  with  its  cnide  distinctions  of  feeling  and  knowledge,  and 
its  unworthy  slavery  to  the  prejudicial  connotations  of  the  term 
feeling.  It  may,  indeed,  be  further  granted  that  precisely  at 
this  point,  in  his  use  of  the  concepts  of  feeling  and  thought, 
Lotze's  presentation  of  the  concept  requires  decided  modification. 
But  with  all  these  admissions,  it  remains  true  that  his  funda 
mental  insight  into  the  identity  of  value  with  reality  and  the 
priority  of  value  to  truth  and  existence  remains  incontestable. 


Valuation  and  Evaluation  427 

This  doctrine  of  priority  means,  when  properly  interpreted, 
that  worth  experience  in  its  entirety  corresponds  to  a  larger 
world  of  reality  than  the  limited  regions  of  existence  and  truth. 
There  are  forms  of  harmony  of  intent  with  content  of  experience 
which  are  not  comprehended  under  the  specific  meanings  of 
these  terms.  The  standards  of  knowledge  are  but  special 
formulations  of  the  ideals  of  unity  and  continuity ;  and  of  the 
total  realm  of  ideal  objects  judged  or  assumed  to  be  real,  only 
a  limited  number  will  conform  to  these  standards,  these  specialised 
demands.  Many  do  not  thus  conform,  and  yet  contribute  to 
the  unity  and  continuity  of  affective-volitional  meaning.  The 
ideal  makes  itself  felt,  and  finds  satisfaction  in  many  ways 
which  fail  to  conform  to  these  definitions.  It  is  no  less  true 
that  the  standards  of  value  are  special  and  relative  formulations 
of  this  demand.  There  are  forms  of  harmony  not  exhaustively 
interpreted  in  terms  of  the  realisation  of  any  of  the  ends  to 
which  these  standards  refer,  forms  of  harmony  in  which  the 
experience  transcends  ideal  construction  of  ends  no  less  than  of 
truths.  This  we  have  fully  recognised  in  the  preceding  chapter 
in  our  conception  of  supreme  moments  in  which  distinctions  of 
ends  and  norms  lapse.  But  this  merely  serves  to  emphasise  the 
priority  of  value.  Practical  absolutes  may  indeed  come  at  the 
limits  of  volition,  relative  values  may  be  sublimated  into  absolute  ; 
but,  so  far  as  I  know,  a  relative  truth  may  become  absolute  only 
by  becoming  the  presupposition  of  an  absolute  value. 

To  the  final  question,  therefore— how  are  judgments  of  exist 
ence  and  truth  related  to  ultimate  appreciations  with  their 
intrinsic  values  ? — our  answer  is  simple  and  evident.  They 
establish  relations  between  experience  and  experience,  between 
idea  and  idea,  which  lead  to  new  appreciations  or  conserve  those 
already  acquired.  The  specific  predicates  of  existence  of  truth, 
already  considered,  have  meaning  only  when  they  add  to  the 
intrinsic  value  or  reality  of  an  impression  or  idea.  There  are 
cases  where  the  acknowledgment  of  the  presupposition  of  reality 
in  these  predicates  does  add  value  and  reality  to  the  idea,  e.g.-,- 
where  the  presupposition  of  reality  already  includes  a  retro 
spective  explicit  definition  of  existence  or  truth.  But  such 
addition  is  not  a  necessary  consequence  of  such  acknowledgment, 
nor  is  such  acknowledgment  a  necessary  condition  of  appreciation. 
As  the  critics  of  the  Ontological  Proof  have  clearly  seen,  the  proof 
of  the  truth  or  existence  of  the  idea  of  the  perfect  being  adds 
nothing  to  its  reality.  In  the  intrinsic  appreciation  of  the  mean- 


428  Valuation:   its  Nature  and  Laws 

ing  of  the  demand  or  postulate  of  such  a  Being,  whatever  reality 
:  has  is  already  included.     Its  central  place  in  the  spiritual  life 
its  value,  must  guarantee  its  reality,  and  the  attempt  to  translate 
such  ultimate  reality  into  the  subordinate  concepts  of  outer 
existence  and  truth  can  certainly  add  nothing  to  that  reality 
and  may  even,   by  falsely  interpreting  the  presupposition  of 
reality,  lead  to  error  and  unreality. 

This  is  the  final  point  to  which  our  axiological  study  brings 
The  implications  of  valuation  when  fully  worked  out  lead 
to  the  claim  of  priority.  Yet,  since  there  is  always  the  further 
implication  of  an  inner  truth  which,  while  not  exhausted  in  the 
predicates  of  existence  and  truth  already  considered,  is  still  not 
in  ultimate  contradiction  with  them,  since,  in  other  words,  life 
and  experience  show  themselves  progressively  more  and  more 
capable  of  statement  as  a  system  of  truth,  there  remains  always 
the  assumption  of  the  ultimate  inteUigibility  of  every  value. 
A  still  higher  form  of  experience  in  which  the  two  claims  are 
equaUy  satisfied,  a  form  of  contemplation  which  transcends  wiU 
and  thought  alike,  must  ever  be  the  goal,  acknowledged  or 
unacknowledged,  of  all  metaphysics.  Such  a  state  of  equilibrium 
would  indeed  be  the  Beatific  Vision.  Whether  it  is  attainable 
or  not,  and  if  attainable  capable  of  being  held,  it  is  not  for  us 
to  say. 


THE   END 


INDEX 


Absolute  values,  19,  335,  342  if.,  3635., 
370,  380,  384  f.,  396  ff.,  402,  406,  425  ff. 

Absolutes,  practical,  150,  277,  295,  374, 
376 ff.,  382,  399,  404,  427 

Act,  cognitive,  as  related  to  dispositions, 

5i»  53,  64,  95 

^Esthetic  feeling,  31,  45,  53,  251 

^Esthetic  mode,  presupposition  of,  45,  7°» 
122,  220,  235  n.,  240,  275 

/Esthetic  values,  of  simple  appreciation, 
205,  217,  221,  224,  229;  of  character 
isation  of  the  person,  266,  277,  296, 
306 ;  in  social  participation,  334 ;  validity 
of,  3°9,  399,  422  ff. 

Affective  continuity  (substitution,  subsump- 
tion,  transition),  121,  130 

Affective  generalisation,  77,  120,  131, 
I33f.,  244 

Affective  abstract  (See  Affective  general 
isation.) 

Affective  memory,  H3f.,  116,  131 

Affective  sign,  105  f.,  112,  129,  241 

Alogism,  395 

Alter,  the  ego  and  the,  211  ;  as  ideal  con 
struction,  234,  248,  261 

Altruism,  egoism  and,  nature  of  distinc 
tion,  261,  264  n.  ;  as  personal  worth, 
301  ;  as  social  value,  314,  352  f.  ;  limits 
of,  359,  374  f- 

Appreciation  and  description,  6ff.,  17,  19, 

57,  424 
Appreciation,  intensity-less,  I28ff.,  274; 

and  hedonic  redundancies,  76,  162,  166, 

185.     (See  also,  Intensity-less  feelings.) 
Appreciation,  simple,  30,  32,  192  ;  objects 

of,    30 ;    meanings   of,    59,    67 ;    value 

movements  of,  199,  234 
Appreciative  description,  nature  of,  8,  14, 

55  ;  of  feeling  and  will,  83  f.  ;  terms  of, 

59  ff.  ;  and  scientific,  57,  62 
Apprehension,  imageless,  125 
Apprehension,  of  inner  life  in  others,  237 


Art,  and  aesthetic  experience,  166  ;  origins 
of,  222  ;  the  element  of  order  in,  225  ff. 

Aspiring  values,  333 f.,  337.  (See  also 
Value  Movements.) 

Assumption,  as  cognitive  act,  38,  42,  47, 
52  ;  as  presupposition  of  worth  attitudes, 
70,  137  f.  ;  implicit  and  explicit,  48,  70, 
115,  132;  in  affective  memory  and 
generalisation,  115,  131  f.;  in  value 
movement,  I99f.  ;  in  Einfuhlung  and 
sympathetic  participation,  233,  245,  259 

Assumption,  implicit,  role  of  in  charac 
terisation,  280,  290 ff.,  309;  in  social 
participation,  313  f.,  316,  334f.,  356  f., 
362 

Assumption-feelings,  48,  115,  118,  138, 
150,  162,  166,  iSiff.,  245.  (See  also 
Feelings  of  imagination  and  Intensity- 
less  appreciation. ) 

Augustine,  399 

Axiology  and  Axiological  method,  l6f., 
24,  1 88,  309 f.,  384 

Axiological  sufficiency,  405,  411.  (See 
Valuation  and  Evaluation. ) 


B 

Baldwin,  J.  M.,  on  genetic  method,  15  ; 
on  cognitive  and  existence  meanings, 
28 n.,  47 n.,  389 n.;  on  semblant  mode, 
70,  127  ;  on  Einfuhlung,  234,  235,  248 

Bentham,  157 

Bernoulli,  157 

Bohm  Bauwerk,  367 

Bosanquet,  B.,  414 

Brentano,  F. ,  on  judgment,  44;  on 
hedonic  redundancies  and  intensity- 
less  feeling,  57,  76,  159;  on  feeling  and 
will,  90 


Characterisation,   values  of,  30,  71,  234, 
260 f.,  284.     (See  Personal  worths.) 


429 


430 


Valuation :   its  Nature  and  Laws 


Complementary  Values,  law  of,  145,  I73ff., 
I78ff.  ;  in  relation  to  other  laws,  151, 
229>  173;  in  economic  valuation,  174; 
in  extra-economic  valuation,  175 ;  in 
simple  appreciation,  193,  214,  229;  in 
characterisation,  266  f.,  306;  in  social 
participation,  340,  348 

Condition,  worths  of,  definition,  30,  147  f., 
190,  197,  294  ;  and  value  movement, 
2O5»  2I3>  2295  and  of  person,  261,  264, 
269,  286,  300 f.,  367,  372,  376 

Consumption,  economic,  143,  160,  170; 
as  modified  by  other  values,  177,  213, 
229 

Continuity  of  value,  postulate  of,  15,  404 ff. 

Correct,  the.    (See  Norm  of  Participation) 

D 

Degree  of  worth  or  value,  72,  74,  76,  108, 
152;  measurement  of,  79,  156 f.,  171, 
282  ff.,  288 ff.,  291,  322 

Demand  and  Supply,  in  economics,  142  f.  ; 
in  personal  worths,  292  ;  in  social 
values,  318,  327,  329 

Depth  and  Breadth,  of  feeling,  50,  72  ff., 
155  ;  as  determined  by  disposition,  50, 
76,  183,  206,  209,  217;  in  the  person 
ality,  264,  269 f.,  286,  296,  304,  371. 
(See  Degree  of  worth. ) 

Desire  and  value,  35  ff.,  70,  82,  85,  94, 
148  ff. 

Diminishing  Utility.  (See  Diminishing 
Value. ) 

Diminishing  Value,  law  of,  145,  156,  158  ; 
extent  of  application  of,  167,  171  ff, 
181  ;  as  related  to  value  movement,  145, 
*93>  203  5  in  personal  worths,  272  ff.  ; 
in  social  values,  326  ff. 

Disposition,  concept  of  in  definition  of 
value,  33 f.,  37,  50,  53,  77;  in  genetic 
method,  I4f.,  82,95,  IO4 ;  and  implicit 
assumption,  121,  127,  133,  138,  248, 
258  f. 

Disposition,  ethical  and  moral,  233,  248 f., 
258f.,  26off.,  283ff,  291,  294,  298, 
311  ff.,  342,  351  f.,  37Sf.  (See  Norm 
of  Characterisation  and  Participation.) 


Economics,  and  theory  of  value,  3,  311, 
367;  laws  of,  142  f.,  169,  174,  177 

Economic  values,  definition  of,  143,  311  ; 
thresholds  of,  146  ff.  ;  laws  of,  169, 


174;  as  related  to  other  values,  213 ff., 
229,  311 

Ehrenfels,  Ch.  von,  on  definition  of  value, 
35  ff.  ;  on  feeling  and  will,  86  ;  on  laws 
of  valuation,  156,  i86ff,  330  f.  ;  on 
value  movements,  195,  333  ;  on  moral 
obligation,  350  n.,  397,  422  n. 

Einfuhlung;  and  valuation,  72,  102, 
234  ff.  ;  psychology  of,  236,  238,  241, 
255 ;  ethical  and  aesthetic,  245,  252 ; 
and  real  feelings,  240,  249  ff.,  273. 
(See  Participation,  sympathetic.) 

Elsenhans,  T.  R.,  129 

Emerson,  335 

Emotion,  as  feeling  of  value,  64  ;  analysis 
of,  88,  97,  100  ff,  119 

Emotional  Logic,  76  f.,  H2f.,  121,  139, 
203  f. 

Ethical  and  Moral  values,  285,  314.  (See 
also  Personal  and  Impersonal  values.) 

Ethical  scepticism,  5,  370 

Ethics  and  theory  of  value,  4,  368,  380 ff., 
416 

Existence,  judgment  of,  42 ;  as  presuppo 
sition  of  feeling  of  value,  38  ff,  66,  80, 
95.  "8,  133,  162,  199,  248;  as  related 
to  other  presuppositions,  47,  51,  66, 
69 ff.,  77  f.,  94.  (See Judgment-feelings.) 

Existence,  meanings  of,  22,  387  ff.  (See 
Presupposition  of  reality. ) 


Fechner,  159,  180 

Feeling,    analysis    of,    62,    93  f.,    103  f.  ; 

dimensions  of,   59 ff.,   62,    83,    looff.  ; 

presuppositions  of,  41  ff,  47,   64,   66; 

genetic  relations  of,  51  ff.,  104  ;  theories 

of,  57,  82,  94,  96  ff. 
Feeling  and  value,  35,   39 ff,   53,   64 f., 

95 
Feeling  and  will,  theories  of,  57,  83,  85, 

89 ff.,  93  ff. 
Feelings  of  value,  presuppositions  of,  35  ff, 

47,  62,  66 
Feelings  of  the  imagination,  in  valuation, 

1 15,  133  ff. ,  138  ;  in  Einfithlung,  245  ff, 

250  ff.  ;  in  ethical  judgment,  263,  276. 

(See  Assumption-feelings. ) 
Flournoy,  Th.,  277 


Genesis  and  Validity,  6,  16,  384,  424  f. 
Genetic  method,  in  worth  analysis,  44  ff, 
68,  81,  137,  191,  234,  252 


Index 


Genetic  levels  of  valuation,  51,  190,  230, 
282  ff.,  314,  366,  379,  384 

Genetic  Theory  of  feeling  and  will,  89, 
104 

Gestalt-qualitat  ("form  of  combination" 
of  elements),  of  emotional  complexes, 
100,  no,  119,  132,  238  ff.  ;  and  com 
plementary  values,  179  ff.  ;  in  aesthetic 
characterisation  of  persons,  225  ff. ,  266  f. 

Giddings,  F.,  325 

Gossen,  157,  166 

Group  Segregation.  (See  Social  Differen 
tiation.  ) 

Gross,  K.,  217,  240,  251 

Guyau,  J.  M.,  210,  342,  397 


H 

Habit,  and  disposition,  5°f-»  83;  an(^ 
feeling  of  value,  5of.,  83,  103 ff.,  120, 
133,  161,  183  f.  ;  and  affective  abstract, 
133,  158;  and  implicit  assumption,  48, 
67,  107,  121,  133,  249,  259;  dulling  of 
sensitivity  with,  158,  161,  183,  203, 
323  ;  and  normal  expectation,  293, 
333 ff.,  356,  361 

Hedonic  redundancies,  of  feelings  of  value, 
77,  108,  129  f.  ;  as  affected  by  repeti 
tion,  163,  166,  273  ;  role  of  in  valua 
tion,  155,  185,  197,  261.  (See  In 
tensity-less  appreciation.) 

Hedonism,  ethical,  84,  370,  417 

Hegel,  203 

Herbart,  397 

Him,  Yrjo,  74,  222 

Hoffding,  H.,  20,  65,  394,  404 

Hoffler,  A.,  152 


Idealisation,  of  the  person,  263,  265  f., 
275  ff.  ;  of  social  wholes,  342  ff. ,  363  f. 

Impartial  Spectator.  (See  Impersonal  at 
titude.) 

Impellent  mode.     (See  Obligation.) 

Impersonal  attitude  in  valuation,  27  f., 
232  f.,  313  f.,  350 f.,  375  f.,  378 

Impersonal  values.  (See  Over-individual 
values.) 

Imputation,  of  merit  and  demerit,  233 ; 
emotional  and  intellectual,  290,  315 ; 
personal  and  ethical,  260  ff. ,  291,  298  ff. , 
307,  309  ;  social  and  moral,  290,  293, 

3°2.  314,  356,  358  ff.,  375  f- 
Imputed  value,  22  f.,  174,  177,  214 


Inner  Peace,  279,  307  f.,  344,  380.  (See 
Tragical  Elevation.) 

Intensity-less  feelings,  34,  77,  106,  128 

Intuitionism,  ethical,  369  ff. 

Isolation  (or  detachment),  aesthetic,  266, 
278  ff.  ;  as  presupposition  of  personal 
values,  285 ff.,  296 f.,  299,  301,  307, 
312  ;  of  social  groups,  339,  344 f.,  359 


James,  William,  12,  137  n.,  126,  182,  375, 

397,  399 

Jevons,  S.,  149 

Judgment-feelings,  38 ff.,  47,  53,  64, 
93  ff,  138,  248.  (See  Existence,  judg 
ment  of.) 

K 

Kant,  205,  278,  285,  336,  414 
Kraus,  Oskar,  157,  172 
Kreibig,  J.  C.,  27,  86,  156 
Kruger,  F.,  39,  50 ff.,  153 


Leibnitz,  396,  407,  411 

Limiting  Value,  law  of.    (See  Diminishing 

Value.) 

Lipps,  Th.,  49 ff.,  264 
Lotze,  130,  426 

M 

Marginal  Participation  Value,  law  of, 
329  f.,  332;  and  social  value  move 
ments,  333  ;  and  moral  judgment,  350 

Marginal  Utility,  law  of,  149,  156,  i69ff., 
1 74>  327.  33°-  (See  Valuation,  Laws  of. ) 

Marshall,  H.  R.,  217 

Meaning,  worth  as  affective-volitional,  15, 

26,  30  ff.,  93  f.;   acquired  and  funded, 
8,  82,   niff.  ;  "common,"  232,  253;, 
recognitive,  113,  117,  248;  generic,  121, 
I25»  J39>  248;  of  existence,  22,  24,  55, 
387 

Meinong,  A.,  on  definition  of  value,  35  ff., 
41  ff.  ;  on  judgment  and  assumption, 
138;  on  ethical  and  moral  judgment, 

27,  290  n.  ;  on  moral  obligation,  350  n. 
Merit  and  Demerit.     (See  Imputation.) 
Minimum  of  Characterisation,  294,  297, 

3°4,  377 
Minimum  of  Existence,   148  ff,  214,  294, 

353,  373,  377 
Minimum  of  Participation,  357,  361,  374 


432 


Valuation :   its  Nature  and  Laws 


Modes  of  consciousness  of  value,  55,  67  ff. 
Monistic  theories  of  value,  368,  380,  416 
Moral  judgment,   the,  313,   350 ff.,   356. 

(See  Impersonal  attitude.) 
Moral  obligation.     (See  Obligation.) 
Munsterberg,  H.,  n,  73,  94,  183,  403 


N 

Nietzsche,  F.,  2,  192,  342,  352  n. 
Norm  of  Characterisation,  293,  299  f.,  304 
Norm    of    Participation,    293 n.,    333 f., 

356  f. 
Normal  values,    333  f. ,  391.     (See  Value 

Movements  (social).) 
Normative  and  Descriptive,  l6ff. ,  384 ff., 

424  f. 
Normative  objectivity.     (See   Objectivity 

of  values. ) 

O 

Objectivity  of  values,  17,  22  f.,  187  f., 
309 f.,  384 ff.  ;  relation  of,  to  fact 
and  truth,  386,  390  ff.  (See  Sufficient 
Reason  and  Well-founded  value. ) 

Objects  of  value,  classification  of,  29 

Obligation,  analysis  of,  68,  207  ;  quasi- 
ethical,  209  ;  ethical  and  personal,  297, 
303  ff.  ;  social  and  moral,  360  ff ;  perfect 
and  imperfect,  285,  304,  364  ;  conflicts 
of,  373  f.,  376,  379 

Ontological  Proof,  and  primacy  of  values, 

395.  427 

Organic  sympathy,  244,  273,  322  f. 
Outlived    values,    333 f.,    338,    341,    354. 

(See  Value  Movement  (social).) 
Over-individual   values,   30,   311  f.,    366; 

laws   of,    328 ff.,    342;    estimation    of, 

350  ff. 


Participation,  sympathetic,  nature  and 
laws  of,  234,  238,  244  ff.  ;  extensive  and 
intensive,  253  ff.,  320  f.  (See  Einfiih- 
lung  and  Sympathy. ) 

Participation,  values  of,  definition  of,  30, 
71,  254 ;  personal  and  impersonal, 
253  ff.  ;  subjective  and  objective,  261, 
316,  328 f.  ;  laws  of,  320,  326,  328 ff., 
333,  342 

Pathetic  fallacies,  310,  359,  410 

Patten,  S.  N.,  197  f. 

Paulhan,  F.  R. ,  on  genetic  theory  of 
affective  attitudes,  105  ff.  ;  on  affective 


memory,     113;    on    "affective    sign," 

124,  129. 
Personal    worths,     definition     of,     270 ; 

origin  and  laws  of,  263  ff,  270,  277  ; 

objects   of,    283  ff.  ;    norms   and  limits 

of,  291  ff,,  348,  359,  361,  373,  376,  419 
Personality,  and  value,  49, 262 ;  ideal  of,  as 

assumed  in  judgments  of  personal  worth, 

262  ff.,    291  ;    construction    of,  248   f., 

263  ff.  ;  idealisation  of,  266,  275.     (See 
Isolation,  aesthetic.) 

Person-project,  248,  263 

Pleasantness-unpleasantness,  and  feeling 
of  value,  39,  50,  57,  62  f.,  96,  108, 
114;  and  the  laws  of  valuation,  142, 
152,  158,  166,  176,  184;  and  aesthetic 
values,  217  ff.  ;  and  personal  worth, 
50,  261,  300,  368  f.,  417 

Poe,  the  Raven,  123,  140 

Practical  Absolutes.  (See  Absolutes,  prac 
tical.  ) 

Pragmatism  and  pragmatic  criterion,  396, 

398 

Praise  and  blame.     (See  Imputation.) 
Presumption    of    existence,    as    cognitive 

attitude,  43,   477  ff.,  as  presupposition 

of  feelings  of  value,  43,  66,   190,   245; 

and  implicit  assumption,  52,  1 18,  257 
Presupposition   of  reality,    in   feelings  of 

value,  24  ff,  38  f.,  47,  53,  66;  ultimate 

meaning  of,  400  ff. 
Presuppositional  method,  as  form  of  the 

genetic,  14  ff.,  81,  384. 
Projection.     (See  Einfiihlung.) 
Psychology  of  valuation,  the,  9  f.,  14 
Puffer,  Miss  EM  219 


Quasi-ethical  and  quasi-moral.    (See  Ethi 
cal  and  Moral  and  Obligation.) 


R 

Reality-meanings.  (See  Meaning  and  Pre 
supposition  of  reality.) 

Religious  value,  346  ff. ,  394  f. 

Ribot,  Th.,  on  affective  memory,  113, 115  ; 
on  affective  generalisation,  124,  129 

Royce,  J.,  62,  126,  234 


Sanction,   ethical,    122 ;    Sufficient. 
Sufficient  Reason  of  valuation.) 


(See 


Index 


433 


Satiety,  law  of,  1648".  ;  and  appreciation, 
183,  203,  221,  237;  and  sympathetic 
participation,  273,  323 

Saxinger,  R.,  119,  1331". 

Schiller,  F.,  418 

Schwartz,  H.,  58,  77,  85,  192  f.,  369 

Self-realisation,  theory  of,  370,  419  f. 

Semblant  mode.     (See  ^Esthetic  mode. ) 

Sensitivity,  dulling  of,  159 ff.,  161,  185; 
in  appreciation,  203,  221,  231  ;  in 
sympathetic  participation,  240,  273, 
323-  (See  Habit.) 

Sensitivity,  feeling  as,  96  ff. 

Shaw,  B.,  352  n. 

Sigwart,  46 

Simmel,  G.,  on  ethical  theory,  12,  370; 
on  feeling  of  value,  57,  69,  96  ;  on  laws 
of  valuation,  171  ff.  ;  on  definition  of 
obligation,  69,  207  ;  on  practical  abso 
lutes,  280  n.  ;  on  social  differentation, 
256  ;  on  moral  obligation,  350  n. 

Social  Differentation,  and  valuation,  322, 
339 

Spencer,  H.,  418 

Spinoza,  348  n. 

Stevenson,  R.  L.,  361 

Stbrring,  G.,  62 

Stout,  G.  F.,  44,  ioo,  125,  128,  155 

Stuart,  H.  W.,  31,  112 

Stumpff,  C.,  155 

Sufficient  Reason  of  valuation,  395,  4oof., 
406  f. 

Sympathy,  social,  322  ff. 

Synergy,  social,  318,  331,  351,  357 

Synthetic  preference,  366,  368 ff.,  371, 
378  f. 

T 
Taylor,  A.  E.,  36 in.,  370 

Tennyson,  Lotos-Eaters,  123,  140,  276 

Thresholds  of  value,  economic  and  extra- 
economic,  146  ff.  ;  as  modified  by  com 
plementary  values,  ethical  and  aesthetic, 
J51*  215,  230;  in  personal  worth, 
293  ff.  ;  in  over-individual  value,  356  f. 

Tolstoi,  342,  373 

Tragical  Elevation,  279,  295,  298,  302, 
30(i>  375.  380-  (See  Inner  Peace.) 

Transgredient  and  immanental  references 
of  feelings,  to  dispositions  presupposed, 
0,  II,  60,  69 f.,  71,  150;  in  simple  ap 
preciation,  201,  206,  210,  216;  in 
personal  values,  269,  277,  303,  307  ;  in 
impersonal  values,  320,  344,  346  ;  axio- 
logical  meaning  of,  401,  404,  410 
2  F 


Truth,  meanings  of  in  evaluation,  389  ff., 
394,  409,  422  ff. 

U 
Universality   in   morals.      (See  Absolute 

values.) 
Utility,    values    of,    30,    143,    169,    311. 

(See  Diminishing  Value  and   Marginal 

Utility.) 


Valuation,  nature  of,  21  ff.,  142  ;  levels  of, 
51,  147,  190,  230,  314,  366,  379,  384 

Valuation,  laws  of,  79,  1426'.,  145,  156, 
167  ff.,  i73ff.,  185  ff.,  193,  202,  27off., 
320 ff.,  328,  37L379,  384 

Valuation  and  Evaluation,  relations  of, 
24,  61,  i85ff.  309  f.,  384  ff.  (See 
Normative  and  Descriptive. ) 

Value,  definition  and  analysis,  21  ff.,  95; 
avctuajl_jyid_jji4juted,  22  f.,  174,  177, 
214,  386  ;  intrinsicaijiU-mstimnental, 
22,  148,  I52,~i(xfr,  174,  195,  20o7^6i, 
26gff.,  311,  327,  342  f.,  386,  391, 
396  ff.,  400;  primajjLJH%d--dejjved,  67, 
71,  192;  rear"and  ideal,  22 f.,  309, 
342 ff.,  386~- 

Value,  the  judgment  of,  as  assertory,  21  f.  ; 
types  of,  31  ;  the  subject  of,  27  ;  objects 
of,  28  ;  relation  of  to  judgments  of  fact 
and  truth,  384  ff, ,  422  ff. 

Value,  theory  of,  I  ff. 

Value  Movements,  15,  67,  191  ff.  ;  and 
laws  of  valuation,  145,  193,  202  f.  ;  of 
simple  appreciation,  67,  199;  of  sympa 
thetic  participation  and  characterisation, 
232,  252  ;  social,  333  ff. 

Vcblcn,  T.  B.,  12,  256 

Voluntarism,  368 f.,  403,  422 ff. 


W 

Well-founded  value,  the  criterion  of,  19, 
185,  309,  405  ff.,  409,  413,  416  ff. 

Wieser,  F.  von,  3 

Witasek,  St.  von,  31,  53,  101,  235,  250 

Worth.     (See  Value.) 

Worth  predicates,  22  ff.,  55  ff.,  78 f.,  233, 
282  f.,  285,  352  ff. 

Worth  suggestions  of  feeling,  61,  64 

Wundt,  W.,  on  three-dimensional  theory 
of  feeling,  63  ;  on  feeling  and  will, 
90  ff,  97,  ioo;  on  law  of  "resultants," 
I  So,  204 


PRINTED    BY 

WILLIAM    BREXDON    AND   SON,    LTD. 
PLYMOUTH