Skip to main content

Full text of "Historical studies in philosophy"

See other formats


UNIVERSITY  OF  ILLINOIS 
LIBRARY. 


Class 


MrlO-20M 


on  or  before 
-  stamped  below, 
charge  is  made  on  all 
books. 

U.  of  I.  Library 


OCT  8'37 

JUNt5'3i 
OCT  £U|S4 
L  29 


JW.  27 1346 


OC!  88H943 


-2 


31 


SEf>  2  0  1989 
08 


11148-S 


HISTORICAL   STUDIES   IN  PHILOSOPHY 


MACMILLAN  AND  CO.,  LIMITED 

LONDON  •  BOMBAY  •  CALCUTTA 
MELBOURNE 

THE  MACMILLAN   COMPANY 

NEW  YORK  •  BOSTON  •  CHICAGO 
DALLAS  •  SAN  FRANCISCO 

THE  MACMILLAN  CO.  OF  CANADA,  LTD. 

TORONTO 


HISTORICAL  STUDIES 


IN 


PHILOSOPHY 

BY 

EMILE    BOUTROUX 

MEMBER    OF    THE    INSTITUTE 
PROFESSOR    OF   THE    UNIVERSITY   OF    PARIS 


AUTHORIZED  TRANSLATION 

BY 

FRED    ROTHWELL,    B.A. 


MACMILLAN    AND   CO.,   LIMITED 
ST.   MARTIN'S    STREET,    LONDON 

1912 


COPYRIGHT 


AUTHOR'S    PREFACE 

THIS  book,  which  a  careful  and  exact  translation  now 
introduces  to  English  readers,  is  a  collection  of  studies, 
independent  of  one  another,  and  written  mostly  at  the 
express  invitation  of  my  pupils  or  colleagues.  As  I 
had  devoted  numerous  lectures  at  the  fccole  Normale 
Superieure  and  the  S or  bonne  to  a  minute  analysis  of 
various  philosophical  systems,  and  as  I  was  unable  to 
find  time  to  set  them  forth  in  detail,  the  suggestion  was 
made  that  I  might,  at  all  events,  publish  a  resume  of 
the  conclusions  at  which  I  had  arrived.  My  first 
impulse  was  to  decline  a  task  which  seemed  somewhat 
ingrate  and  risky.  Indeed,  it  is  no  easy  matter  to  act 
upon  the  mind  of  the  reader,  unless  you  lead  him, 
to  some  extent,  along  the  paths  you  have  already 
traversed,  on  to  the  position  you  yourself  have  reached. 
And  even  though  he  may  not  feel  inclined  to  make 
your  conclusions  his  own,  he  will  regard  them  with  a 
certain  amount  of  indulgence  if  he  sees  that  you  have 
been  at  considerable  pains  to  reach  them.  Frequently, 
too,  quite  apart  from  your  results,  he  will  fully 

V 


26630 


vi  STUDIES  IN  PHILOSOPHY 

appreciate  the  investigations  you  have  made.  As 
Victor  Hugo  said  : 

.  .  .  Dieu  benit  1'homme 
Non  pour  avoir  trouve,  mais  pour  avoir  cherche. 

It  may  be  that  there  are  critics  good  enough  to  be  less 
severe  than  God,  but  if  scarcely  anything  beyond  results 
are  set  forth,  the  reader  will  not  feel  inclined  to  show 
you  much  indulgence  ! 

In  spite  of  these  scruples,  the  reason  I  consented  to 
write  these  short  studies,  almost  devoid,  as  they  are,  of 
the  critical  basis  on  which  they  are  constructed,  is  that 
I  cannot  accustom  myself  to  the  idea — often  tacitly 
admitted  if  it  be  not  openly  avowed — that  in  dealing 
with  the  history  of  philosophy  all  inquiry  into  the  true 
and  inmost  thought  of  a  master  can  never  be  other 
than  premature,  and  that  a  genuine  savant  should 
concentrate  his  attention  upon  the  search  after  texts, 
their  comparison  with  one  another,  and  the  discussions 
to  which  they  give  rise.  I  am  aware  that  Fustel 
de  Coulanges,  one  of  our  great  historians,  has  said  that 
a  single  hour  of  synthesis  presupposes  centuries  of 
analysis.  His  own  admirable  generalisations,  however, 
form  a  magnificent  contradiction  of  this  formal,  abstract 
doctrine.  Perhaps  we  should  regard  as  more  truly 
in  conformity  both  with  the  conditions  of  scientific 
research  and  with  the  method  actually  adopted  by 
Fustel  de  Coulanges,  the  following  maxim  of  an 
eminent  geographer,  Professor  W.  M.  Davis,  of 


PREFACE  vii 

Harvard  University  :  Invention  may  advisedly  go  hand 
in  hand  with  observation.  The  analytical  study  of  the 
texts  suggested  to  me  a  certain  interpretation  of  the 
works  of  the  masters,  just  as  certain  previous  hypo- 
theses as  to  the  meaning  of  these  works  had  served  me 
as  heuristic  principles  in  the  analytical  study  itself: 
some  of  the  best  established  results  are  given  in  the 
present  volume.  I  am  disposed  to  regard  them  as 
nothing  more  than  starting-points  for  subsequent 
research,  ideas  to  be  tested  anew  and  revised  by  a  com- 
parison with  facts  ;  still  they  form,  in  some  measure, 
the  living  synthesis  of  several  of  my  previous  studies. 

Moreover,  I  do  not  think  it  will  ever  be  possible, 
in  setting  forth  a  system  of  philosophy,  to  content 
oneself  with  collecting  extant  documents,  manipulating 
them,  and  finally  extracting  their  substance,  by  quite 
mechanical  processes,  after  the  fashion  of  a  chemist. 
One  of  our  cleverest  linguists,  M.  Michel  Breal,  in  his 
famous  work,  Essai  sur  la  Stmantique,  is  altogether 
opposed  to  the  opinion  that  language  possesses  an 
existence  of  its  own,  and  is  capable  of  being  studied 
per  se,  independently  of  the  living  mind  of  man  who 
is  continually  building  it  up  and  perfecting  it.  "  Under- 
neath the  phenomena  presented  by  language,"  he  con- 
cludes, "  we  feel  the  action  of  a  thought  releasing 
itself  from  the  form  to  which  it  is  chained  down  .  .  . 
Mens  agitat  molem" 

A  fortiori  is  this  the   case  with  systems  of  philo- 


Vlll 


sophy,  which,  assuredly,  are  something  more  than 
a  blind  impulse,  an  enthusiastic  aspiration  of  the  mind  : 
they  represent  the  methodical  effort  of  an  intelligence 
employing  all  its  knowledge  and  dialectic  power  in 
an  attempt  to  confine  reality  within  clear  and  well- 
connected  formulas.  Still,  the  living  mind  of  the 
philosopher  is  never  absent  from  his  work ;  the 
system  should  never  be  conceived  as  separate  from 
its  creator,  like  fruit  as  distinct  from  the  tree  on 
which  it  has  grown.  Consequently,  in  order  to  under- 
stand an  author's  work  in  the  way  he  meant  it  to  be 
understood,  i.e.  to  understand  it  aright,  we  must  make 
it  our  constant  endeavour  not  merely  to  search  into  the 
visible  letter  of  the  text  and  all  the  details  of  docu- 
ments, but  also  to  think  and  live  with  the  author 
himself,  to  enter  into  his  spirit.  In  reality,  it  is  this 
interior  principle  of  development, — which,  in  truth, 
cannot  be  isolated  from  the  visible  forms  but  rather 
governs  them  and  gives  them  their  particular  aspect,— 
it  is  the  active,  ever-present  soul  of  the  author,  that 
the  historian  should  endeavour  to  set  before  us, 
enabling  us  to  enter  intuitively,  as  it  were,  into  that 
soul  and  attain  to  direct  participation  therewith. 

If  this  effort  be  made,  our  understanding  of  the 
work  becomes  as  profound  and  adequate  as  we  can  make 
it.  Nor  is  this  all.  Just  as  we  must  enlarge  our  own 
mind,  if  we  would  thus  comprehend  thoughts  greater 
than  our  own,  so  it  would  seem  that  to  cultivate  the 


PREFACE  ix 

history  of  philosophy  in  this  way  is  not  only  to  learn  to 
know  philosophers  but  also  to  become  more  capable  of 
philosophizing  ourselves.  To  what  heights  might  we 
not  aspire,  what  claims  might  we  not  make,  if  some- 
thing of  the  genius  of  the  masters  could  really  live 
again  within  us  and  enter  into  our  thought  !  "  Das  ist," 
said  Goethe,  "  die  Eigenschaft  des  Geistes,  dass  er  den 
Geist  ewig  anregt."  Is  not  this  submission — in  a  happy 
blend  of  effort  and  abandon — to  the  potent  influence  of 
the  masters,  this  attempt  to  carry  on  their  work  and 
message,  the  natural  and  legitimate  object  and  end  of 
our  historical  investigations  ? 

£MILE  BOUTROUX. 

PARIS,  March  7,  1912. 


TRANSLATOR'S    NOTE 

THE  word  science  throughout,  and  especially  in  the 
essays  on  Socrates  and  Aristotle,  has  been  frequently 
used  both  in  this  translation  and  in  the  original  work, 
to  render  the  word  a-o<j>ia,  as  being  preferable  to  wisdom 
and  to  sagesse  alike.  It  is  therefore  to  be  interpreted 
as  connoting  the  highest  branches  of  learning  and 
philosophy. 


CONTENTS 

PAGE 

THE  HISTORY  OF  PHILOSOPHY      ......          I 

SOCRATES,  THE  FOUNDER  OF  MORAL  SCIENCE  ...          8 

ARISTOTLE  .         .          .         .  .          .          .         .74 

JACOB  BOEHME,  THE  GERMAN  PHILOSOPHER  .         .          .         .169 

DESCARTES.         .          .          .         .          .          .         .         .         .234 

KANT .  .     255 

INDEX         ...  ......      331 


THE   HISTORY   OF  PHILOSOPHY 

i  rb  iSiov.  —  ARISTOTLE,  Eth.  Nic.  i.  7.  1097  b  35. 


THE  more  historical  works  on  every  subject  multiply, 
the  more  difficult  does  it  become  to  find  agreement  as 
to  the  object  of  history  itself.  Can  the  science  studied 
by  a  Renan,  when  investigating  the  moral  laws  of  man- 
kind and  the  universe,  be  the  same  as  that  studied  by 
a  Fustel  de  Coulanges,  who  is  ignorant  as  to  the  very 
existence  of  historical  laws,  and  whose  sole  ambition  is 
to  connect  a  few  facts  with  their  immediate  causes  ? 

The  history  of  philosophy  cannot  escape  this  con- 
dition of  things.  Hegel  understands  it  in  a  far  different 
fashion  from  Grote.  In  turn,  it  is  philosophical, 
psychological,  social,  philological  and  naturalistic,  nor 
do  we  clearly  see  what  definite  form  it  tends  to  assume. 
It  has  become  necessary  for  any  one  who  undertakes 
this  study,  unless  he  wishes  to  confine  himself  to  some 
particular  current  of  thought,  to  reflect  upon  the  end 
of  this  science  and  examine  the  various  definitions  that 
may  be  given  thereof. 

What,  then,  is  the  proper  object  of  the  history  of 
philosophy  ?  What  is  the  most  suitable  method  to 
adopt  ? 

Have  we  simply  to  collect  and  classify,  geographic- 
ally and  chronologically,  such  facts  as  may  rightly  be 
called  philosophical  ? 


2  STUDIES  IN  PHILOSOPHY 

Once  this  selection  is  effected,  have  we  to  connect 
each  of  these  facts  with  the  particular  environment  in 
which  it  happened,  and  also  with  its-  conditions  or 
causes  ? 

Or  rather,  if  we  consider  that  philosophy,  up  to  a 
certain  point,  has  an  existence  and  development  of  its 
own,  and  constitutes  a  kind  of  organism,  shall  we  un- 
ravel, as  it  were,  and  follow  up  this  autonomous  de- 
velopment through  the  apparently  capricious  inventions 
of  individuals  ?  Shall  we  consider  each  philosopher  as 
the  more  or  less  docile  instrument  of  an  immanent  and 
universal  spirit?  Has  our  task  to  consist  in  finding 
and  completing  those  parts  of  each  thinker's  work  which 
are  productive  and  likely  to  live,  and  neglecting  those 
which,  sooner  or  later,  time  must  destroy?  Is  it  not 
expected  that  a  historian  should  read,  study  and  criticise 
everything,  so  that  he  may  relegate  to  the  waters  of 
oblivion  such  events  as  have  no  claim  upon  the  memory 
of  mankind  ? 

But  if  we  have  scruples  about  thus  judging  of 
philosophical  productions  in  the  name  of  the  more  or 
less  mystical  idea  of  an  eternal  philosophy,  should  we 
not  like,  at  all  events,  to  distinguish  those  conceptions 
of  a  man  of  genius  in  which  he  is  really  himself  and 
is  introducing  innovations  or  preparing  the  future,  from 
those  in  which  he  shows  himself  nothing  more,  at  that 
stage,  than  an  echo  of  his  predecessors  ? 

In  short,  is  there  not  a  conception  of  the  history  of 
philosophy — a  very  plausible  one  by  reason  of  its  con- 
nection with  the  positive  sciences — according  to  which 
it  seems  to  be  the  historian's  task  to  take  philosophers, 
not  philosophy,  as  the  object  of  his  investigations  ;  and, 
by  a  process  of  psychological  analysis,  to  show  with 
reference  to  each  of  them,  the  line  of  evolution  he  has 


THE  HISTORY  OF  PHILOSOPHY          3 

of  necessity  had  to  follow — taking  into  consideration 
his  temperament,  his  education  and  the  circumstances  of 
his  life — in  the  production  of  the  ideas  he  has  given  to 
the  world  ? 

Evidently  each  of  these  points  of  view  has  an  interest 
and  importance  of  its  own,  but  none  of  them  seems 
to  be  the  special  point  of  view  of  the  historian  of 
philosophy. 

To  confine  oneself  to  the  collection  and  chronological 
arrangement  of  philosophical  manifestations  would  be 
too  modest  a  task  ;  for  though  we  may  somewhere  find 
a  logical  concatenation  of  facts  along  with  the  facts 
themselves,  still  it  is  in  doctrines  and  systems  that 
philosophy  finds  its  realisation. 

On  the  other  hand,  he  would  be  a  bold  man  who 
would  affirm  that  some  particular  conception  has  a  future 
before  it,  whereas  some  other  has  had  its  day.  In 
Voltaire's  time,  metaphysics  was  an  utterly  exploded 
doctrine.  Now,  that  was  the  very  period  when  German 
philosophy  was  beginning  to  become  known. 

And  what  an  ambition,  to  find  the  historical  and 
unconscious  origins,  the  mechanical  genesis  of  a  thinker's 
ideas  !  Which  of  us,  even  the  most  wide-awake  and 
skilled  in  analysing  mental  states,  could  correctly  explain 
the  origin  of  his  opinions  and  doctrines  ?  Amongst  the 
many  influences  under  which  our  increasingly  complex 
life  is  continually  bringing  us,  how  can  we  set  apart 
those  that  have  been  deep  and  lasting,  or  state  exactly 
in  what  direction  they  have  been  exercised  ?  Besides, 
why  do  we  so  strongly  insist  that  our  ideas  spring  only 
from  external  influences  and  that  we  ourselves  have 
nothing  to  do  with  their  production  ? 

Apart  from  these  various  conceptions  of  the  history 
of  philosophy — conceptions  in  turn  excessively  timid  and 


4  STUDIES  IN  PHILOSOPHY 

boldly  venturesome — there  is  one  of  a  less  striking 
nature,  from  the  fact  that  it  does  not  seem  to  be  so 
scientific,  though  perhaps  it  is  more  in  accordance  with 
the  nature  of  the  subject  under  investigation.  It  is 
also  the  one,  unless  I  am  mistaken,  generally  applied  by 
writers  whose  distinctive  object  is  to  take  up  the  history 
of  philosophy,  without  troubling  about  anything  else. 

It  consists  in  regarding  as  the  subject  of  investigation 
from  the  very  outset,  what  to  us  are  immediate  data, 
to  wit,  any  particular  doctrine  that  is  one  in  its  greater 
or  less  complexity,  any  collection  of  ideas  set  forth  by 
the  philosopher  as  forming  a  whole.  Where  this  con- 
dition is  not  fulfilled,  we  may  indeed  be  dealing  with  a 
shrewd  moralist  or  a  profound,  original  thinker,  but 
certainly  we  are  not  dealing  with  a  philosopher.  The 
problem  to  be  solved  is  that  of  finding  out  what  logical 
connection  the  philosopher  has  really  set  up  between  his 
ideas,  which  he  has  taken  as  principles,  and  in  what 
order  and  fashion  he  makes  the  rest  depend  on  the  main 
ideas.  A  philosopher  is  a  man  who  sets  up  men's 
knowledge  over  against  their  beliefs,  and  tries  to  find 
their  relations  to  one  another.  We  want  to  know  how 
a  Plato  or  a  Leibnitz  conceived  of  these  relations.  And 
since  the  philosopher  is  not  a  seer  to  whom  truth  is 
revealed  in  a  flash,  but  rather  a  patient  seeker  who  re- 
flects and  criticises,  doubts  and  hesitates,  and  listens  to 
reason  alone,  we  want  to  know  the  methods,  observations 
and  reasonings  by  which  our  author  has  reached  his 
conclusions.  We  are  not  now  dealing  with  the  uncon- 
scious, mechanical  work  of  his  brain,  but  with  his 
conscious,  determined  efforts  to  overcome  the  limits  of 
his  individuality,  to  think  in  an  all-embracing  manner, 
and  to  bring  the  truth  to  light. 

If  such  be  the  case,  it  is  neither  philosophy  in  general, 


THE  HISTORY  OF  PHILOSOPHY          5 

throughout  its  development,  nor  the  psychological 
evolution  of  each  philosopher  in  particular,  that  is  the 
immediate  object  of  the  history  of  philosophy  :  it  is  the 
doctrines  that  have  been  thought  out  by  philosophers. 
To  know  and  understand  these  doctrines  well,  to  explain 
them — to  the  extent  of  one's  capacity — as  the  author 
himself  would  do,  to  set  them  forth  in  the  spirit  of  that 
author,  and  to  some  extent,  in  his  style  :  this  is  the  one 
essential  task,  to  which  all  the  rest  must  be  subordinated. 

It  is,  indeed,  useful  to  consider  the  man,  and  not 
the  work  only,  but  it  is  so  because,  in  most  cases,  the 
man  helps  us  to  understand  the  work.  Cartesianism  is 
indebted  for  more  than  one  of  its  characteristics  to 
Descartes,  the  man.  And  yet,  what  a  mistake  to  insist 
on  regarding  Cartesianism  as  nothing  more  than  the 
history  of  an  individual  mind  ! 

It  is  likewise  an  interesting  question  to  ask  oneself : 
What  becomes  of  philosophy  fer  se  throughout  the 
succession  of  systems?  Does  it  advance  or  remain 
stationary  ?  This  general  study  of  philosophy,  however, 
cannot  replace  that  of  the  doctrines  considered  in  them- 
selves from  each  author's  point  of  view  :  rather  it  pre- 
supposes it. 

Let  it  not,  then,  be  said  that  some  particular  portion 
of  a  philosopher's  doctrines  may  be  neglected,  under  the 
plea  that  it  is  to  be  found  in  the  writings  of  some 
earlier  philosopher.  That  is  an  insufficient  reason.  A 
great  mind  does  not  seek  after  novelty  or  originality  ; 
it  seeks  after  truth.  Why  should  it  refuse  any  portion 
thereof  under  the  plea  that  it  was  discovered  by  some- 
one else  ?  In  the  classic  ages  of  literature,  authors  did 
not  feel  themselves  bound  to  create,  ex  nihilo,  after  the 
fashion  of  God.  A  Corneille  and  a  Moliere  make  lavish 
use  of  the  works  of  their  predecessors.  No  one  finds 


6  STUDIES  IN  PHILOSOPHY 

fault  with  their  lack  of  originality  when  they  take 
advantage  of  this  material  in  writing  fine  and  noble 
works.  With  still  more  reason,  an  Aristotle,  a  Leibnitz 
and  a  Kant  carefully  retain  whatever  those  who  have 
gone  before  have  found  to  be  of  advantage.  In  reality, 
they  make  it  their  own  by  the  way  in  which  they  use  it. 
"  When  two  men  are  playing  tennis,"  said  Pascal,  "  both 
play  with  the  same  ball,  but  one  of  them  places  it  better 
than  the  other."  It  often  happens  that  the  most  common- 
place idea  assumes  a  new  aspect  by  reason  of  the  new 
relations  in  which  it  finds  itself. 

On  the  other  hand,  some  idea  destined  at  a  later 
period  to  prove  an  important  and  fruitful  one,  may 
have  played  only  a  secondary  or  eclipsed  part  in  the 
system  in  which  it  first  appeared.  Though  picking  it  up 
as  a  chance  find,  so  to  speak,  or  regarding  it  as  an  interest- 
ing presentiment,  we  must  be  careful  not  to  place  it  in 
the  foreground  under  pretence  of  serving  the  author  by 
giving  him  a  more  modern  aspect.  It  is  not  Descartes 
as  one  would  imagine  he  would  be  at  the  present  time, 
but  the  Descartes  of  1 644,  referring  every  problem  to  the 
one  of  certainty,  whom  it  is  our  object  to  make  known. 

The  task  in  hand  determines  exactly  the  means  it 
behoves  us  to  put  into  operation.  In  the  ulterior 
developments  a  system  may  have  gone  through,  the 
doctrines  to  which  it  has  given  birth,  the  appreciations 
and  interpretations  of  contemporaries  and  successors,  or 
even  the  historical  and  biographical  information  regard- 
ing the  author's  person  and  works,  we  ought  not  to 
look  for  anything  else  than  sign-posts  of  the  problems 
we  must  set  before  ourselves,  or  material  data  which 
determine  the  ground  on  which  to  work.  The  spring 
and  origin  of  the  history  of  philosophy  can  be  found 
only  in  the  monuments  left  by  philosophers  themselves. 


THE  HISTORY  OF  PHILOSOPHY          7 

Each  philosophical  work  requires  to  be  considered 
both  as  a  whole  and  in  detail.  The  work  of  the  mind 
is  one  continual  oscillation  from  the  whole  to  the  parts 
and  from  the  parts  to  the  whole.  Such  is  the  method 
applied  in  the  understanding  of  a  drama,  a  poem,  or  a 
work  of  art.  This  alternate  movement  of  induction  and 
deduction  is  the  origin  of  the  sciences.  In  the  same 
way,  if  we  explain  the  author  by  himself,  his  general 
ideas  by  his  particular  doctrines  and  his  particular 
doctrines  by  his  general  ideas,  the  probabilities  are  that 
we  shall  thoroughly  grasp  his  meaning  and  enter  into 
his  thought. 

It  is  not  enough  to  discover  curious — even  un- 
published— texts.  Which  of  us  can  enter  completely 
into  all  an  author  says  ?  What  likelihood  is  there  that 
a  letter  written  to  some  correspondent  or  other,  how- 
ever ill  fitted  he  be  to  understand  the  philosopher, 
should  prove  of  greater  importance  than  treatises  that 
have  been  slowly  matured  and  are  destined  for  posterity  ? 
The  historian,  whose  aim  it  is  not  to  make  a  collection  of 
anecdotes,  but  to  form  a  correct  appreciation  of  a  great 
man's  work,  will  be  less  anxious  to  marshal  and  array 
an  imposing  number  of  disconnected  texts  than  to 
become  increasingly  imbued  with  the  author's  thought, 
by  reading  the  whole  of  his  works  not  once,  but  many 
times.  His  aim  will  be  to  see  things  from  the  author's 
point  of  view,  following  him  along  the  winding  by- 
paths of  meditation,  sharing  in  his  emotions  as  a  philo- 
sopher, and,  along  with  him,  enjoying  that  harmony 
wherein  his  intelligence  has  found  repose. 

Systems  of  philosophy  are  living  thoughts.  It  is  by 
seeking  in  the  written  word  for  the  means  of  reviving 
these  thoughts  within  ourselves  that  we  may  hope  to 
hear  them  deliver  their  message. 


SOCRATES, 
THE  FOUNDER  OF  MORAL  SCIENCE l 

"  Les  memes  pens6es  poussent  quelquefois  tout  autrement  dans 
un  autre  que  dans  leur  auteur." — PASCAL. 


AFTER  the  keen  rivalry  that  has  existed  amongst  those 
most  capable  of  dissipating  the  clouds  and  mists  that 
hang  about  the  figure  of  Socrates,  inquiring  men  of 
letters,  shrewd  moralists,  philosophers  of  penetrating 
intellect,  learned  historians,  and  even  doctors,  whose 
object  it  has  been  to  collect  and  interpret  such  docu- 
ments as  are  calculated  to  make  him  known  to  us,  is 
there  anything  left  to  say  about  him  ?  Is  not  the 
writer  on  such  a  theme  compelled  to  repeat  mere 
commonplaces  if  he  is  determined  to  say  only  what  is 
true,  to  give  expression  to  paradoxes  if  he  claims  to 
have  anything  new  to  advance  ? 

In  this  connection  it  would  seem  as  though  a  dis- 
tinction must  be  made.     Doubtless,  all  possible  light 

1  The  present  essay  deals  less  with  the  feelings  and  the  soul  of  Socrates 
than  with  his  philosophy  and  his  work.  In  one  aspect  Socrates,  as  a  man, 
is  an  enthusiast  in  the  literal  meaning  of  the  word,  almost  a  mystic.  He 
is  Apollo's  messenger,  and  feels  within  himself  the  divine  afflatus.  His 
unique  originality  consisted  in  introducing  religious  zeal  into  the  preach- 
ing of  rational  morals.  Here  we  confine  ourselves  to  the  consideration  of 
the  doctrine  which  Socrates  taught  his  disciples  and  bequeathed  to  man- 
kind. 

8 


SOCRATES  9 

has  been  shed  on  most  of  the  details  of  the  life  and 
teaching  of  Socrates,  but  it  is  not  so  certain  that  this 
could  be  said  concerning  the  ensemble  of  the  man  and 
his  doctrine.  The  reader  is  astonished  when  he  com- 
pares with  one  another  those  works  of  our  contemporaries 
that  deal  with  Socrates.  If  we  would  know  of  his 
life,  the  causes  of  his  condemnation,  the  meaning  of 
maieutics,  the  doctrines  of  virtue,  or  some  other 
portion  of  Socratic  philosophy,  each  of  these  authors 
gives  almost  identical  answers.  But  if  we  ask  what 
Socrates  was  in  himself,  the  basis  of  his  character  and 
the  root  idea  of  his  teaching  :  regarding  this  question 
— in  which  all  the  rest  culminate — opinions  are 
contradictory. 

Thus,  according  to  Zeller,1  ancient  physics  having 
finally  disappeared  beneath  the  action  of  sophistic, 
Socrates  regenerated  philosophy  by  founding  it  upon  a 
new  principle  :  the  general  or  concept,  regarded  as  the 
object  of  science.  The  work  of  Socrates,  then,  was 
the  invention  of  a  principle  of  theoretical  logic. 

Grote,  in  a  series  of  life-like  sketches,  presents 
Socrates  as  a  religious  missionary,  appointed  by  the 
oracle  of  Delphi  for  putting  the  would-be  wise  on  the 
rack  and  inducing  them  to  confess  their  ignorance. 
Socrates  is  the  god  of  debate,  "  an  elenchtic  or  cross- 
examining  god."  2  His  work,  religious  in  its  inspira- 
tion, is  a  living  dialectic  in  itself. 

Fouillee  regards  Socrates  as  a  speculative  philosopher, 
who  substitutes  final  causes  for  physical  ones  in  the 
explanation  of  all  phenomena,  both  physical  and  moral. 
He  is  the  creator  of  spiritual  metaphysics. 

Leveque    considers    that    Socrates    endeavoured    to 

1  Die  Philosophie  der  Griechen,  3rd  edit.  vol.  ii.  p.  93-94. 
2  History  of  Greece,  vol.  viii.  p.  566. 


io  STUDIES  IN  PHILOSOPHY 

bring  about  the  moral  and  political  reform  of  Athens, 
and,  with  this  end  in  view,  established  morals  as  a 
science  independent  of  the  physical  sciences. 

Janet's  brief  though  important  sketch  in  the  Dic- 
tionnaire  philosophique  shows  us  Socrates  as  a  philo- 
sopher above  all  else  ;  he  mentions  two  main  char- 
acteristics of  his  :  the  moral  sentiment,  which  dominates 
his  personality  and  appears  throughout  his  doctrine ; 
and  maieutics,  from  which  the  Platonic  dialectic  was 
to  originate. 

In  a  short  work,  published  in  1881,  Gustave 
d'Eichthal  considers  the  outstanding  feature  of  the 
Socratic  doctrine  to  be  religious  instruction.  Socrates, 
he  says,  with  a  view  to  checking  the  evils  he  saw 
ravaging  his  country,  wished  to  give  his  fellow-citizens 
what,  to  him,  was  the  principle  of  all  virtue  and  the 
first  condition  of  every  reform,  namely,  religious  faith, 
especially  faith  in  divine  Providence. 

Finally,  Franck,  in  an  article  that  appeared  in  the 
Journal  des  savants  on  d'Eichthal's  book,  likewise 
admits  that  Socrates  was  not  only  a  reasoner  and  a 
philosopher,  but  more  than  all  else  a  profoundly 
religious  soul,  in  the  real  meaning  of  the  word,  a  soul 
in  whom  faith  in  God,  admiration  of  his  works,  the 
certainty  of  his  kingdom  throughout  nature  and  of  his 
providence  over  men,  were  tinged  with  a  certain  degree 
of  mysticism. 

All  these  interpretations,  moreover,  are  based  on 
texts  of  the  greatest  importance.  Thus,  confining 
ourselves  to  the  three  of  our  contemporaries  who  have 
written  most  about  Socrates,  Zeller,  in  support  of  his 
position,  quotes  that  clear,  precise  passage  in  Aristotle * 
where  it  is  mentioned  that  Socrates  seeks  the  ri 

1  Met.  xiii.  4.  1878  b  23  sqq. 


SOCRATES  ii 

the  general  essence,  though  without  regarding  this 
essence  as  existing  apart,  as  Plato  did.  Grote  draws 
his  conclusions  from  the  Apology?  which,  indeed, 
mainly  shows  us  Socrates  as  having  received  from  the 
gods  the  mission  of  convincing  men  of  their  ignorance. 
And  lastly,  the  statement  of  Fouillee  2  appears  inspired 
by  those  luminous  passages  of  the  Phaedo?  in  which  we 
find  Socrates  reproaching  Anaxagoras  for  omitting  to 
take  into  account,  in  his  explanation  of  the  details  of 
the  world,  that  ordaining  and  regulating  intelligence  he 
had  so  wisely  proclaimed  to  be  the  universal  cause  ; 
considering,  for  his  part,  that  any  purely  mechanical 
explanation  was  superficial  ;  and  satisfied  only  with  such 
explanations  as,  in  the  ultimate  analysis,  were  given  by 
final  causes.4 

But  why  is  it  that  each  of  these  authors  has  taken 
up  some  particular  text  in  preference  to  others  ?  In 
all  likelihood,  personal  preoccupation  or  different  mental 
habits  may  give  a  partial  explanation.  An  old  Hegelian 
like  Zeller,  whose  object  above  all  is  to  find  out  the 
place  occupied  by  men  and  doctrines  in  the  general 
development  of  the  human  mind,  was  bound  to  take, 
as  his  main  guide,  Aristotle,  who  emphasises  in  his 
predecessors  just  those  ideas  that  have  paved  the  way 
for  his  own.  Grote,  the  historian,  who  would  point 
out  the  part  played  by  famous  men  throughout  the 
entire  social  and  political  life  of  their  times,  was  bound 
to  rely  mainly  on  the  Apology^  a  life-like  picture,  it 
would  seem,  of  Socrates  as  he  appeared  to  his  fellow- 
citizens.  Lastly,  Fouillee,  the  eloquent  and  profound 
interpreter  of  the  theory  of  Ideas,  was  naturally  disposed 

1  Grote,  History  of  Greece,  viii.  565. 
2  La  Philosophie  de  Platan,  vol.  i.  p.  17  sqq. 

3  Ch.  xlv.  sqq. 
*  Phaedo,  ch.  xlvi.  p.  97  B. 


12  STUDIES  IN  PHILOSOPHY 

to  regard  Socrates  as  the  precursor  of  Plato,  and  to 
find  in  his  doctrines  the  germ  of  Platonic  metaphysics. 
It  is  not  surprising  that  he  should  take  as  his  starting- 
point  those  passages  in  which  Plato  himself  connects 
his  theory  of  Ideas  with  the  speculations  of  his  master. 

In  these  investigations  into  the  real  character  of 
Socrates,  Zeller  appears  to  have  adopted  the  standpoint 
of  absolute  mind,  Grote  that  of  a  cultured  Athenian  of 
the  fifth  century,  and  Fouillee  that  of  Plato.  What 
would  be  the  result  were  we  to  adopt  the  standpoint 
of  Socrates  himself,  asking  ourselves  what  Socrates 
must  have  been,  not  in  the  eyes  of  others  but  in  his 
own  ?  The  apostle  of  the  yv&Oi  o-avrov  must  have 
been  acquainted  with  himself.  We  should  regard  our- 
selves as  having  sufficient  knowledge  of  him  were  we 
acquainted  with  him  to  the  same  extent. 

But  then,  how  can  we  enter  into  the  soul  of  Socrates, 
since  he  left  nothing  in  writing?  Is  it  not  this  very 
difficulty  of  adopting  his  point  of  view  which  induces 
historians  to  seek  one  from  without  ? 

Perhaps  the  difficulty  is  partly  artificial.  It  showed 
itself  most  prominently  when  Schleiermacher  advanced 
the  principle  that,  for  an  exposition  of  the  Socratic 
doctrine  to  be  a  faithful  one,  it  must  above  all  else 
explain  how  Plato  came  to  regard  Socrates  as  the 
promoter  of  his  philosophic  activity.  From  this  stand- 
point a  comparison  was  made  between  the  Socrates  of 
Xenophon  and  the  Socrates  of  Plato  and  Aristotle,  and 
the  two  were  found  to  differ  widely.  Naturally,  the 
followers  of  Schleiermacher  adopted  the  views  of  Plato 
and  Aristotle,  and  so  the  authority  of  the  only  one  of 
our  witnesses  who  was  a  historian  by  profession,  and 
who  had  made  it  his  business  to  tell  us  what  Socrates 
had  really  been  in  his  own  person,  was  discredited. 


SOCRATES  13 

A  change,  however,  has  come  about  since  then. 
Whilst  the  champions  of  Xenophon  and  Plato  were 
wrangling  over  Schleiermacher's  theory,  a  less  biassed 
criticism  compared  the  testimonies  of  Xenophon,  Plato 
and  Aristotle  with  one  another.  Now,  these  testimonies 
have  been  found  to  agree  as  regards  the  main  issue.1 
Henceforth,  to  an  impartial  critic,  the  authority  of 
Xenophon  was  restored.  The  charge  might  still  be 
brought  against  him  that  he  set  forth  the  person  and 
teachings  of  his  master  more  or  less  incompletely,  though 
not  that  he  presented  them  in  a  wrong  light.  If  such  be 
the  case,  the  historian  of  the  present  day  has  the  right, 
not  only  to  invoke  the  testimony  of  Xenophon,  along 
with  those  of  Plato  and  Aristotle,  but  even  to  assign 
the  greatest  importance  to  this  testimony,  for  Xenophon 
is  the  only  one  of  the  three  who  does  nothing  more 
than  repeat  what  he  personally  knew.  True,  the 
immediate  object  of  his  work  would  appear  to  have 
been  to  refute  the  harangue  of  Polycrates,  the  rhetor, 
about  the  year  393  B.C.  ;  none  the  less,  Xenophon 
must  have  brought  to  his  task  those  qualities  of  fidelity 
and  impartiality  that  distinguish  his  strictly  historical 
narratives. 

Of  course,  we  must  not  repeat  the  mistake  made  by 
the  historians  of  old,  who,  reading  Xenophon  in  a  very 
superficial  manner,  saw  depicted  only  the  account  of  a 
simple-minded  moralist  ;  we  must  allow  Plato  and 
Aristotle  to  breathe  life  into  and  complete  the  picture 
sketched  by  Xenophon.  Still,  it  would  be  wise  to  use 
the  contributions  of  the  two  former  only  as  a  scholar 
uses  a  hypothesis,  that  is,  in  stating  or  asking  questions, 
not  in  solving  or  answering  them.  To  analyse  the 
data  of  Xenophon,  interpreting  and  developing  them 

1  Such  is  the  opinion  of  Zeller,  Grote  and  Fouiltee. 


i4  STUDIES  IN  PHILOSOPHY 

according  to  a  scientific  induction  whose  leading  ideas 
are  to  be  supplied  by  Plato  and  Aristotle  :  such  seems 
to  be  the  method  we  must  pursue,  if  we  would  know 
Socrates  in  a  really  historical  fashion. 

Along  with  Xenophon's  Memorabilia  we  must  con- 
sider Plato's  Apology,  which  most  critics1  look  upon  as 
trustworthy  with  regard  to  facts  ;  also  certain  portions, 
difficult  to  define,  it  must  be  confessed,  of  the  Crito, 
Phaedo,  Laches  and  The  Banquet. 

Now,  what  is  the  root  thought  of  Socrates,  regarded, 
as  far  as  possible,  from  his  own  point  of  view  ? 

II 

The  first  result  we  obtain,  if  we  take  the  Memorabilia 
as  the  main  source  of  the  history  of  Socratic  thought,  is 
a  confession  of  ignorance  as  to  what  happened  previous 
to  the  last  ten  years  of  the  philosopher's  career.  The 
temptation  is  almost  irresistible  to  seek  in  other  works 
for  some  means  of  going  back  to  an  earlier  year  in 
Socrates'  life  than  the  Memorabilia  allow.  For  instance, 
Fouillee  believes  he  has  found,  in  the  famous  passage  of 
the  Phaedo  on  the  early  philosophical  reflections  of 
Socrates,2  and  the  coincidence  of  this  text  with  the 
Clouds  of  Aristophanes,  the  proof  that  Socrates,  before 
devoting  himself  to  moral  research,  passed  through  a 
previous  stage,  in  which  he  was  engaged  in  speculations 
on  nature.  Disappointed  in  this  direction,  he  would 
appear  to  have  fallen  back  on  morals  for  a  solution 
of  the  very  problem  of  ancient  Greek  philosophy  :  that 
of  the  explanation  of  the  universe.  Besides  the  fact, 
however,  that  the  Memorabilia  contain  no  indication 

1  Schleiermacher,  Zeller,  tlberweg  and  Grote. 
2  Ch.  xlv.  sqq. 


SOCRATES  15 

whatever  of  such  a  starting-point,  the  statement  of  the 
Socrates  of  the  Phaedo  contradicts  the  formal  declara- 
tions of  the  Socrates  of  the  Apology,  where  it  is  affirmed 
that  he  never  studied  physics.1  The  objection  may  be 
urged  that  the  character  of  Socrates  in  the  Clouds  must 
rest  upon  some  historical  basis.  But  it  is  precisely 
when  speaking  of  the  Clouds  that  Socrates  makes  this 
solemn  declaration  in  the  Apology.  True,  the  question 
is  decided  by  dismissing  the  Apology ,  under  the  pretext 
that  it  is  a  speech,  and  alleging  that  the  text  of  the 
Phaedo  itself  gives  one  the  impression  of  historical 
reality.  Such  preference,  however,  is  unjustified.  As 
it  is  the  object  of  the  text  of  the  Phaedo  to  show  us  the 
origin  of  the  theory  of  ideas,  which  theory,  moreover, 
is  likewise  attributed  to  Socrates,  it  is  best  to  attribute  to 
Plato  himself  the  reflections  with  which  this  exposition 
commences.  The  Apology  is  certainly  possessed  of 
historical  value,  as  is  proved,  along  with  other  details, 
by  the  strange  prediction  Socrates  made  to  the  judges,2 
that,  when  he  was  dead,  the  Athenians  would  find  a  far 
greater  number  of  censors  (eXey^oi/re?)  rising  up  against 
them,  and  these  would  be  all  the  more  unpleasant 
because  they  would  be  younger.  .  This  prediction, 
which  does  not  appear  to  have  come  to  pass,  would 
certainly  have  been  omitted  in  an  apology  invented  by 
Plato  himself.  But  if  Socrates  indeed  challenged  his 
listeners  to  prove  that  he  ever  even  mentioned  physics,3 
how  could  we  affirm  the  contrary  ?  Shall  we  set  the 
fiction  of  a  comic  poet  above  the  testimony  of  Socrates 
himself  ? 

Consequently,  we  will  abandon  the  attempt  to  dis- 
cover what  ideas  Socrates  held  in  youth  and  even  in 

1  Ch.  iii.  p.  19  c  D.  2  Ch.  xxx.  p.  39  c  D.     Cf.  Grote. 

3  Ch.  iii.  p.  19  D. 


1 6  STUDIES  IN  PHILOSOPHY 

mature  age.  Besides,  we  have  ground  to  suppose  they 
were  in  conformity  with  those  he  held  at  the  end  of  his 
life,  for  Socrates,  in  the  Apology^  tells  his  listeners  that 
the  reason  they  are  prejudiced  against  him,  and  look 
upon  him  as  a  physicist  and  a  sophist,  is  that  ever 
since  they  were  children  they  have  been  deceived  by 
his  enemies  regarding  himself.1  At  all  events,  to 
claim  to  throw  light  on  the  Socrates  of  his  latter  years 
by  the  Socrates  of  the  Clouds  period  is  trying  to 
explain  the  known  by  the  unknown. 

The  starting-point  of  the  established  doctrine  of 
Socrates  we  shall  find  in  his  critical  reflections  on  the 
two  disciplines  which  then  occupied  men's  minds  : 
physics  and  sophistic. 

Socrates  never  applied  himself  to  physics.  The 
testimony  of  Plato  2  and  Aristotle  3  is  a  proof  of  this, 
without  mentioning  that  of  Xenophon.  There  can  be 
no  doubt,  however,  that  he  had  studied  the  subject, 
though  it  was  principally  as  a  philosopher  that  he 
was  interested  in  it.  It  was  not  the  details  of  the 
science,  the  particular  theories  which  in  all  probability 
were  the  main  object  of  research  on  the  part  of  the 
ancient  physiologists,  that  he  cared  about,  but  rather 
those  general  principles  that  controlled  all  the  rest,  the 
mechanical  or  dynamical  conceptions  of  nature  which 
led  philosophers  to  explain  everything  without  having 
recourse  to  supernatural  powers.  Is  being  one  or 
multiple  ;  is  it  in  motion  or  at  rest ;  is  it  subject  to  a 
state  of  becoming  and  destruction,  or  is  it  exempt  from 
generation  and  corruption  ?  Such  were  the  philosophical 
questions  that  physiologists  asked  themselves.4 

Socrates  wasted  no  time  in  examining  one  by  one 

1  Ch.  ii.  p.  18  c.  3  Met.  i.  6.  987  b  i. 

2  Apol.  ch.  iii.  p.  19  D.  4  Xenophon,  Mem.  i.  i.  14. 


SOCRATES  17 

the  different  doctrines  to  which  the  idea  of  natural 
physics  had  given  birth.  He  condemned  them  en  bloc, 
as  being  useless,  barren  and  sacrilegious. 

Physics  was  useless,  for  physicists  were  unable  to 
agree  on  a  single  point.  Some  maintained  that  being  is 
one,  the  rest  that  it  is  infinitely  multiple  ;  some  that 
everything  is  in  motion,  the  rest  that  everything  is  for 
ever  motionless,  and  so  on.1  Now,  contradiction  is  a 
sign  of  ignorance. 

And  it  was  barren  as  well.  Do  those  who  trouble 
about  such  matters,  said  Socrates,  imagine  that  when 
they  have  discovered  the  law  of  necessity  according  to 
which  everything  is  produced,  they  will  be  able  to  make 
the  winds,  waters  and  seasons  at  their  own  pleasure  ? 2 

And  these  two  features  were  themselves  the  result 
of  a  radical  vice  :  to  wit,  the  sacrilegious  nature  of 
the  task.  All  that  is,  said  Socrates,  may  be  divided 
into  two  categories,3  human  things  (ra  avOpaTreia),  such 
as  pious  and  impious,  beautiful  and  ugly,  just  and 
unjust,  matters  dealing  with  civic  life  and  authority,4 
and  divine  things  (&u/*owa),  such  as  the  formation  of 
the  world,5  or  even  the  distant  and  final  consequences 
of  our  actions.6  Now,  the  gods  have  given  us  power 
to  know  the  former  by  reasoning  ;  the  latter  they  have 
reserved  for  themselves.7  Physicists,  when  speculating 
on  things  divine  and  neglecting  the  human,  invert  the 
order  set  up  by  the  gods  themselves  :  they  disdain 
knowledge  which  the  gods  have  placed  within  our 
reach,  and  try  to  acquire  that  knowledge  the  gods  have 
reserved  for  themselves. 

It  is   noteworthy  that  Pascal  makes   a  similar  dis- 

1  Xenophon,  Mem.  iv.  2.  4  Ibid.  i.  i.  16. 

2  Ibid.  i.  i.  z5.  6  Ibid.  i.  i.  n. 

3  Ibid.  i.  i.  12.  °  Ibid.  i.  i.  8. 

7  Ibid.  i.  i.  7-8. 


i8 

tinction.  He  also1  divides  things  into  human  and 
divine,  and  accuses  men  of  having  perverted  the  order 
established  by  God  when  they  use  profane  things  as  they 
ought  to  use  sacred  ones,  and  vice  versa,  that  is  to  say, 
when  they  consider  profane  things  with  the  heart  and 
divine  ones  with  the  mind.  To  Pascal,  however,  it  is 
physical  things  that  are  profane  and  moral  ones  that  are 
divine. 

Both  this  resemblance  and  this  difference  enable  us 
all  the  better  to  understand  the  thought  of  Socrates. 
The  same  religious  spirit,  both  in  Socrates  and  Pascal, 
sets  a  limit  upon  human  reason.  To  the  Hellene, 
however,  man  himself  is  his  own  master  ;  it  is  nature, 
with  its  mysteries  and  remoteness,  that  is  divine.  To 
the  modern  man  and  the  Christian  it  is  the  infinitude  of 
the  interior  life  that  is  divine,  and  nature,  brute,  passive 
matter,  that  is  the  object  set  before  human  activity. 

The  original  cause  of  Socrates'  condemnation  of 
ancient  physics  may  be  found  in  the  stock  or  fund  of 
ideas  peculiar  to  his  nation.  Greece  could  not  wholly 
adapt  herself  to  those  speculations  on  the  principles  of 
things  into  which  physiologists  had  plunged.  Doubt- 
less the  power  of  reasoning,  the  ingenious  subtilty  and 
wonderful  sense  of  harmony  displayed  by  these  pro- 
found investigators,  were  a  good  thing,  but  the  imme- 
diate application  of  these  mental  qualities  to  material 
objects  most  foreign  to  mankind  was  opposed  to  the 
genius  of  a  race  that  was  essentially  political,  and 
mightily  enamoured  of  fine  speeches  and  noble  deeds. 
Besides,  how  could  one  reconcile  a  philosophy  which 
undertook  to  explain  physical  phenomena  by  perfectly 
natural  causes  with  a  religion  which  everywhere  intro- 
duced the  immediate  action  of  the  gods?  Certainly 

1  De  I' Esprit  giom.  2nd  frag. 


SOCRATES  19 

they  were  Greeks  who  had  planned  these  beautiful 
systems  in  which  nature  was  subject  to  the  laws  of 
thought ;  still,  they  were  citizens  of  the  colonies,  and  had 
dealings  with  the  Egyptians,  Phoenicians  and  Babylonians. 
They  had  created  the  form  :  the  East  had  supplied 
them  with  the  matter.  To  detach  human  affairs  from 
the  totality  of  things,  to  make  them  the  proper  domain 
of  man's  activity  and  intelligence,  and  at  the  same  time  to 
restore  physical  phenomena  once  more  to  the  gods,  was 
to  place  oneself  again  in  the  position  of  the  Hellene  : 
more  especially  of  the  Athenian.  This  was  quite 
natural  for  the  philosopher  who  never  left  Athens  except 
to  fight  in  the  ranks  of  his  fellow-citizens. 

Socrates' judgment  on  physics,  therefore,  is  no  for- 
tuitous accidental  fact ;  it  is  not  the  outcome  of  a  posi- 
tive, a  prosaically  utilitarian  mind.  It  is  not  even 
altogether  that  depreciation  of  the  past,  habitual  to 
innovators,  that  antagonism  to  a  rival  idea  :  the  condi- 
tion of  the  realisation  and  development  of  the  new  idea 
which  claims  the  right  to  exist.  Socrates'  objections  to 
physics  are  the  philosophic  expression  of  that  antipathy 
of  a  religious,  artistic  people  to  a  mechanical  explana- 
tion of  things,  whereof  Aristophanes  set  himself  up  as 
the  interpreter  in  the  Clouds.  The  real  Socrates  flouts 
the  Socrates  of  Aristophanes,  as  do  the  people.  The 
only  difference  is  that  he  knows  better  why  he  does  so. 

But  this  very  discernment  of  his  prevents  him  from 
altogether  condemning  the  work  of  the  physicists. 
Though  declaring  it  useless,  barren  and  sacrilegious,  he 
yet  discovers  in  it  a  principle  which  he  is  jealously 
anxious  to  adopt.  This  principle  is  the  form  and 
mould,  so  to  speak,  of  Hellenic  thought  into  which  the 
physiologists  cast  the  matter  they  borrowed  from  the 
East :  it  is  the  consciousness,  henceforth  acquired  by 


20  STUDIES  IN  PHILOSOPHY 

the  human  mind,  of  its  need  of  unity  and  harmony  ; 
the  notion  of  an  impersonal  truth,  distinct  from  opinion 
and  fancy  ;  the  abstract  idea  of  science.  When 
Socrates  asks  the  physiologists l  if  the  reason  they 
undertake  to  speculate  on  divine  things  lies  in  the  fact 
that  they  think  they  know  enough  of  human  things,  he 
evidently  retains  of  ancient  physics  the  general  idea  of 
science  as  being  a  special,  a  superior  mode  of  know- 
ledge, whilst  leaving  on  one  side  the  object  to  which  this 
idea  has  hitherto  been  applied. 

And  so  the  general  idea  of  science  does  not  spring 
forth  all  at  once  in  the  mind  of  Socrates,  with  the  intui- 
tion of  genius,  as  one  might  imagine  from  Schleier- 
macher's  profound  though  abstruse  dissertation.  Nor 
is  it  the  reaction  of  subjectivism  against  objectivism,  a 
reaction  which  was  evidently  determined  by  the  excesses 
of  objectivism  itself,  in  accordance  with  the  general  law 
of  the  development  of  the  human  mind,  as  appears  to 
be  admitted  by  the  former  Hegelian,  Zeller.  This  idea 
of  science  is  nothing  else  than  the  proper  share  of  the 
Hellenic  genius  in  the  formation  of  ancient  physics. 
Socrates'  work  lies  in  freeing  it  from  the  foreign 
elements  with  which  it  was  confused,  owing  to  a  subtle 
distinction  between  matter  and  form  which  the  different 
opponents  of  the  physiologists  had  been  unable  to  draw. 
In  this  he  was  doubtless  aided  by  his  power  of  inven- 
tion, as  well  as  by  his  singularly  Hellenic  turn  of 
mind.  In  him  the  Greek  genius  recognised  its  own 
good  fortune  through  the  scientific  form  that  the 
physiologists  had  given  to  the  practical  knowledge  or 
astronomical  speculations  of  the  Orientals. 

Though  Socrates  concerned  himself  with  physics,  he 
paid  even  more  attention  to  sophistic.  Here  he  dis- 

1  Mem.  i.  i.  12. 


SOCRATES  21 

tinguished  two  things :  the  end  and  the  means.  In  his 
opinion,  the  end  or  object  of  sophistic  was  to  make 
men  capable  of  speaking  and  acting  well,  of  managing 
efficiently  the  business  of  city  and  home,  in  a  word,  of 
being  useful  to  others  as  well  as  to  themselves.1  The 
means  consisted  solely  of  exercise  and  routine,  the 
immediate  practice  of  that  action  the  capacity  for 
which  it  is  one's  object  to  acquire,  and  so  the  Sophist, 
in  the  mind  of  Socrates,  is  the  man  who  identifies  the 
means  with  the  end,  who  considers,  for  instance,  that, 
in  order  to  learn  to  speak  well,  all  that  is  necessary  is 
to  hear  others  speak  and  to  speak  oneself,  without 
taking  the  trouble  to  study  theoretically  the  conditions 
of  eloquence.  Practice  is  sufficient  in  itself.  Talent  is 
like  some  physical  aptitude  which  men  acquire  by  being 
shaped  and  drilled  to  acquire  it. 

Socrates  approved  of  the  object  of  this  discipline, 
though  he  condemned  the  method  employed. 

It  was  not  ironically  that  he  called  the  sophistic  art 
the  finest  and  greatest  of  them  all,  a  truly  royal  art.2 
If  we  consider  nothing  but  the  end  set  up  for  human 
activity,  we  find  that  Socrates  is  not  only  in  agreement 
with  the  Sophists,  he  is  himself  one  of  them.  Like  the 
Sophists,  he  considers  that  man  should  trouble  himself 
about  none  but  human  affairs.  Like  them,  he  thinks 
that,  apart  from  and  above  men  engaged  in  special  pro- 
fessions and  trades,  carpenters,  pilots,  and  doctors,  etc., 
there  is  man,  pure  and  simple,  who  calls  for  and 
deserves  distinct  culture  and  training.  Evidently 
Socrates  does  not  limit  philosophy  to  the  study  of 
human  things  for  the  same  reason  as  do  the  Sophists. 
The  latter  extolled  mankind  because  they  denied  the 
existence  of  the  gods.  Socrates  sees  the  proof  of  the 

1  Mem.  iv.  3.  i  ;  iv.  2.  n.  2  Ibid.  iv.  2.  ti. 


22  STUDIES  IN  PHILOSOPHY 

existence  and  greatness  of  the  gods  in  the  very  limits 
imposed  on  man.  Socrates  and  the  Sophists  arrive  at 
the  same  conclusion,  though  along  different  paths. 

In  this  comparison  between  Socrates  and  the  Sophists 
there  is  nothing  disparaging  to  our  philosopher  if  we 
form  a  correct  idea  of  sophistic.  The  Sophists  were 
something  more  than  the  destroyers  of  whom  Zeller 
speaks,  something  more  than  that  impersonal  echo  of  the 
prevailing  morals  that  Grote  would  have  us  believe.  It 
fell  to  the  creators  of  sophistic,  men  like  Protagoras  and 
Gorgias,  to  be  the  first  to  conceive  of  the  legitimacy 
and  utility  of  intellectual  culture  of  a  general  nature, 
applied  not  to  some  particular  faculty,  but  to  the  man  him- 
self, in  such  a  way  as  to  make  him  capable  of  acting  nobly 
under  all  circumstances.  To  gymnastics  the  national 
education  had  now  added  music,  or  the  teaching  of 
knowledge  which  moulds  the  intelligence.!  The  Sophists, 
however,  rose  to  a  loftier  conception  of  education,  the 
end  of  which  they  regarded  as  being  not  only  the  intro- 
duction into  the  mind  of  more  or  less  determined 
knowledge,  but  also  the  creation  of  universal  aptitudes. 
In  doing  this  it  may  be  said  that  they  brought  within 
the  sphere  of  consciousness  the  principle  which  had 
long  controlled  the  practical  life  of  the  Hellene^  and 
which  showed  itself  in  a  strange  admiration  for  men 
fertile  in  expedients,  and  skilled  in  getting  out  of  a 
difficulty  under  all  circumstances  j  men  like  Ulysses, 
Themistocles  or  Alcibiades.  The  special  form  the 
Sophists  gave  to  their  principle  indicates  even  more 
clearly  its  Hellenic  nature,  for  it  was  essentially  in  the 
ability  and  skill  to  speak  and  debate  that  they  placed  a 
man's  peculiar  wortrj  it  was  to  develop  this  virtue  in 
their  pupils  that  they  established  what  might  be  called 
intellectual  gymnastics. 


SOCRATES  23 

No  wonder  Socrates  approved  of  whatever  there  was 
in  sophistic  that  was  lofty  and  in  conformity  with  the 
genius  of  the  race.  But  he  did  not  therefore  accept  the 
principles  of  the  Sophists. 

Indeed,  the  thought  came  to  him  to  find  out  if  per- 
formance came  up  to  promise,  and  if  the  Sophists  really 
carried  out  that  intellectual  and  moral  education  the 
excellence  of  which  they  well  understood.  It  must  be 
confessed  that  the  process  he  adopted  to  assure  himself 
thereof  was  that  of  a  man  prepossessed  in  favour  of  a 
contrary  doctrine,  rather  than  that  of  an  impartial  critic 
who  unreservedly  sees  things  from  the  point  of  view  of 
his  interlocutors.  He  did  not  trouble  about  seeing 
people  at  work,  or  trying  to  discover  if  the  pupils  of 
the  Sophists  behaved  as  clever  politicians,  just,  clear- 
sighted men.  He  started  with  the  idea  that  the  proof 
of  ability  was  knowledge,  and  that  the  proof  of  know- 
ledge was  the  power  to  explain  to  others  what  one 
knows.1  Then  he  went  about  the  town,  questioning 
the  Sophists  and  their  pupils,  calling  upon  them  to  tell 
him  what  piety,  justice,  courage  and  virtue  were,  and 
satisfactorily  to  answer  all  possible  questions  thereon, 
without  ever  contradicting  themselves.  Not  one  came 
successfully  out  of  this  test ;  so  Socrates  concluded  that, 
though  the  Sophists  made  fine  promises,  their  perform- 
ances were  not  in  conformity  with  them. 

Now,  what  but  the  method  employed  by  the  Sophists 
could  be  the  cause  of  their  failure  ?  This  method  con- 
sisted of  practice  left  to  itself  and  rejecting  all  theory 
as  vain  and  useless  ;  it  was  art  considered  as  its  own 
means  and  end. 

Here  Socrates  saw  a  double  error.  In  the  first  place, 
art  cannot  be  an  end  unto  itself.  Consider  bodily 

1  Mem.  iv.  6.  i  ;  iii.  3.  n.     Cf.  Laches,  190  C. 


24  STUDIES  IN  PHILOSOPHY 

gymnastics.  If  you  admit  this  to  be  an  absolute  end, 
you  will  be  led  to  attach  as  much  importance  to  tricks 
of  strength  which  deform  the  body  as  to  the  well- 
planned  exercises  which  make  it  supple  and  strong. 
It  is  the  same  with  intellectual  gymnastics.  In  itself 
it  is  quite  as  likely  to  make  men  more  unjust  and 
wicked  as  to  make  them  more  just  and  noble.1  Will  it 
have  the  same  value,  then,  in  both  cases  ? 

There  is  more  than  this,  however.  Not  only  cannot 
art  be  an  end  unto  itself;  it  cannot  come  into  being 
from  exercise  and  practice  alone.  If  art  for  art's  sake 
is  dangerous,  art  by  means  of  art  is  impossible.  Is  it 
imagined,  as  Aristotle  says  later  on,  that,  according  to 
Socrates'  meaning,  in  teaching  a  man  the  trade  of  a 
shoemaker,  it  is  sufficient  to  place  in  his  hands  a  collec- 
tion of  ready-made  shoes  ? 2  To  call  forth  art  itself  is 
a  very  different  thing  from  imparting  the  products  of 
art.  A  pupil  trained  by  external  means  can,  more  or 
less  faithfully,  reproduce  whatever  he  has  seen  his  master 
do,  but  he  has  not  within  him  that  general,  self-sufficing 
ability  which  constitutes  true  art.  Art  is  independence, 
whereas  such  a  pupil  is  his  master's  slave.3 

Art  by  means  of  art  is,  in  a  word,  routine,  ignorance, 
chance.  Now,  a  man  must  be  very  simple-minded  if 
he  thinks  that,  whereas  it  is  impossible  to  become  a 
carpenter,  pilot,  or  general  unless  one  possesses  special 
knowledge  of  these  different  professions,  all  the  same, 
skill  in  the  general  conduct  of  life  can  spring  up  within 
us  as  the  result  of  mere  chancel  Take  any  mental 
quality  you  please,  if,  in  acquiring  it,  you  restrict  your- 
self to  practice  alone,  you  can  never  be  certain  you  will 

1  Mem.  iv.  3.  i.  2  Arist.  Soph.  Elench.  184  a  i. 

3  Mt'm.  iv.  7.  i  :  a&TdpKeis  iv  rcus  7rpocnr;Koij<rais  Trpd^effiv. 

4  I*>id.  iv.  2.  2  sqq.  ;  iii.  5.  21  sqq. 


SOCRATES  25 

not  end  in  the  very  opposite  of  what  you  are  aiming  at. 
Take  justice,  for  instance.  The  man  who  has  learnt  it 
by  nothing  but  practice  and  routine  will  regard  it  as 
consisting  of  certain  determined  modes  of  action  :  e.g. 
never  stealing  or  deceiving  another.  Deceit  is  just,  when 
you  are  dealing  with  enemies  ;  and  so  is  pillage,  when 
it  is  the  foe  you  are  plundering.1 

But  if  art  is  insufficient  unto  itself,  where  can  it  find 
the  rule  and  principle  it  needs  ?  Nowhere  but  in  just 
ideas  on  the  use  of  mental  qualities,  and  on  the  con- 
ditions of  these  very  qualities  :  in  a  word,  it  can  find 
them  only  in  science.  \The  Sophists  missed  their  goal 
because  they  were  too  eager,  and  made  straight  for  it, 
instead  of  proceeding  along  the  winding  path  which 
alone  could  have  led  them  to  it.  Before  laying  claim 
to  skill  in  practical  speech  or  deed,  one  must  acquire 
that  theoretical  knowledge  which  alone  confers  general 
ability.2  We  are  good  at  the  things  with  which  we  are 
acquainted,  and  bad  at  those  we  know  nothing  about.3 
Art  implies  science :  a  thing  the  Sophists  did  not  see. 

Such  were  the  views  of  Socrates  regarding  physics 
and  sophistic.  One  judgment  was  the  reverse  of  the 
other.  He  blamed  the  physiologists  for  not  having 
that  sense  of  human  affairs  which  he  praised  the 
Sophists  for  possessing  :  he  blamed  the  Sophists  for 
being  without  that  conception  of  science  which  he 
found  in  the  physiologists.  The  latter  had  applied  the 
form  of  science  to  something  that  goes  beyond  it:  the 
Sophists  had  omitted  to  apply  it  to  the  thing  that  re- 
quires and  admits  of  it.  Physics  was  science  isolated 
from  art  and  practical  life,  losing  itself  in  empty  specu- 
lations; sophistic  was  art  isolated  from  science  and  so 
reduced  to  dangerous  routine. 

1  Mem.  iv.  2.  14  sqq.         2  Ibid.  iv.  3.  i  ;  iii.  9.  4.         3  Laches,  194  D. 


26  STUDIES  IN  PHILOSOPHY 

Such  an  appreciation  of  physics  and  sophistic  natur- 
ally led  Socrates  to  collect  and  combine  the  principles 
which  to  him  appeared  viable  in  each  of  these  two 
disciplines,  i.e.  scientific  form,  on  the  one  hand,  and 
exclusive  preoccupation  about  human  things,  on  the 
other.  By  applying  to  the  object  of  sophistic  the  scien- 
tific form  invented  by  the  physiologists,  there  would  be 
established  a  wisdom  as  useful  as  art  and  as  universal 
and  communicable  as  science,  capable  of  moulding  man 
and  influencing  his  morals,  capable  also  of  being  self- 
sufficient  and  defending  itself  against  objections,  in  a 
word,  proportioned  to  the  forces  and  the  needs  of 
human  nature. 

iThis  idea  of  a  union  of  science  and  art  is  the  very 
germ  of  Socratic  philosophy.)  Socrates  does  not  first 
cultivate  science  and  art  separately,  and  make  them 
serve  each  other  afterwards.  To  his  mind,  each  strays 
from  the  path  whenever  it  claims  to  be  journeying  alone. 
In  their  close  co-operation,  their  mutual  penetration,  lies 
the  condition  of  their  existence  and  success. 

Here  we  find  determined  the  general  object  of 
Socrates'  investigations.  This  object  is  the  domain  he 
clearly  discerned  and  circumscribed  between  things  divine 
and  the  mechanical  arts,  i.e.  human  nature  in  what- 
ever it  offers  of  a  general  and  definable  character  ; l  it  is 
real  and  substantial  human  happiness,  as  distinct  from 
imaginary,  fragile  and  delusive  happiness  ; 2  it  is  the  art  of 
using  men  and  human  things  well,  not  only  under  certain 
circumstances  and  by  chance,  but  with  certainty  and  under 
all  circumstances ; 3  in  a  word,  it  is  all  that  is  necessary 
and  sufficient  for  the  making  of  an  honest  man. 

Such  was  his  idea  when  he  went  about  repeating 
the  Apollonian  maxim  :  Tif&Ot  a-avrov.  According  to 

1  Mem.  i.  i.  16.  2  Apol.  36  D.  3  Mem.  iv.  i.  2. 


SOCRATES  27 

Socrates  to  know  oneself  was  not  simply  to  be  con- 
scious, under  all  circumstances,  of  what  one  is  or  is  not, 
capable.  It  meant  entering  deep  into  one's  own  soul, 
beyond  the  particular  and  the  fleeting,  to  find  the  one 
identical,  permanent  substratum.  It  meant  finding  that 
secret  nature  we  carry  about  everywhere  with  us,  and 
which  contains  within  itself  the  conditions  of  our  wisdom 
and  happiness  far  more  than  do  external  things.  In  a 
word,  the  Socratic  maxim  is  an  exhortation  to  become 
conscious  of  whatever  in  us  is  general. 

Socrates  does  not  consider  the  Tvwdt  <ravrov  as  simply 
the  first  step  in  the  search  after  the  whole  of  truth.  He 
does  not  mean  that  knowledge  of  self  is  the  condition  of 
attaining  to  all  other  knowledge,  and  that  once  it  is 
acquired  we  shall  be  in  a  position  to  enter  upon  the 
search  for  all  the  rest.  The  TvwOt,  a-avrov  is  the  end 
as  well  as  the  beginning  of  science  ;  there  can  be  no 
other  science  for  man  to  acquire  than  that  of  himself. 

True,  we  read  in  the  Phaedrus  of  Plato l  that  Socrates 
regards  it  as  ridiculous  to  trouble  about  other  things, 
when  one  is  still  ignorant  of  oneself ;  this  passage  would 
seem  to  indicate  that  Socrates  merely  postpones  physical 
and  theological  research,  not  that  he  rejects  it.  Here, 
however,  he  is  speaking  ironically.  To  his  mind,  the 
time  will  never  come  for  taking  up  the  science  of 
universal  being,  because  man  will  never  know  himself 
completely.  Probably  no  one,  before  the  time  of  Socrates, 
was  as  conscious  as  he  was  of  the  infinite  complexity  and 
the  unfathomable  profundity  of  man's  moral  nature,  as 
we  see  from  the  passage  just  quoted  in  the  Phaedrus  : 
"  I  am  trying  to  find  out,"  he  says,  "  whether  I  am 
more  complicated  and  wicked  than  the  serpent  Typhon, 
or  if  I  am  of  a  simple  nature,  participating  in  divinity." 

1    229  E. 


28  STUDIES  IN  PHILOSOPHY 

How  could  Socrates  recognise  research — even  so  far  as 
to  postpone  it — which  had  not  man  for  its  object  ?  Apart 
from  human  things,  there  are  none  but  physical  or  divine 
things  and  the  mechanical  arts.  Now,  the  former  are 
beyond  man's  reach,1  and  the  rest,  such  as  the  art  of  the 
shoemaker,  the  carpenter,  the  wrestler,  and  the  pan- 
cratiast,  are  practised  very  well  by  special  men,  without 
the  aid  of  theoretical  science.2 

Moreover,  wisdom,  when  thus  restricted  to  man,  is 
that  which  is  of  the  greatest  interest  to  him.  Indeed, 
what  most  dignifies  human  nature  if  it  be  not  freedom, 
independence  with  regard  to  other  men  and  external 
matters,  and  the  possession  of  everything  necessary  for 
good  conduct  and  happiness  ?  Now,  what  kind  of 
occupation  is  capable  of  conferring  on  us  this  divine 
independence  ?  Not  the  mechanical  arts,  subjected  to 
the  needs  of  the  body  ;  not  advanced  astronomy  and 
geometry,  difficult  and  useless  sciences,  whose  object 
is  quite  foreign  to  the  human  soul.3  Close  investiga- 
tion will  reveal  to  us  the  fact  that,  under  all  circumstances, 
it  is  one  and  the  same  thing  that  makes  man  dependent 
and  a  slave,  to  wit,  ignorance  of  real  good  and  evil, 
ignorance  of  himself.4  Therefore,  what  is  to  set  man 
free  and  enable  him  to  be  sufficient  unto  himself,  under  all 
circumstances,5  is  science,  not  any  particular  science,  but 
the  knowledge  of  what  we  are  and  of  what  tallies  with 
our  nature. 

(Thus,  Socrates  regards  the  science  of  human  things 
as  the  object  most  worthy  of  man's  powers^  Great  is 
the  distance,  however,  between  the  idea  of  such  a  science 
and  its  realisation.  The  scientific  form,  as  we  find  it  in 
ancient  physiology,  is  not  adapted  to  things  dealing  with 

1  Mem.  iv.  7.  6.  2  Ibid.  iii.  5.  21  ;  iv.  2.  12. 

3  Ibid.  iv.  7.  2.          4  Ibid.  iv.  2.  22-23  ;  i.  i    16.          6  Ibid,  iv,  7    i. 


SOCRATES 


29 


the  moral  life,  nor  does  art,  as  the  Sophists  conceived  it, 
lend  itself  to  scientific  development.  For  the  physicists 
science  consisted  in  knowing  the  generation  of  things, 
in  being  able  to  say  whether  there  is  only  one  substance 
or  several,  whether  everything  is  immovable  or  in 
motion.  How  can  these  categories  be  applied  to 
intellectual  and  moral  things?  (On  the  other  hand, 
for  the  Sophists  there  is  nothing  fixed  or  universal  in 
human  nature,  good  and  happiness  are  entirely  relative 
to  individuals)  Human  things  offer  for  our  study  only 
an  infinite  number  of  particular  cases  unconnected 
with  one  another.  How  can  such  material  be  an  object 
of  science  ? 

The  idea,  then,  of  a  moral  science  such  as  Socrates 
had  conceived  it,  called  forth  a  double  task.  On  the 
one  hand,  the  idea  of  science  had  to  be  elaborated  so 
that  it  might  be  adapted  to  moral  things  ;  on  the  other, 
moral  things  had  to  be  looked  upon  from  such  a  bias, 
so  to  speak,  as  to  make  them  appear  fit  to  become 
objects  of  science.  A  mould  suited  to  the  matter  had 
to  be  made,  and  the  matter  rendered  capable  of  flowing 
into  the  mould.  The  mind  of  Socrates  was  directed  to 
the  solution  of  this  double  problem.  The  results  of 
his  reflections  on  both  points  may  be  grouped  together 
under  the  terms  dialectic  and  ethic.  Still,  we  must 
not  attribute  to  Socrates  a  dialectic  and  an  ethic  distinct 
from  each  other.  The  characteristic  of  his  dialectic  is 
that  it  is  built  up  with  a  view  to  his  ethic,  and  the 
characteristic  of  his  ethic  is  that  it  is  the  working  out 
of  his  dialectic.  They  are  only  two  phases  of  one  and 
the  same  discipline  :  the  more  or  less  artificial  duplica- 
tion of  the  "  Moral  Science." 

In  this  sense,  in  what  do  the  dialectic  and  ethic  of 
Socrates  consist  ?     In  the  details  of  his  philosophy  shall 


30  STUDIES  IN  PHILOSOPHY 

we  find  those  characteristics  that  seem  to  us  to  have 
indicated  his  general  conception  of  human  wisdom  ? 


Ill 

Both  Zeller  and  Schleiermacher  maintain  that 
Socrates,  far  from  being  a  merely  popular  moralist, 
does  not  limit  himself  to  moral  philosophy  :  he  follows 
after  true  science,  the  science  of  the  essence  of  things. 
First  of  all,  he  forms  a  universal  conception  of  science, 
regarding  it  as  consisting  of  the  methodical  determina- 
tion of  the  concept  or  the  expression  of  the  general 
element  of  the  things  given.  Then,  in  accordance  with 
the  law  of  the  human  mind,  he  applies  this  universal 
form  to  the  particular  incomplete  object  with  which 
experience  supplies  him.  This  object  happens  to  be 
human  life.  The  subsequent  task  of  the  Socratics 
consists  in  applying  this  very  form  to  the  other  domains 
of  reality.1 

According  to  this  interpretation,  the  Socratic  theory 
of  science  would  appear  to  have  a  distinct  existence. 
Logically,  if  not  chronologically,  it  would  seem  to  be 
anterior  to  and  independent  of  the  Socratic  ethic  ;  a 
system  of  symbols  which  the  philosopher  had  created 
from  quite  an  abstract  point  of  view,  and  without  con- 
sidering the  peculiar  nature  of  the  things  he  had 
undertaken  to  investigate. 

It  cannot  be  denied  but  that  this  interpretation 
accords  with  the  destiny  of  Socratic  philosophy.  Indeed, 
we  find  Plato  and  Aristotle  applying  to  the  whole  study 
of  nature  a  method  analogous  to  the  one  employed  by 
Socrates  in  the  investigation  of  moral  questions. 

1  Schleiermacher,   Werke,  iii.   2,  p.   300  sqq.  ;  Zeller,  Phil.  d.   Gr.  3rd 
edition,  vol.  ii.  93  sqq. 


SOCRATES  31 

But  does  an  interpretation  need  only  to  be  in 
agreement  with  the  historical  fortune  of  a  philosophy 
for  us  to  regard  it  as  the  faithful  expression  of  the 
thought  of  the  philosopher  himself?  To  find  out  what 
a  thing  is  in  its  true  nature  by  what  it  subsequently 
becomes  is  a  method  dear  to  Hegelians.  Indeed, 
to  their  mind,  creation  is  being  itself.  It  does  not 
seem,  however,  to  be  without  reason  that  Pascal  said  : 
"Sometimes  the  self-same  thoughts  develop  quite 
differently  in  others  from  the  way  in  which  they 
develop  in  their  author."  How  many  principles 
expand  or  shrink,  become  modified  or  transformed, 
when  they  pass  from  one  mind  to  another  which 
examines  them  from  its  own  point  of  view !  We 
could  not  say  with  Schleiermacher  and  the  Hegelians  : 
u  To  know  what  Socrates  was,  we  must  above  all  find 
out  how  Plato  came  to  regard  him  as  his  master." 
For  Plato  may  have  applied  the  Socratic  method  to 
objects  for  which  it  was  not  meant. 

Now,  if  we  consider  the  main  elements  of  this 
method,  one  by  one,  we  shall  find  that,  in  the  form 
in  which  they  appear  in  Socrates'  speeches,  they  can 
be  explained  only  by  a  continual  preoccupation  upon 
the  moral  object  to  which  they  are  to  be  applied.  We 
shall  not  find  Socrates  determining  the  idea  of  science 
for  itself,  and  afterwards  applying  it  to  morals.  Science, 
he  imagines,  can  be  separated  from  morals  only  in  a 
totally  abstract  manner,  in  language,  if  you  will,  never 
in  the  nature  of  things.  In  a  word,  we  shall  find 
Socrates  stating  the  logical  problem  in  the  following 
terms  :  of  what  should  science  consist,  in  order  that 
virtue  and  happiness  may  become  objects  of  science  ? 

First  of  all,  the  criterion  of  science,  in  the  mind  of 


32  STUDIES  IN  PHILOSOPHY 

Socrates,  is  agreement  with  itself,  and  the  power  to 
get  accepted  infallibly  by  all,  what  one  thinks  he 
knows.1  Socrates  does  not  show  himself  anxious  to 
confront  philosophical  doctrines  with  the  nature  of 
things  as  this  nature  is  capable  of  existing  in  itself, 
independently  of  the  conceptions  of  the  human  mind. 
According  to  him,  the  necessary  and  all-sufficient  con- 
dition of  certainty  lies  in  the  double  agreement  of  man 
with  himself  and  with  the  rest  of  mankind  ;  in  other 
words,  in  the  agreement  of  the  human  mind  with  itself. 

Now,  this  principle,  new  to  philosophy,  would 
indeed  be  strange  were  the  knowledge  of  being  and 
of  the  universal  principles  of  nature  the  object  of 
philosophy.  In  that  case,  if  we  would  understand 
Socrates'  doctrine,  we  should  have  to  infer  that  he  was 
already  identifying  human  thought  with  the  principle 
of  being  in  general.  But  such  identification  was 
possible  only  when  several  regions  had  been  distin- 
guished in  the  human  mind,  and  the  existence  of  an 
eternal  reason  had  therein  been  found.  Such  an 
analysis  was  the  distinctive  work  of  Plato  and  Aris- 
totle. Socrates,  for  his  part,  clearly  distinguishes 
opinion  from  reasoning,  but  he  goes  no  farther  ;  he 
considers  that  our  power  of  reasoning  cannot  claim  to 
know  the  first  principles  and  final  ends  of  things. 

On  the  other  hand,  it  may  well  be  understood  that 
the  agreement  of  the  human  mind  with  itself  should  be 
looked  upon  as  the  criterium  of  truth,  if  we  are  dealing 
only  with  truth  in  moral  affairs.  For  it  is  quite 
natural  to  admit  that,  innate  in  the  human  mind,  there 
is  the  general  idea  of  what  is  suitable  for  man,  and  that 
this  intellectual  substratum  is  the  same  in  all  individuals. 
It  is  this  that  is  called  common  sense,  a  sure  guide  so 

1  Alcibiades  I.  iii.  D-E  ;  Mem.  iv.  6.  i  and  15. 


SOCRATES  33 

long  as  we  are  concerned  with  the  conduct  of  life,  but 
pregnant  with  error  when  dealing  with  the  knowledge 
of  the  laws  of  the  universe. 

Now,  to  what  object  must  one  apply  oneself  to 
realise  that  agreement  with  oneself  and  the  rest  of  the 
world  which  forms  the  condition  of  certainty  ?  In 
other  words,  what  is  the  matter  proper  to  science  ? 

Here  we  find  what  constitutes  the  essence  of  the 
logical  doctrine  of  Socrates,  that  original  and  fruitful 
principle  which  was  to  remain  the  guide  of  the  human 
mind  for  two  thousand  years.  Science,  asserted  Socrates, 
has  for  its  object  that  which  is  general.  There  is  no 
science  of  the  individual,  of  the  accidental,  of  particular 
things  as  they  are  presented  to  us.  The  object  of  the 
science  of  courage,  for  instance,  is  not  courageous 
deeds,  it  is  that  which  is  common  to  all  courageous 
deeds,  it  is  the  answer  to  the  question  :  rL  ea-rcv  q 
avSpeia ;  it  is,  as  Plato  says  later  on,1  TO  Bia  Travrav 
Trepl  avSpeias  7r€(f>v/c6<;.2 

This  maxim  is  the  very  one  advanced  to  prove  that 
Socrates  considered  science  in  itself,  apart  from  the 
matter  to  which  it  must  be  applied.  But  though  it 
is  true  that  the  maxim  of  Socrates  became  after  his 
time  a  logical  and  even  metaphysical  doctrine  superior 
to  any  particular  domain,  it  does  not  therefore  follow 
that  it  was  so  to  himself.  This  will  be  evident  if, 
instead  of  considering  it  separately,  it  is  replaced  in 
the  ensemble  of  the  Socratic  philosophy. 

The  whole  work  of  Xenophon 3  clearly  shows  that 
Socrates  never  sought  the  general  except  in  human  things. 

Consequently  the  matter  at  issue  has  less  bearing  on 
the  question  of  fact  than  on  that  of  right. 

1  See  Mem.  i.  i.  16.  2  Laches,  192  E. 

3  See  principally  Mem.  i.  i.  16. 


34  STUDIES  IN  PHILOSOPHY 

What  was  it  that  Socrates  meant  by  the  general,  and 
why  did  he  see  in  it  the  only  object  that  admitted  of 
scientific  knowledge  ? 

By  the  general,  Socrates  did  not  mean  the  simple 
permanent  element  which  may  lie  hidden  in  the  com- 
pound things  that  strike  our  senses.  In  reality,  that 
is  not  the  general  ;  it  is  rather  substance,  that  is  to  say, 
the  very  object  which  physicists  had  considered  and 
which  Socrates  regards  as  inaccessible.  On  the  other 
hand,  the  general  is  not  yet  to  him  what  it  will  be  to 
Plato  and  Aristotle  :  the  normal  type  of  a  species,  the 
natural  being  as  it  would  be  if  the  cause  peculiar  to  it 
were  acting  alone  without  being  opposed,  as  happens 
in  the  sensible  world,  by  outer  influences.  The  general, 
of  which  Socrates  speaks,  is  not  related  to  the  material 
world,  nor  even  to  an  intelligible  world  :  strictly 
speaking,  it  is  the  common  substratum  of  men's 
speeches  and  actions.  Socrates  starts  with  the  idea 
that  the  reason  we  use  one  and  the  same  word,  justice, 
to  designate  quite  different  modes  of  action,  such  as 
doing  good  to  one's  friends  and  doing  evil  to  one's 
enemies,  lies  in  the  fact  that  we  have  in  mind  a  certain 
notion  which  is  single  in  its  nature,  and  the  object  of 
which  we  find  in  the  various  actions  we  designate  as 
just.  And  as  men,  when  they  talk  to  one  another  in 
good  faith,  come  sooner  or  later  to  agree  as  to  the  use 
of  their  words,  the  ideas  represented  by  these  words  are 
bound  to  be  identical  in  the  minds  of  all. 

And  now,  why  does  Socrates  make  the  general,  thus 
understood,  the  proper  object  of  science  ? 

Because  he  finds  in  it  the  necessary  and  sufficient 
condition  of  that  agreement  with  oneself  and  others 
which,  in  his  mind,  is  the  mark  of  knowledge. 

Apart  from  these  determined,  fixed  notions,  which 


SOCRATES  35 

form  the  foundations  of  words,  there  is  no  guiding- 
mark  for  the  mind  in  its  reasonings,  and  therefore  no 
means  of  coming  to  an  understanding  with  oneself  and 
others.  On  the  other  hand,  it  is  sufficient  to  make 
one's  discourse  conform  to  those  general  notions  on 
which  all  men  are  agreed,  to  be  sure  of  obtaining  the 
assent  of  one's  interlocutors.  Why  does  Homer  call 
Ulysses  an  orator  sure  of  success?  Because  Ulysses, 
in  his  discourse,  is  guided  by  ideas  which  all  men 
accept  :  Sia  rwv  SOKOVVTOW  dvdpcorroi^.1 

Francis  Bacon,  the  modern  legislator  of  the  sciences 
of  nature,  said,  not  without  reason,  that  from  human 
language  one  can  deduce  only  words,  not  things,  if  we 
would  know  the  nature  of  the  external  world  ;  but 
human  language  is  certainly  the  first  testimony  that 
must  be  consulted  if  it  is  desired  to  become  acquainted 
with  the  thoughts  and  wishes  of  the  human  mind. 
There  is  nothing  to  indicate  that  the  categories  of 
language  reproduce  those  of  things  ;  but  it  is  evident 
they  are  an  image  of  the  categories  of  our  thoughts 
and  actions.  The  discourse  of  men  can  supply  the 
physicist  only  with  an  altogether  provisional  ensemble 
of  signs  and  conjectures.  Such  language,  when  dealing 
with  moral  philosophy,  is  the  very  thing  we  have  to. 
fathom. 

If  we  now  consider  in  detail  the  method  of  Socrates, 
we  find  that  it  consists  of  two  parts  which  may  be 
designated  by  the  names  of  exterior  form  and  logical 
substratum.  The  former  consists  of  dialogue  along 
with  certain  features  peculiar  to  Socrates,  such  as  irony 
and  maieutics,  as  well  as  the  leading  rble  assigned  to 
self-possession  and  love.  Logical  substratum  consists 

1  Mem.  iv.  6.  15. 


36  STUDIES  IN  PHILOSOPHY 

of  definition  and  induction.     Each  of  these  parts,  accord- 
ing to  Socrates,  has  a  special  aspect. 

Zeller  says  that  the  reason  Socrates  makes  use  of  the 
dialogue  form  is  that  he  is  conscious  of  his  own  ignor- 
ance, because  of  the  contradictions  he  finds  in  the 
various  systems  of  philosophy,  and  that  it  is  his  desire 
to  escape  from  this  state  of  ignorance.  Hence,  accord- 
ing to  Zeller,  his  disposition  to  appeal  to  others,  with 
the  object  of  discovering  if  perchance  they  are  in 
possession  of  the  very  science  he  lacks. 

This  explanation  is  not  altogether  satisfactory.  In 
the  first  place,  Socrates  does  not  consult  his  interlocutors 
on  things  in  general,  but  only  on  what  concerns  man- 
kind :  he  expects  to  learn  nothing  from  the  dialogue 
form — any  more  than  from  any  other  method  of  in- 
vestigation— about  physical  things.  Then,  too,  Socrates 
does  not  see  in  the  dialogue  form  merely  a  convenient 
and  suggestive  method  of  philosophising ;  to  his 
mind  dialectic  cannot  be  distinguished  from  wisdom 
itself. 

Though  investigation  into  the  causes  of  the  world  is 
a  matter  of  solitary  speculation,  it  is  not  the  same  with 
investigation  into  the  conditions  of  human  life.  How 
can  man  be  known,  except  by  conversing  with  men  ? 
And  if  science  consists  in  discovering  the  points  on 
which  all  men  are  agreed,  and  which  form  the  substratum 
of  all  their  judgments  (ra  fiaKicrra  opoX.oyovfjieva),  what 
shorter  and  more  certain  method  can  one  have  than  to 
bring  together  men's  opinions  and  compare  them  with 
one  another  ?  In  a  word,  if  science  must  be  used  for 
instructing  men  and  persuading  them  of  things  of  which 
we  have  become  certain,  once  for  all,  is  not  methodical 
conversation,  from  its  beginning  right  on  to  its  end, 
an  integral  part  of  philosophy  and  wisdom  ? 


SOCRATES  37 

Consequently,  it  is  not  from  modesty,  from  deference 
to  the  science  of  others,  that  Socrates  constantly  speaks 
of  examining  things  in  common,  Kotvfj  (3ov\€ve<r0ai,,1 
Koivrf  a-KeTTTecrOai,,  tcoivf)  tyreiv,  (rvtyrelv  :  this  form  of 
investigation  is  implied  in  the  very  object  he  has  in 
view.  For  a  dissertation  on  the  principles  of  nature, 
writing  is  sufficient ;  but  to  know  men  and  succeed  in 
convincing  them,  one  must  converse  with  them. 

Socratic  dialogue  frequently  assumes  the  form  of 
irony.  Socrates  puts  his  questions  without  ever  answer- 
ing them,2  and  thus  brings  his  interlocutor  either  to  the 
point  of  contradicting  himself  or  coming  to  a  dead  stop, 
and  acknowledging  his  ignorance  of  the  very  things  he 
thought  he  knew.3 

Now,  the  use  of  such  a  method  is  far  more  compre- 
hensible when  dealing  with  the  knowledge  of  human 
things  than  when  dealing  with  that  of  nature.  How, 
when  dealing  with  external  things,  can  a  man  confine 
himself  to  questioning  others  without  bringing  their  asser- 
tions face  to  face  with  reality  itself?  In  order  profitably 
to  undertake  such  questioning,  would  not  a  man  need 
previously  to  have  shown  himself  competent  in  both 
physics  and  metaphysics  ?  And  again,  would  not  the 
listeners  also  need  to  be  specially  competent  if  their 
judgment  on  the  discussion  is  to  be  of  any  value  ?  But 
when  dealing  with  human  things,  every  one  is  competent, 
for  he  bears  within  himself  just  the  touchstone  needed 
for  the  testing  of  opinions.  In  conversation  itself,  the 
questioner  may  find  all  that  is  needed  for  proving  that 
his  interlocutor  is  not  only  in  flagrant  contradiction  with 
himself,  but  with  the  very  nature  of  things  as  well. 
Moreover,  is  it  not  especially  such  human  qualities  as 

1  Mem.  iv.  5.  12.  2  Arist.  Soph.  el.  ch.  xxxiii. 

3  Plat.  Repub.  i.  337  A  E  ;  Sophistes,  183  B. 


38  STUDIES  IN  PHILOSOPHY 

piety  and  justice,  courage  and  virtue,  with  whose  nature 
every  man  thinks  he  is  acquainted,  though  really  such 
is  not  the  case  ?  The  physiologists  would  not  have 
accepted  the  contest  to  which  Socrates  invited  his  inter- 
locutors. Only  such  men  as  were  occupied  in  moral 
affairs  could  submit  to  such  a  mode  of  questioning  : 
only  they,  in  fact,  did  so. 

It  is  the  same  with  maieutics.  Socrates  would  have 
us  think  that  he  is  barren  as  regards  wisdom  ;  but  by 
his  questions  he  helps  others  to  bring  to  birth  what  they 
bear  in  their  own  mind,  and  that  unconsciously.  Then, 
after  thus  eliciting  the  secret  ideas  of  his  interlocutors, 
he  carefully  examines  whether  the  offspring  of  their  soul 
is  nothing  but  fancy,  or  fruit  that  is  real  and  capable  of 
living.1  What  must  we  think  of  such  a  method  ? 

Socrates,  we  are  told,  considers  himself  barren  as 
regards  wisdom.  What  kind  of  wisdom  is  here  meant, 
if  not  practical  wisdom,  which  indeed  has  the  strange 
characteristic  of  being,  in  one  aspect,  incapable  of  com- 
munication, of  existing  within  us  only  if  it  is  ourself, 
of  being  produced  within  our  person  only  if  it  springs 
forth  from  our  own  inmost  nature  ? 

How  is  Socrates  able  to  generate,  in  the  minds  of 
his  interlocutors,  ideas  likely  to  be  true  and  capable  of 
living  ?  This  doctrine  is  a  very  strange  one,  if  we  are 
dealing  with  physical  or  metaphysical  truths.  That 
audacious  doctrine  which  identifies  the  mind  of  man 
with  the  principle  of  things  is  nowhere  to  be  found  in 
Socrates  :  if  it  happens  that  he  predicts  the  future 2  it 
is  not  by  the  might  of  his  intelligence  alone,  but  owing 
to  a  mysterious  and  quite  supernatural  revelation. 
Maieutics,  however,  is  a  very  reasonable  and  legitimate 
method,  if  our  object  is  to  bring  moral  truths  before 

1  Theaet.  149,  157  c.  2  Mem.  i.  i.  5. 


SOCRATES  39 

men,  for  these  truths  are  nothing  but  the  expression 
and  reflective  knowledge  of  human  nature  :  and  human 
nature  is  what  every  man  has  within  himself.  The 
fiction  of  Meno  is  a  Platonic  and  paradoxical  extension 
of  Socratic  maieutics.  Socrates,  for  his  part,  elicits 
from  the  minds  of  his  listeners  only  knowledge  that 
relates  to  piety,  justice,  temperance,  courage,  urban 
government,  and  everything  that  goes  to  make  up  an 
honourable  man.1 

Finally,  how  can  Socrates,  who  professes  to  be  ignor- 
ant, rightly  estimate  the  true  value  of  the  fruit  which 
he  assists  human  intelligence  in  bringing  forth  ?  Are 
we  not  here  dealing  exclusively  with  those  moral  and 
practical  ideas  upon  which  every  man,  in  his  human 
capacity,  is  competent,  when  in  forming  his  judgments 
he  imposes  silence  on  his  distinctive  tastes  and  passions 
and  puts  himself  just  at  that  point  of  view,  superior  to 
the  individual,  which  Socrates  had  defined  ? 

Dialectic,  besides,  possesses  two  very  remarkable 
moral  conditions  :  self-possession  and  love  :  eyKpdreia 
and  6/9o>9* 

"  To  those  who  are  self-possessed,  and  to  them  only, 
is  it  given  to  investigate  the  best  in  everything,  and, 
distinguishing  things  by  a  dialectic  of  actions  and  words, 
according  as  they  belong  to  the  good  or  the  evil,  to 
choose  the  one  and  abstain  from  the  other." 2  It  is 
because  dialectic  has  for  its  object  the  determination  of 
the  value  of  things,  from  the  moral  and  human  point 
of  view,  that  self-possession  is  its  essential  condition. 
Indeed,  the  true  moral  value  of  things  lies  in  the  interest 
they  offer  to  human  nature  in  general,  not  to  the  indi- 
vidual, regarded  from  the  standpoint  of  his  tastes  and 
passions,  which  are  superficial  and  fleeting.  Now,  it  is 

1  Mem.  i.  i.  16.  2  Ibid.  iv.  5.  n. 


40  STUDIES  IN  PHILOSOPHY 

owing  to  self-control  that  man,  in  his  judgments,  lays 
his  individual  and  accidental  preferences  on  one  side. 

And  finally,  love,  epw?,  plays  an  important  part  in 
the  dialectic  of  Socrates.  The  same  may  be  said  regard- 
ing all  the  Socratics.  Not  only  Xenophon  and  Plato, 
but  also  Euclid,  Crito,  Simmias  and  Antisthenes,  have 
written  on  the  subject  of  love.  What  is  the  love\that 
is  here  meant  ?  Doubtless  Socrates  does  not  mean 
friendship,  pure  and  simple,  but  rather  affection  mingled 
with  sense  attraction.  It  is  a  kind  of  spiritual  ardour 
that  enters  the  whole  man,  causing  in  him  an  emotion 
that  has  nothing  to  do  with  mere  friendship.  Evidently 
Socrates  disparages  physical  love,  though  not  in  all  its 
elements.  He  retains  its  soul-uplifting  charm,  which 
is  lacking  when  the  intelligence  alone  is  at  play. 
He  keeps,  one  might  say,  its  vital  impulse,  if  not  its 
object.1 

This  love,  moreover,  could  not  go  to  the  point  of 
passion  and  frenzy,  like  the  love  of  which  Plato  speaks  in 
Phaedrus.  Even  here  self-possession  is  still  a  superior, 
inviolable  duty.  The  Platonic  distinction  between  good 
and  evil  frenzy  would  have  been  rejected  by  Socrates, 
to  whom  all  frenzy  is  slavery. 

How  is  the  rule  that  governs  such  a  mental  state  to 
be  explained  ? 

Certainly  Socrates  does  not  dream  of  investing  love 
with  the  rble  that  Plato  assigns  to  it,  and  which  consists 
in  introducing  us  into  the  world  of  beauty,  as  into  the 
vestibule,  as  it  were,  of  divine,  transcendent  truth.  In 
order  that  love  might  appear  as  endowed  with  such 
power,  it  would  have  to  be  a  state  of  rapture  and  ecstasy, 
whereas  Socratic  love  is  inseparable  from  self-possession. 
Already  Socrates  condemns  poets  for  writing  poetry 

1  Xen.  Banquet,  ch.  viii. 


SOCRATES  41 

not  by  science  but  by  enthusiasm,  a  kind  of  divine 
inspiration.1  With  all  the  more  reason  would  he  have 
condemned  as  sacrilegious  the  claim  that  the  secrets 
the  gods  have  removed  from  our  mental  grasp  could 
be  reached  in  a  state  of  frenzy. 

In  investigations  upon  human  things  there  is  room 
for  a  kind  of  love  which  combines  sense  attraction  and 
self-possession.  In  accordance  with  the  principle  of 
maieutics,  the  soul  must  bring  forth  its  wisdom  from 
itself,  just  as  the  body  brings  forth  from  itself  the  fruit 
to  which  it  gives  birth.  Therefore  the  soul,  as  well  as 
the  body,  must  be  impregnated.  Love  here  intervenes 
for  the  purpose  of  playing  a  part  similar  to  that  it 
plays  in  physical  procreation.  Intelligences  impregnate 
each  other,  as  bodies  do.  By  the  influence  of  noble 
love  the  soul  becomes  big  with  noble  thoughts  and 
feelings.  "  Orestes  and  Pylades,  Theseus  and  Pirithous, 
and  several  other  demi-gods  are  famous  .  .  .  because, 
admiring  one  another,  they  performed  together  the  most 
glorious  deeds." 2  Moreover,  it  was  a  familiar  idea 
amongst  the  Greeks  that  the  mutual  love  of  youths 
exalted  their  courage,  and  made  them  capable  of  mighty 
actions. 

And  so  we  find  that  dialogue,  irony,  maieutics,  self- 
possession  and  love,  all  of  which  are  elements  of  the 
Socratic  method,  if  regarded  not  as  abstract  formulas 
but  rather  in  their  historical  aspect,  testify  to  a  reflective 
and  exclusive  preoccupation  to  establish  the  science  of 
morals.  But,  so  far,  these  are  nothing  but  the  externals 
of  the  method.  What  must  we  think  of  that  which 
constitutes  their  basis,  to  wit,  of  the  process  of  refu- 
tation which,  in  some  way,  makes  up  the  negative 
method,  and  of  the  processes  of  definition  and  induction 

1  Plat.  Apol.  22  B-C.  2  Xen.  Banquet,  ch.  viii. 


42  STUDIES  IN  PHILOSOPHY 

of  which  the  positive  method  consists  ?  Does  it  not 
appear  that  here,  at  all  events,  we  have  to  do  with 
instruments  that  are  really  of  universal  importance,  and 
with  conditions,  not  merely  of  the  science  of  morals, 
but  of  science  in  general,  whatever  be  its  object  ? 

Of  what  does  the  Socratic  refutation  consist  ?  Socrates 
begins  by  eliciting  or  drawing  forth  from  the  problem 
in  question  the  very  datum  he  presupposes.1  For 
instance,  if  he  is  told  that  any  one  man  is  a  better 
citizen  than  another,  he  asks  his  interlocutor  what,  in 
his  mind,  constitutes  a  good  citizen.  When  the  other 
man  has  replied,  Socrates  asks  him  additional  questions, 
dealing  with  cases  to  which  the  term  "  good  citizen  "  is 
generally  applied.  By  this  method  he  makes  him  give 
answers  that  are  incompatible  with  the  original  reply  : 
the  result  being  that  the  definition  put  forward  was 
either  too  restricted  or  too  wide,  or  defective  in  some 
other  way.2 

Socrates  applies  this  method  of  refutation  to  the 
judgments  either  of  ordinary  men,  politicians,  poets  and 
artists  of  renown,3  professors  of  eloquence  and  virtue, 
or  of  sophists  ;  in  a  word,  he  applies  it  to  all  ideas  that 
deal  with  morals  ;  but  we  do  not  find  that  he  made  use 
of  it  to  refute  physical  or  metaphysical  doctrines.  As 
regards  the  latter,  he  contents  himself  with  emphasising 
the  contradiction  that  prevails  between  the  various  ideas 
of  philosophers. 

Naturally,  the  Socratic  method  of  refutation  may  be 
employed  under  all  circumstances,  but  its  most  legitimate 
use  is  in  regard  to  morals.  If  we  carefully  notice,  we  find 
that  Socrates  bases  the  truth  of  any  given  particular  asser- 

1  Mem.  iv.  6.  1 3  :  tvl  rrjv  vir&deffw  fTravrjyev  &v  Trdvra  rbv  \6yov. 

2  For    instance,    Mem.    iv.    2  :    Conversation   between    Socrates    and 
Euthydemus. 

3  Apol.  ch.  vi.  to  viii. 


SOCRATES  43 

tion  on  knowledge  of  the  general  principle  relating  to 
that  assertion.  Now,  such  a  method  is  incomprehensible, 
if  we  are  dealing  with  the  order  of  physical  realities,  where 
the  particular  is  given  before  the  general.  Is  it  con- 
ceivable that,  when  affirming  we  see  the  sun  turning 
round  the  earth,  we  should  be  interrupted  and  asked 
whether,  before  expressing  ourselves  in  this  way,  we 
have  assured  ourselves  that  we  know  what  sight  and 
movement  are  ?  All  philosophies — even  ancient  philo- 
sophy— have  necessarily  subordinated  knowledge  of  the 
principles  of  physics  to  the  facts  and  appearances  that 
have  to  be  explained,  not  the  existence  of  facts  or 
appearances  to  a  knowledge  of  the  principles.  In  the 
moral  order  of  things,  however,  the  particular  is  not 
"  given "  :  it  is  to  be  sought  for.  Aristides  is  not 
"  given  "  to  me  as  a  virtuous  man  :  I  ask  myself  if  I 
ought  to  declare  him  to  be  so.  The  conduct  I  should 
observe  if  I  would  practise  piety  is  not  "  given  "  :  it  is 
to  come,  it  is  only  possible.  And  how  can  it  be  deter- 
mined except  by  starting  from  the  general  idea  of  piety? 
Socrates  is  therefore  right  in  subordinating  the  truth  of 
particular  judgments  to  the  knowledge  of  the  general, 
if  he  is  specially  considering  the  moral  domain  ;  for 
here  the  particular  is  nothing  more  than  we  make  it ; 
and  we  make  it  of  such  or  such  a  nature  only  by  virtue 
of  the  ideas  inherent  in  our  own  mind.  Now,  universal 
principles  exist  in  most  men  only  under  the  form  of 
habits  or  blind  instincts  ;  hence  the  principitancy  and 
inconsistency  noticed  in  their  judgments.  It  is  the 
very  object  of  the  method  of  Socrates  to  substitute 
deliberate,  resolute  maxims  for  these  blind,  wavering 
opinions. 

But  we  have  not  yet  entered  upon  the  two  Socratic 
processes,  which,  more  than   all   others,  appear  to   be 


44  STUDIES  IN  PHILOSOPHY 

of  universal,  theoretical  application :  definition  and 
induction  ; l  definition,  the  supreme  object  of  dialectic  ; 
induction,  the  methodical  march  leading  to  definition. 

Definition  is  the  adequate  expression  of  that  general 
essence  which  is  the  object  of  science.  The  Socratic 
definition  possesses  this  in  particular  :  it  does  not  con- 
fine itself  to  offering  a  distinctive  sign  of  things  ;  it 
claims  to  set  forth  the  necessary  and  all  -  sufficient 
condition  of  their  existence.  It  not  only  states  what 
the  thing  is,  seen  from  without,  it  even  tries  to  discover 
what  is  capable  of  producing  it.  For  instance,  to  call 
a  just  man  the  one  who  does  just  things  is  not  to 
define  him.  It  is  possible  to  do  just  things  by  chance, 
not  by  justice ;  and  one  may  be  just  without  mani- 
festing justice  within  oneself.  On  the  other  hand,  to 
say  that  the  just  man  is  he  who  knows  what  the 
laws  ordain  with  reference  to  men,  is  to  offer  a  true 
definition.  For  we  do  not  find  that  men  ever  do 
anything  else  than  what  they  'think  they  ought  to  do, 
and  those  who  know  justice  will  necessarily  do  just 
things  under  all  circumstances.2  They  have  within 
themselves  the  universal  capacity  for  justice. 

Thus  the  Socratic  definition  consists  of  the  declara- 
tion of  the  inner  capacity,  of  which  the  thing  to  be 
defined  is  the  outer  manifestation. 

Now,  where,  in  the  first  place,  is  this  distinction 
between  the  concrete,  particular  thing  and  the  invisible, 
general  power  to  be  found,  if  not  in  man  ?  And  does 
not  this  search  after  a  metaphysical  essence — justified, 
if  we  are  dealing  with  the  human  soul,  by  consciousness 
itself — become  extremely  rash  and  dangerous  if  we 
claim  to  practise  it  with  regard  to  the  outer  phenomena 
of  nature. 

1  Arist.  Met.  xiii.  4.  1078  b  25.  2  Mem.  iv.  6.  6. 


SOCRATES  45 

Why,  too,  does  Socrates  regard  the  capacity,  or 
total  principle  of  the  action  as  reposing  in  an  idea,  in 
the  knowledge,  pure  and  simple,  of  the  conditions  of 
action,  leaving  aside  the  force  necessary  to  realise  it? 
The  reason  is  that,  in  man,  force  or  activity  is  ever 
present,  and  is  always  determined  conformably  with 
knowledge.  Such,  at  all  events,  is  the  opinion  of 
Socrates  regarding  will.  Will  is,  as  it  were,  a  constant 
datum  which  it  is  practically  needless  to  mention. 
It  would  not  be  so  were  we  dealing  with  the  pro- 
duction of  physical  phenomena,  for  in  the  latter  case 
the  nature  of  the  generating  causes  and  their  mode 
of  action  are  unknown  and  inaccessible. 

To  arrive  at  definition,  thus  regarded,  the  method 
Socrates  uses  is  induction.  This  operation  consists  of 
two  parts,  which  may  be  called  invention  and  discussion. 

To  discover  the  general  essence,  Socrates  takes  as 
his  starting-point  a  certain  number  of  instances  of  the 
thing  to  be  defined.  These  instances,  however,  do 
not  consist  of  natural  facts,  directly  observed  :  Socrates 
takes  them  exclusively  from  human  discourse.  Language, 
opinions,  ordinary  judgments  or  even  nature  seen 
through  man  :  such  is  the  material  of  which  his 
induction  is  formed,  such  the  ground  in  which  it 
must  germinate.  From  the  outset  Socrates  interests 
himself  preferably  in  the  feelings  of  men  regarding 
paltry  matters  and  commonplace  pursuits.1  Initiation 
into  the  lesser  mysteries,  he  says,  must  precede  initiation 
into  the  greater.  This  is  the  reason  he  is  constantly 
speaking  of  shoemakers  and  smiths,  carpenters  and 
drovers :  a  reproach  brought  against  him  by  his 
enemies.2 

To  observation,  as  thus  understood,  Socrates  adds 

1  Gorgias,  ch.  li.  p.  497  B-c.  2  Mem.  i.  2,  37. 


46  STUDIES  IN  PHILOSOPHY 

analogy.  He  appeals  to  things  his  interlocutor  knows ; 
and,  showing  him  the  resemblance  between  these  things 
and  those  that  form  the  subject  of  conversation,  he 
draws  him  on  to  the  discovery  that  even  the  latter 
were  not  really  unknown  to  him.1  What,  for  instance, 
constitutes  a  just  man  ?  We  know  that  a  carpenter 
is  a  man  who  knows  carpentry  ;  a  musician  is  one  who 
knows  music ;  a  doctor,  one  who  knows  medicine. 
Our  conclusion,  by  analogy,  will  be  that  the  just  man 
is  the  man  who  knows  justice.2  The  usual  and,  as  it 
were,  essential  theme  of  these  analogies  consists  in  the 
transition  from  mechanical,  special  arts  to  moral  and 
general  art ;  in  a  word,  the  transition  from  things  of 
the  body  to  those  of  the  soul. 

Still,  observation  and  analogy  give  only  provisional 
results  :  discussion  alone  affords  decisive  ones.  Having 
once  invented  a  general  formula  by  means  of  carefully 
chosen  cases,  Socrates  considers  the  greatest  possible 
number  of  cases  and  applies  his  formula  to  all  these 
instances,  retaining  it  unmodified  if  it  emerges  success- 
fully from  the  test,  and  suitably  modifying  it  if  it  docs 
not.  Not  only  does  he  vary,  he  even  reverses  the 
experiment,  trying  to  find  a  definition  for  the  contrary 
object,  and  ascertaining  whether  this  new  definition 
is  to  the  former  what  negation  is  to  affirmation. 

Such  is  Socratic  induction.  Now,  all  the  details  of 
this  process  are  applicable  to  human  things,  whereas 
they  are  inapplicable  to  physical  or  metaphysical  things. 

To  take  as  one's  starting-point  the  language  and 
discourse  of  ordinary  life,  and  not  external  facts,  is  a 
method  that  may  rightly  be  regarded  as  meaningless 
and  fantastic  if  our  object  is  to  know  the  absolute 
essence  of  being  and  of  things  ;  but  it  is  a  very  natural 

1  Xen,  Economicus,  19.  15.  2  Gorgias,  460  B. 


SOCRATES  47 

and  legitimate  method  if  our  object  is  to  find  out  what 
lies  at  the  bottom  of  human  judgments.  It  is  also 
quite  conceivable  that  the  philosopher  should  bestow 
particular  attention  upon  common  and  ordinary  things, 
if  his  express  purpose  is  to  know  man,  for  it  is  in  this 
order  of  things  that  human  nature  appears  as  it  really 
is,  stripped  of  the  mask  of  convention  and  false 
knowledge. 

The  complaisant  use  of  the  method  of  analogy  and 
the  fact  that  this  mode  of  reasoning  is  regarded  as 
proof,  would  indicate  anything  but  a  scientific  mind 
if  one's  investigation  were  compelled  to  cover  every 
domain  of  reality.  But  if  we  are  to  move  in  one  and 
the  same  domain  only,  and  that  the  domain  of  human 
things,  then  analogy  is  a  useful  method  to  follow. 
For  then,  its  action  is  limited  to  passing  from  one 
species  to  another  in  the  same  genus,  and  that,  too, 
in  the  order  of  things  most  familiar  to  us,  in  which  we 
need  only  retire  within  ourselves  to  find  points  of 
reference  at  each  step. 

In  short,  the  Socratic  process  of  discussion  and 
control  is  a  very  uncertain  and  inadequate  method, 
if  we  would  have  knowledge  of  the  things  of  nature. 
Socrates  endeavours  to  verify  his  induction  by  examin- 
ing every  instance  that  offers  itself.  But  how  can  one 
gather  together  all  the  instances  of  one  and  the  same 
genus  in  the  order  of  physical  and  material  things? 
How  can  we  call  up  at  will  the  manifestations  of  the 
essence  contrary  to  the  one  whose  definition  we  are 
seeking  ?  Doubtless  modern  experimentation  must 
have  realised  these  conditions  to  some  extent ;  but 
the  ancients  had  no  idea  whatever  of  such  a  method 
of  investigation.  On  the  other  hand,  they  must  have 
thought  that,  in  human  things,  the  conditions  in 


48  STUDIES  IN  PHILOSOPHY 

question  were  quite  realisable.  Though  it  is  foolish 
to  claim  to  know  all  the  different  cases  in  which  cold 
and  heat,  generation  and  destruction  may  be  met  with, 
it  would  seem  easier  to  set  forth  a  complete  list  of 
such  actions  as  we  call  just  and  of  those  we  call  unjust. 
The  number  of  names  representing  these  actions  is 
limited  and  they  are  all  at  the  disposal  of  man,  for 
they  are  his  work.  This  possibility  of  comprehending 
the  entire  domain  of  moral  things  must,  above  all, 
have  been  recognised  in  a  nation  where  the  conditions 
of  human  life  were  relatively  simple,  where  the  totality 
of  human  duties  naturally  clustered  round  a  few  pre- 
cise, concordant  ideas,  and  there  was  entire  ignorance  of 
those  conflicts  between  the  individual  and  society,  con- 
science and  public  interests,  family  and  country,  country 
and  humanity,  physical  comfort  and  lofty  culture,  that 
have  introduced  inextricable  confusion  into  the  moral 
life  of  modern  nations. 

The  logical  method  of  Socrates  is  limited  to  in- 
duction and  definition  as  thus  understood.  Aristotle 
finds  fault  with  this  dialectic,  which  is  carried  on 
exclusively  by  a  process  of  questioning,  because  it  pins 
its  faith  to  common  opinion,  and  goes  no  farther  than 
probability.  His  appeal  is  to  special,  direct  intuition, 
the  indispensable  condition  of  a  complete,  infallible 
demonstration.  Aristotle's  reproach  is  comprehensible, 
if  our  object  is  to  go  back  to  the  first  principles  of  all 
things.  But  if  we  have  only  to  find  in  human  nature 
a  rule  for  human  judgment  and  conduct,  to  discover  and 
set  forth  the  principles  applied  by  human  reason  when 
tranquil  and  free  from  routine  and  passion,  with  the 
object  of  discovering  in  these  principles,  which  are  now 
objects  of  clear  consciousness,  a  weapon  against  routine 
and  passion  themselves  ;  in  a  word,  if  we  have  to  set 


SOCRATES  49 

man  free  by  enabling  him  to  know  mankind,  we  can 
understand  why  Socrates  contented  himself  with  the 
observation  of  human  phenomena,  and  made  no  attempt 
to  pierce,  by  metaphysical  intuition,  into  the  mysteries 
of  absolute  thought. 

IV 

Thus  we  see  that  the  nature  and  import  of  the 
Socratic  method  are  in  exact  proportion  to  the  object 
Socrates  had  in  view,  which  was  nothing  less  than  the 
constitution  of  ethics  as  a  science.  Conversely,  the 
concrete  doctrine  of  Socrates,  his  conclusions  on  things 
and  on  man  himself  are  exactly  what  might  have  been 
expected  from  the  use  of  such  a  method.  Matter 
responds  to  form  as  form  responds  to  matter. 

It  may  seem,  if  we  cast  a  general  glance  at  the 
teaching  of  Socrates,  that  the  science  of  which  it  con- 
sists does  indeed  go  beyond  the  limits  marked  out  by 
his  method,  and,  in  a  sense,  includes  not  only  human 
but  also  physical  and  divine  things. 

Is  not  his  reason  for  throwing  overboard  the 
mechanistic  physics  of  the  ancient  philosophers,  that  he 
wishes  to  substitute  therefor  a  teleological  system  of 
physics  ? l  Though  he  condemns  cosmological  theology, 
the  investigation  into  the  way  in  which  the  gods 
formed  the  universe,  does  he  not  extol  what  may  be 
called  moral  theology  in  his  endeavour  to  demonstrate 
the  existence  of  a  divine  intelligence  and  providence  ? 
The  considerable  importance  given,  in  the  Memorabilia, 
to  speculations  of  this  kind,  the  originality  of  Socrates' 
views  on  these  matters,  have  induced  several  critics 
to  regard  them  not  only  as  significant  parts  of  his 

1  Mem.  i.  4,  iv.  3. 


50  STUDIES  IN  PHILOSOPHY 

philosophy,  but  even  as  its  very  centre  and  ground- 
work. Thus,  to  Fouillee,  Socrates  is  essentially  the 
promoter  of  a  system  of  teleological  metaphysics, 
whereas  to  Franck1  he  is  above  all  else  a  theological 
philosopher. 

But  in  order  to  discover  if  teleology  and  moral 
theology  form  an  integral  part  of  the  object  of  science 
according  to  Socrates,  it  is  not  sufficient  to  examine  and 
see  whether  Socrates  advanced  profound  ideas  on  these 
subjects  or  not.  We  must  also  ask  ourselves  what 
relation  these  ideas  bear  to  the  fundamental  principles 
of  his  philosophy. 

Now,  one  can,  it  would  appear,  divide  the  teleo- 
logical and  the  theological  ideas  of  Socrates  into  two 
parts  ;  the  one  overstepping  the  limits  of  ethics,  though 
at  the  same  time  offered  us  as  the  fruit  of  supernatural 
inspiration  superior  to  science  ;  the  other,  of  a  more 
scientific  nature,  though  connected  with  ethics  as  its 
source  and  raison  d'etre.  When  Socrates  speaks  of  his 
daemonic  sign  and  of  the  power  it  sometimes  affords 
him  of  foreseeing  the  future  ; 2  when  he  speaks  of  the 
divinity  that  is  not  far  from  each  one  of  us  and  is 
ready  to  utter  a  warning  call  to  the  man  who  listens  in 
silence  ;  when  he  declares  that  to  fear  death  is  to  believe 
oneself  wise  without  really  being  so,  for  it  is  to  believe 
that  one  knows  what  one  does  not  know,3  he  is  evi- 
dently speaking  of  those  things  which,  as  they  are 
beyond  our  power  to  control,  are  also  beyond  the  reach 
of  our  science.4 

When,  on  the  contrary,  he  deals  with  physical  and 
divine  things  in  a  scientific  method,  we  see  him  pre- 
occupied about  considering  things,  not  in  themselves, 

1  Journal  ties  Savants,  October  1881. 
2  Mem.  i.  i.  3-5.  3  Apol.  29  A.  *  Mem.  i.  i.  9. 


SOCRATES  51 

but  from  without  and  with  reference  to  man.  Thus, 
he  constantly  tends  to  substitute  for  the  gods  the 
daemons,  who  are  nearer  to  ourselves,  and  for  the 
daemons  the  mere  daemonic  phenomena  or  visible  signs 
of  the  gods,  perceived  directly  by  man.1  He  believes 
that  we  cannot  see  the  gods  ;  that  we  see  nothing  but 
their  manifestations  to  us.2  The  order  and  harmony 
the  gods  have  introduced  into  things  consists  in  the 
appropriation  of  these  things  to  our  needs.3  In  this 
way,  physical  or  teleological  objects  are  brought  within 
the  compass  of  moral  and  human  ones. 

These  conjectures  on  the  adaptation  of  the  outer 
nature  to  the  needs  of  man,  besides  the  fact  that,  in  the 
case  of  Socrates,  they  spring  naturally  from  a  very 
sincere  and  deep  religious  sentiment,  are  called  for,  or 
required,  by  his  ethical  doctrine,  in  accordance  with 
which  the  happiness  of  man  depends  on  himself,  on 
nothing  whatever  but  self-knowledge.  Since,  in  spite 
of  his  efforts  to  suffice  unto  himself,  man  cannot  free 
himself  from  physical  nature,  he.  must  admit,  if  he 
claims  to  be  good  and  happy  without  occupying  himself 
with  externals,  that  the  gods  are  occupied  with  them  on 
his  behalf,  and  control  them  so  as  to  meet  his  needs. 
Teleology  and  the  doctrine  of  providence  were  the 
necessary  postulates  of  Socratic  morals. 

This  very  role  shows  us  that  they  are  complementary, 
not  essential  parts,  of  the  philosophy  of  Socrates. 

The  proper  object  of  this  philosophy,  not  only  in 
theory  but  in  fact,  is  the  one  that  the  Sophists  had 
brought  into  credit ;  that  is,  art,  or  practical  skill, 
understood,  however,  in  an  original  manner,  which  we 
have  now  thoroughly  to  investigate. 

Art,  in  the  mind  of  Socrates,  is  not  the  search  after 

1  Apcl.  27  B,  E.  2  Mem.  iv.  3.  13.  3  Ibid.  iv.  3.  i,  4. 


52  STUDIES  IN  PHILOSOPHY 

absolute  good,  the  power  to  regulate  our  actions  by 
the  whole  of  the  consequences  which  must  result  there- 
from, so  as  to  perform  only  those  whose  consequences, 
even  the  most  far-reaching,  are  in  conformity  with  our 
wishes.  The  gods  have  reserved  to  themselves  the 
knowledge  of  the  final  result  of  our  actions.  Does  the 
man  who  plants  an  orchard  know  who  will  gather  the 
fruit  thereof?  Does  he  who  builds  a  house  know  who 
will  dwell  in  it  ? 1 

On  the  other  hand,  however,  art  worthy  of  the 
name  resembles  no  special  profession  such  as  that  of 
the  carpenter,  the  shoemaker,  or  the  armourer.  These 
men  have  in  mind  the  realisation  of  some  particular 
material  object ;  whereas  art  pursues  a  general,  im- 
material end,  viz.  the  happiness  and  good  of  man. 
This  is  what  the  Sophists  had  already  taught,  and 
rightly  taught.  But  though  they  had  the  idea  of  what 
may  be  called  the  moral  end,  they  were  mistaken  as  to 
the  manner  of  attaining  to  it.  They  imagined  this 
could  be  effected  by  regular  practice,  similar  to  that 
which  proves  successful  in  special  professions.  But, 
even  in  these  latter,  regularity  or  routine  is  far  from 
being  sufficient.  Every  good  artisan  possesses  not  only 
the  practice  but  also  the  science  of  his  trade,  in  so  far 
as  his  trade  is  capable  of  being  an  object  of  science.  A 
well-drawn  analogy  will  lead  us  to  think  that  moral  art 
also  must  be  a  science,  according  to  the  acceptation  of 
this  word  in  the  moral  domain. 

To  sum  up,  moral  art,  occupying  a  position  midway 
between  religion  and  the  special  professions,  art  which 
has  for  its  aim  the  present  good  and  happiness  of  man, 
and  for  its  province  the  science  of  human  things  :  such 
is  the  object  of  Socrates'  reflections. 

1  Mem.  i.  i.  8. 


SOCRATES  53 

'It  is  this  object  that  exactly  answers  to  his  idea  of 
science.  Science  tries  to  discover  that  which  is  general, 
and  which  forms  the  material  for  the  discourse  of  men  ; 
i.e.  the  categories  in  which  they  place  particular  things. 
But  is  it  not  in  moral  things  that  we  find  a  perfect 
instance  of  that  relation  of  genus  to  species,  of  principle 
to  application,  of  latent  to  manifest  knowledge,  which 
such  an  idea  of  science  implies  ?  Moral  things  do  not 
contain  within  themselves  the  absolute,  the  one  in  itself, 
the  supreme  principle  of  being  and  knowing  ;  but  then, 
Socratic  science  does  not  aim  so  high  as  that.  On  the 
other  hand,  however,  and  in  contradistinction  to  the 
opinion  of  the  Sophists,  in  human  nature  itself  there 
are  certainly  fixed,  solid  points,  which  enable  one  to 
gain  a  satisfactory  science  of  the  general. 

Moreover,  is  it  not  moral  things  that  form  the  usual 
matter  of  human  discourse  ?  Is  it  not  on  these  questions 
that  each  man  has  acquired  experience  and  is  capable  of 
advancing  an  opinion  worthy  of  consideration?  If  so, 
then  it  is  along  this  line  that  there  will  be  the  best 
chance  of  success  for  a  science  that  seeks  its  various 
elements  in  the  discourse  of  men,  even  of  the  humblest 
and  most  ignorant. 

When  considered  with  a  view  to  the  knowledge  of 
moral  principles  the  Socratic  method  thus  reacts  on  the 
conception  of  moral  things  themselves.  In  the  light  of 
the  idea  of  science,  Socrates  found  in  human  nature  that 
substratum  of  common  and  invariable  notions  that  had 
escaped  the  notice  of  the  Sophists  ;  thereupon,  every- 
thing human  was  invested  with  new  dignity  in  the  mind 
of  the  philosopher. 

This  reaction  of  method  on  object  appears  no  less 
clearly  in  the  details  of  Socratic  morals. 

Here  two  essential  parts  may  be  distinguished  :  ist, 


54  STUDIES  IN  PHILOSOPHY 

the  general  principle  :  all  virtues  are  sciences  ; l  2nd, 
the  deduction  of  the  virtues,  which  deduction  is  supplied 
by  this  principle. 

In  what  sense  did  Socrates  claim  that  all  virtues  are 
sciences  ? 

According  to  Zeller,2  the  science  here  in  question  is 
evidently  science  in  general,  the  science  of  the  nature  of 
things.  But  in  none  of  the  texts  dealing  with  our 
question  do  we  find  this  abstract  expression  :  science. 
They  all  state  more  or  less  explicitly :  the  virtues  are 
sciences.8 

Consequently,  virtue  is  not  identified  with  science  in 
general,  but  with  some  particular  science.  Now,  what 
is  this  science  ? 

Fouillee4  says  that  the  science  of  which  Socrates 
speaks  is  probably  the  science  of  good  in  itself,  i.e.  the 
science  of  the  real  and  absolute  worth  of  things. 

Such  an  object,  however,  would  go  beyond  the  end 
aimed  at.  When,  says  Socrates,  one  is  thinking  of 
becoming  a  good  shoemaker,  or  pilot  or  musician,  the 
science  each  man  regards  as  indispensable  is  that  of 
shoemaking,  ship-management  or  music:  that  special 
science  alone,  in  each  category,  makes  the  man  com- 
petent. Now,  competency  is  also  what  Socrates  extols 
in  moral  matters.  The  analogy  he  is  constantly  drawing 
between  the  special  professions  and  the  practice  of 
virtue  shows  that  he  places  the  condition  of  this  new 
competency  not  in  a  universal  and  necessarily  vague 
science,  but  in  the  science  of  virtue  itself.  Though 
Socrates  does  not  agree  with  the  Sophists,  who  made  too 
close  a  comparison  between  moral  art  and  the  mechanical 

1  Aristot.  Eth.  Nic.  vi.  13.  1144  b  28.  2  ii.  (3rd  edit.),  93,  117. 

3  Mem.  iii.  9.  5,  iv.  2.  22,  iv.  6-7  ;  Aristot.  Eth.  Nic.  vi.  13.  1144  b  17. 
*  La  Phil,  de  Socrate,  i.  177,  281,  285. 


SOCRATES  55 

arts,  he  yet  does  not  go  so  far  as  to  abolish  all  analogy 
between  these  latter  arts  and  the  former.  Virtue  is  still 
a  special,  determinate  art ;  just  men  as  well  as  artisans 
have  their  own  distinctive  work.1 

Science,  thus  determined,  i.e.  the  special  science  of 
virtue,  is,  according  to  Socrates,  the  very  definition  or 
essence  of  science.  Socrates  means  thereby  that  it  is 
its  necessary  and  all-sufficient  condition. 

It  is  the  necessary  condition  of  virtue.  If  com- 
petency is  necessary  in  mechanical  arts,  how  can  it  be 
superfluous  in  an  art  that  is  surely  more  delicate  and 
complicated,  since  it  has  to  work  upon  things  that  are 
invisible,  accessible  to  the  understanding  alone  ?  The 
masses  are  wrong  when  they  think  that  nature  in  moral 
matters  is  all-sufficient.  In  vain  did  the  Sophists  substi- 
tute practice  for  nature.  He  who  is  ignorant  of  the 
definition  of  good  may,  by  a  happy  chance,  sometimes 
meet  with  it,  but  he  will  never  be  certain  that  he  has 
not  altogether  passed  it  by.  He  will  even  run  the  risk 
of  taking  evil  for  good,  and  vice  versa.  For  instance,  if 
one  does  not  possess  a  definition  of  justice,  one  may  regard 
it  invariably  as  unjust  to  deceive  and  injure  others, 
whereas  it  is  just  to  deceive  the  enemies  of  the  State,  and 
to  reduce  an  unjust  nation  to  a  state  of  subjection.2 
Again,  if  one  is  without  this  definition,  one  will  stop  to 
examine  such  a  question  as  the  following  :  "  Who  is  the 
more  unjust  :  the  man  who  wittingly  deceives,  or  the 
man  who  unwittingly  deceives  ?  "  One  will  be  astonished 
at  finding  arguments  in  support  of  both  positions, 
whereas,  at  bottom,  the  question  is  an  absurd  one,  since 
the  terms  "  unjust  "  and  "  unwittingly  "  immediately 
exclude  each  other.  Science  renders  certain  actions  good, 
which,  without  it,  would  be  indifferent  or  even  evil ;  for 

1  Mem.  iv.  2.  12.  2  Ibid.  iv.  2.  14-15.  3  Ibid.  iv.  2.  19. 


56  STUDIES  IN  PHILOSOPHY 

instance,  the  use  of  money.  By  science  and  science 
alone  does  skill  in  speech  and  action  become  a  virtue  : 
this  skill,  if  left  to  itself,  might  readily  cause  men  to 
become  more  unjust  and  maleficent  than  nature  made 
them.1 

Science  is  not  only  necessary,  it  is  all-sufficient  for  the 
engendering  of  virtue.  This  doctrine  is  what  may  be 
called  the  Socratic  paradox.  Perhaps  the  paradox  is  not 
so  pronounced  as  it  at  first  seems. 

It  would,  indeed,  be  strange  for  Socrates  to  attribute 
such  efficacy  to  science,  if  we  were  dealing  with  a 
purely  theoretical  science,  or  even  with  the  science  of 
good  in  itself  and  of  the  rational  value  of  things.  At 
the  outset  the  objection  will  be  made  that  such  know- 
ledge supplies  a  law  to  the  intelligence,  but  that  it  does 
not  determine  the  will. 

The  science,  however,  of  which  Socrates  speaks,  is 
distinctly  the  science  of  the  suitability  and  utility  of 
things  from  the  human  point  of  view  ;  it  is  the  know- 
ledge of  the  relation  that  exists  between  things,  and  the 
end  that  man  follows  of  his  own  accord,  naturally  and 
of  necessity.  u  In  order  to  be  obeyed  by  my  sub- 
ordinates," said  a  cavalry  officer  to  Socrates,2  "  will  it 
suffice  if  I  show  them  that  I  am  their  superior  ?  "  "  Yes," 
was  the  answer,  "  provided  you  prove  that  obedience 
to  you  is  safer  and  more  beautiful  for  them  than 
the  contrary  (/caXXtoi/  re  ical  awT'rjpKOTepov  aurot?)." 
Socrates  reasons  in  this  fashion  :  it  is  acknowledged  that 
men  invariably  do  what  they  believe  they  ought  to  do, 
i.e.  what  they  look  upon  as  most  profitable  to  them- 
selves. If,  then,  it  is  demonstrated  to  them  that  virtue 
is  most  profitable,  they  will  infallibly  practise  virtue. 
In  a  word,  our  philosopher  transfers  to  the  science  of 

1  Mem.  iv.  3.  i.  2  Ibid.  iii.  3.  10. 


SOCRATES  57 

the  good  the  practical  efficacy  he  usually  notes  in  the 
mere  opinion  of  the  good.  More  than  this  :  the  science 
of  the  good  seems  to  him  as  though  it  must  be  even 
more  efficacious  in  determining  the  human  will  than 
the  mere  opinion  of  the  good  can  be,  because  science  is 
immovable,  whereas  opinion  is  at  the  mercy  of  circum- 
stances. 

Fouillee  l  considers  that  the  Socratic  paradox  consists 
essentially  in  the  negation  of  free-will.  It  rather  con- 
sists in  the  claim  to  demonstrate  that  virtue  is  always 
that  which  is  most  advantageous  to  man. 

(As  regards  free-will,  Socrates  neglects  to  take  it 
into  consideration  rather  than  denies  it)  And,  indeed, 
free-will  is  almost  useless  in  a  doctrine  which  only 
requires  man  to  decide  in  the  way  he  considers  most 
beautiful  and  advantageous.  This  method  of  determi- 
nation, according  to  Socrates,  is  that  of  the  masses  ;  it 
is  quite  spontaneous,  and  does  not  imply  the  conscious- 
ness of  being  able  to  determine  in  favour  of  the  opposite 
course  of  action. 

True,  the  objection  may  be  advanced  that,  for  a  man 
to  regard  as  insufficient  the  mere  opinion  of  good,  and 
try  to  discover  the  constituents  of  real  good,  he  must 
make  an  effort  which  involves  the  intervention  of 
free-will. 

Socrates  is  far  from  denying  the  necessity  of  such  an 
effort  ;  though  he  connects  it  with  self-control  and 
temperance,  which  latter  is  itself,  in  his  mind,  a  science, 
and  the  most  important  of  them  all.2  The  obligation 
of  self-control  and  temperance  is  demonstrated  in  the 
same  fashion  as  that  of  all  the  other  virtues  :  by  its 
useful  effects.  It  by  no  means  follows  that,  in  acquir- 
ing this  virtue,  the  first  condition  of  all  the  rest,  free-will 

1  La  Phil,  de  Socr.  i.  173.  2  Mem.  i.  5.  4. 


58  STUDIES  IN  PHILOSOPHY 

has  no  part  to  play.  The  negation  of  free-will  might 
be  deduced  from  the  doctrine  if  Socrates  distinctly  set 
up  self-control  (eyre  pare  to)  between  science  (cro^ta)  and 
temperance  (a-atfypoa-vvrj)  as  being  a  consequence  of  the 
former  and  nothing  more,  as  Fouillee l  states.  Socrates, 
however,  regards  self-control  as  both  a  condition  and  a 
result  of  science.  "  Do  you  not  think,"  he  says,  "  that 
lack  of  self-control  (arcpaa-ia)  turns  men  away  from 
science  (<ro<£ia),  the  greatest  of  all  things,  and  drives  them 
to  its  opposite  ?  "  "  Only  to  such  as  are  self-possessed," 
he  says  in  another  place,  "is  it  given  to  practise 
dialectic."  3  It  is,  therefore,  no  abstract  science,  but  a 
living  science,  action  and  knowledge  combined,  which 
is  the  root  of  virtue. 

Thereby  we  find  clearly  determined  the  relation 
Socrates  sets  up  between  science  and  practice.  He 
maintains  that  science  engenders  virtue  to  which  it  plays 
the  part  of  an  efficient  cause  ;  but  he  also  maintains 
that  the  search  for  science  has,  for  its  province,  the 
desire  to  attain  to  virtue,  and  thus  virtue  plays  the  part 
of  final  cause,  as  regards  science.  Science  is  both  cause 
and  means,  virtue  both  end  and  result.  Between  the  two 
terms  there  is  solidarity,  mutual  action.  It  must  be 
granted  that  such  a  relationship  raises  difficulties  for 
him  who  would  understand  it  thoroughly.  Socrates, 
however,  must  have  found  it  tolerably  clear,  at  a  period 
when  neither  the  efficient  nor  the  final  cause  had  yet 
been  studied  for  themselves  and  no  clear  line  of  demar- 
cation drawn  between  will  and  intelligence. 

Though  such  is  the  Socratic  doctrine  as  to  the 
relation  between  science  and  virtue,  Socrates,  doubtless 
explicitly,  went  beyond  the  stand-point  of  ordinary 

1  La  Phil,  de  Socr.  i.  173.  2  Mem.  ix.  5.  6. 

3  Ibid.  iv.  5.  n. 


SOCRATES  59 

morals  which  merely  sets  forth  isolated  precepts  without 
connecting  them  with  any  principle.  He  also  went 
beyond  that  of  the  ancient  sages,  as  well  as  of  the  great 
writers  of  his  time,  who  confined  themselves  to  deriving, 
direct  from  their  own  consciousness,  maxims  that  were 
frequently  profound,  without  attempting  to  demon- 
strate them  scientifically.  He  was  the  first  to  make 
science  an  integral  element  of  morals ;  the  first  to 
bring  action,  which  appears  as  individual,  within  the 
compass  of  true  knowledge,  which  is  universal. 

But  this  does  not  mean  that  he  applied  to  morals 
the  universal  idea  of  science,  and  not  merely  that  idea 
of  a  science  of  man  which  appears  as  the  term  of  his 
dialectic.  Where  can  Socrates  obtain  the  rational 
knowledge  of  good  and  virtue,  which  is  all  that  he  here 
means  by  science,  except  in  the  discourse  of  men,  that 
immediate  testimony  of  their  desires,  their  needs  and 
experience  ?  What  more  certain  method  of  giving  a 
practical  definition  of  things,  expressing  the  interest 
they  offer  to  man,  than  that  of  using  the  analogy  and 
induction  which  take  human  facts  themselves  for  their 
basis,  and  interpret  them  in  the  light  of  human 
reason  ?  Likewise,  what  science  will  have  most  chance 
of  acting  upon  the  will,  what  science  will  better 
merit  those  bold  words  of  praise  :  ovSev  lo-^vporepov 
ffrpovrjcrews,1  than  that  truly  living  science  which  Socratic 
maieutics  evolves  from  our  soul,  and  which  is,  at  bottom, 
only  the  consciousness  of  our  own  nature  ?  If  care  be 
taken,  the  details  of  the  doctrine  of  the  relations  that 
exist  between  virtue  and  science,  coincide,  step  by  step, 
with  the  details  of  dialectic  ;  in  such  fashion  that,  the 
latter  being  posited,  the  former  necessarily  follows. 

Dialectic,  sprung  from  the  general  and  still  vague 

1  Eth.  Eud.  vii.  13. 


60  STUDIES  IN  PHILOSOPHY 

idea  of  moral  science,  reacts  upon  this  idea  and 
determines  it.  Moral  science  is  but  dialectic  in  action. 

We  reach  a  similar  conclusion  if  we  examine  the 
second  part  of  Socratic  morals,  to  wit,  the  deduction  of 
the  virtues,  supplied  by  the  general  principle  of  morals. 

What  are  the  chief  maxims  of  this  science  of  good 
which  is  the  necessary  and  all-sufficient  condition  of 
virtue  ? 

In  this  connection  Socrates  distinguishes  between 
good  in  general  and  particular  good. 

Good  in  general  is  the  truly  useful  as  distinguished 
from  the  pleasant.1  The  whole  of  morals  consists  in 
distinguishing  what  distinctly  constitutes  our  own  good 
from  what  seems  to  do  so,  though  in  reality  giving  us 
only  passing  pleasure,  perhaps  even  loss.  Why  is  in- 
temperance evil  ?  Because,  says  Socrates,  it  turns  man 
aside  from  useful  things  (w^eXoOz/ra)  and  inclines  him 
towards  pleasant  things  (^Sea).2 

Though  Socrates  makes  a  broad  distinction  between 
what  is  good  in  appearance  and  what  is  good  in  reality, 
we  do  not  find  that  he  is  thinking  of  an  absolute  good, 
of  which  the  good  of  man  would  seem  to  be  only  one 
particular  manifestation.  He  appears  to  have  com- 
pletely identified  the  good  with  the  useful,3  and  the 
reason  he  recommends  the  acquisition  of  science,  the 
practice  of  justice,  soul-culture  and  the  attainment  of 
the  loftiest  virtues,  is  that  he  regards  them  as  useful  for 
man's  happiness.  Even  when  he  prefers  death  to 
shame,  the  reason  he  gives  is  that,  in  the  absence  of  the 
daemonic  sign  which  usually  warns  him  whenever  he  is 
about  to  do  something  destined  to  injure  him,  he  is 
convinced  death  will  do  him  no  harm.4 

1  Mem.  iv.  6.  8.  2  Ibid.  iv.  5.  6.  3  Ibid.  iv.  6.  8. 

4  Apol.  CC.  xxix.  sqq. 


SOCRATES  6 1 

Clearly,  this  doctrine,  in  the  Socratic  philosophy,  is 
the  reaction  of  form  on  matter.  Matter  was  first  the 
vague  idea  of  pleasure  and  well-being,  as  found  in  the 
reasonings  of  the  Sophists  concerning  the  goal  of  our 
actions.  Now,  science,  according  to  Socrates,  is  the 
search  after  the  general.  Therefore,  when  brought 
into  contact  with  the  idea  of  science,  the  idea  of  well- 
being  becomes  two-fold,  engendering,  on  the  one  hand, 
the  idea  of  pleasure,  pure  and  simple,  or  a  chance,  fleet- 
ing enjoyment,  incapable  of  becoming  an  object  of 
science,  and,  on  the  other  hand,  the  idea  of  true  utility 
and  happiness,  corresponding,  in  its  generality,  with  the 
conditions  of  dialectic.  True  utility  is  that  object,  at  once 
stable  and  human,  the  type  and  standard  of  which  each 
of  us  bears  within  himself  and  which  it  is  for  maieutics, 
induction  and  definition  to  find  out  and  determine. 

Now,  what  is  the  teaching  of  Socrates  regarding 
particular  good  ? 

Socrates  is  sometimes  represented  as  deducing  a 
priori  particular  good  from  the  idea  of  absolute  good,  and 
judging  custom  and  legality  in  the  name  of  reason  and 
justice.  This  is  by  no  means  his  method  of  procedure. 
Instead  of  criticizing  tradition  and  the  positive  law 
in  the  name  of  reason,  it  is  in  the  traditional  and  the 
positive  that  he  seeks  the  expression  of  the  rational. 
According  to  Socrates,  particular  good  consists  of  those 
things  that  men  are  agreed  in  regarding  as  good  : 
health  and  strength  of  body  and  soul,1  easy  domestic 
circumstances,2  useful  knowledge,3  family  and  friendly 
relations,4  civil  society  and  the  country's  prosperity,5 
good  repute,6  and,  speaking  generally,  skill  in  the 
management  of  life. 

1  Mem.  in.  12.  4,  6.  2  Ibid.  ii.  17.  3  Ibid.  iv.  2.  23-35. 

4  Ibid.  ii.  3.  19.  6  Ibid.  iii.  7.  9.  °  Ibid.  ii.  i.  31. 


62  STUDIES  IN  PHILOSOPHY 

Socrates  distinctly  identifies  justice  with  legality,  and 
piety  with  the  observance  of  the  religious  laws  of  one's 
country.  <f>r)fu  yap  eyo> l  TO  vofupov  Si/caiov  elvai  .  .  ., 
TO  avro  vofit/jiov  re  ical  Siiccuov  :  "  I  say  that  justice  con- 
sists in  the  observance  of  the  law  ;  that  the  just  and 
the  legal  are  both  the  same  thing."  After  all,  what  is 
law  ?  It  is  what  the  citizens,  gathered  together,  have 
decreed,  in  writing,  as  something  that  must  either  be 
done  or  avoided.2  Piety  itself  is  nothing  else  than  the 
knowledge  and  practice  of  those  laws  of  one's  country 
which  refer  to  the  gods  :  TO,  frepl  rov<s  0eou?  vofupa.3 

True,  Socrates  also  speaks  of  divine,  unwritten 
laws.4  By  these  he  means  not  laws  of  an  abstract, 
universal  nature,  but  laws  that  are  quite  as  positive 
(yofiL^ov)  as  human  laws.  These  laws  are  written  in 
the  soul,  though  they  may  not  be  found  on  material 
tablets.  When  Socrates  wishes  to  cite  examples  thereof, 
he  speaks  of  the  recommendation  to  honour  the  gods, 
the  prohibition  from  marrying  one's  own  children : 
maxims  that  partake  of  the  nature  of  particular  and 
positive  statutes.  In  his  own  words  :  "  In  the  divine 
as  in  the  human  order  of  things  justice  is  identical  with 
legality."6 

The  doctrine  of  Socrates  regarding  particular  good 
is,  however,  not  limited  to  this.  To  common,  tradi- 
tional morals  as  matter  he  connects  the  idea  of  science 
as  form ;  and,  by  contact  with  this  new  element, 
morals  is  completely  transformed  without  this  appearing, 
externally,  to  be  so. 

The  first  function  of  science  is  to  justify,  to  deduce 
what  common  sense  and  tradition  offer  to  us  only  as 
independent  facts. 

1  Mem.  iv.  4.  12.  2  Ibid.  iv.  4.  13.  3  Ibid.  iv.  6.  4. 

4  Ibid.  iv.  4.  19.  6  Ibid.  iv.  4.  25. 


SOCRATES  63 

This  deduction  is  effected  by  demonstrating  that  all 
actions  which  common  sense  and  tradition  prescribe  to 
us  are  calculated  to  procure  advantages  for  us,  whereas 
the  opposite  of  these  actions  must  sooner  or  later  do 
us  harm.  For  instance,  temperance  is  a  good  thing, 
because  it  is  the  condition  of  pleasure,  helps  us  to  bear 
privation,  and  makes  us  esteemed  by  our  fellow-beings. 
If  a  general,  a  tutor,  or  a  steward  is  wanted,  the 
temperate  man,  not  the  intemperate,  is  the  one  who 
will  be  chosen.1  The  observance  of  civil  laws  is  a  good 
thing,  for,  under  all  circumstances,  those  who  observe 
the  laws  are  the  ones  best  treated  in  the  State  ;  in 
public  or  private  life  it  is  they  who  inspire  most  confi- 
dence.2 The  same  reasoning  holds  regarding  unwritten 
laws.  It  is  good  to  observe  them,  for  the  man  who 
violates  them  is  punished  :  thus,  parents  who  marry 
their  own  children  have  misshapen  offspring.3  In  this 
sense  Socrates  affirms  that  what  is  legal  is  likewise 
just.  A  law  is  just,  in  so  far  as  its  observance  procures 
advantages,  whilst  its  violation  has  disastrous  conse- 
quences.4 

Science  thus  deduces  and  justifies  the  established  laws. 
Nor  is  this  all.  As  the  wise  man,  by  means  of  science, 
searches  into  and  understands  the  rational  value  of 
tradition  and  legality,  and  thus  learns  to  conform 
with  the  laws  of  his  country,  not  blindly,  as  do  the 
masses,  but  by  reflection  and  reason,  he  regards  action 
inspired  by  science  as  superior  to  that  emanating  from 
instinct  or  custom.  Science  no  longer  seems  to  him 
merely  to  confirm  the  positive  rules  of  morals  :  it 
becomes  itself  an  indispensable  condition  of  virtue,  the 
root  of  all  virtue  :  virtuepar  excellence.  To  act  under  the 

1  Mem.  iv.  5.  2  Ibid.  iv.  4.  17.  3  Ibid.  iv.  4.  19  sqq. 

4  Ibid.  iv.  4.  25. 


64  STUDIES  IN  PHILOSOPHY 

influence  of  nature  alone,  like  prophets  and  soothsayers,1 
means  not  only  exposing  oneself  to  continual  failure  in 
some  direction  or  other,  it  likewise  means  having 
nothing  but  the  mask  of  art  or  virtue.  He  alone  who 
is  virtuous  through  science  (o-o^ta)  truly  merits  the  title 
of  virtuous.  Nothing  blind  or  inconsiderate  could  be 
really  good  :  on  the  other  hand,  once  a  man  acquires 
self-possession,  his  actions  are  of  necessity  good.  And 
so  Socrates,  when  accused,  refuses  to  move  his 
listeners  to  compassion,  because  compassion  is  a  blind 
sentiment.2  On  the  other  hand,  he  declares  that,  as  he 
has  never,  willingly  and  knowingly  (e/e&>i>),  done  wrong, 
he  is  certain  he  has  never  really  done  wrong  at  all.3 

The  mental  state  which  immediately  corresponds  to 
science,  because  it  is  both  its  condition  and  first  result, 
is  self-control  (ejicpdreia)  or  freedom  (e^evOepLa}.  f  Self- 
control  thus  becomes  the  first  of  all  virtues^  the  one 
whose  possession  is  both  necessary  and  all-sufficient  for 
the  performance  of  good  under  all  circumstances.  To 
know  how  he  ought  to  act,  the  wise  man  has  definitively 
only  one  question  to  ask  himself:  is  this  particular 
line  of  conduct  seemly  in  a  free  man,  or  not  ? 

On  several  puzzling  occasions,  this  doctrine  explains 
^jhe  line  of  conduct  adopted  by  Socrates.]  The  reason 
he  refused  to  accept  money  from  his  listeners,  was  not 
liberality  on  his  part  or  the  fear  of  slanderers,  it  was 
because  he  considered  that  to  receive  money  from 
another  was  equivalent  to  acknowledging  that  man  to 
be  one's  master.5  The  reason  he  extolled  manual  work 
was  not  from  sympathy  with  the  occupations  of  the 
humble,  but  because  he  saw  in  such  work  a  source  of 
independence  and  easy  circumstances  from  a  material 

1  Apol.  22  B.  2  Ibid.  35  B.  3  Ibid.  37  A. 

4  Mem.  i.  5.  4.  5  Ibid.  i.  5.  6. 


SOCRATES  65 

point  of  view.1  If  it  is  true  that,  on  one  occasion,  he 
walked  barefoot  on  the  ice,  and  on  another,  remained 
standing  for  a  whole  day  and  night  in  the  self-same 
spot,2  this  was  not  done  in  a  spirit  of  folly  or  boasting, 
though  it  might  have  been  an  instance  of  mystic  con- 
templation ;  perhaps,  too,  these  experiments  were  made 
for  the  purpose  of  seeing  how  far  his  independence  of 
the  external  world  could  be  carried.  Again,  the  reason 
he  endured  the  peevish  temper  of  Xanthippe  his  wife, 
was  not  from  resignation  or  good  temper,  it  was  be- 
cause his  wife  offered  him  a  splendid  opportunity  for 
practising  self-control.  The  reason  he  delighted  in 
banquets  and  feasts,  conversed  in  perfect  freedom  with 
Tljeodota,  the  courtezan,8  considered  it  quite  right  that, 
in  the  relations  between  the  sexes,  one  should  obey 
the  promptings  of  nature,  provided  one  is  caused  no 
embarrassment  thereby,4  acknowledged  so  strange  and 
dangerous  a  kind  of  love  between  young  men  ;  was  to 
be  found  in  the  fact  that  he  saw  nothing  in  all  this, 
irreconcilable  with  self-possession,  nothing  but  a  witness 
to  or  an  instrument  of  freedom. 

In  this  dignified  conception  of  life,  the  positive  and 
traditional  rules  of  morals  are  by  no  means  neglected  ; 
but  from  the  role  of  principles  they  descend  to  that  of 
matter  or  external  conditions.  The  wise  man  has •  self- 
possession,  and  that  is  enough  for  him  ;  after  all,  he 
speaks  and  acts  like  the  rest  of  mankind.  He  is  con- 
scious of  his  freedom  in  the  very  act  of  observing  the 
laws  and  customs  of  his  country.  These  laws  govern 
his  outward  actions,  just  as  science  governs  his  inner 
disposition,  and  harmony  between  the  two  disciplines 
is  all  the  better  established  in  that  self-possession,  the 

1  Mem.  ii.  7.  4.  2  Plat.  Banquet,  c.  xxxv-xxxvi. 

3  Mem.  iii.  n.  4  Ibid.  i.  3.  14. 

F 


66  STUDIES  IN  PHILOSOPHY 

supreme  command  of  the  inner  law,  becomes  reconciled 
of  itself  with  the  most  multiple  and  diverse  modes  of 
outer  action.  Besides,  it  is  evident  that  amongst  the 
various  positive  disciplines  conceivable,  the  wise  man 
will  decide  for  that  of  his  own  nation.  What,  indeed, 
could  be  more  favourable  to  the  inner  freedom  after 
which  he  aspires,  than  to  live  in  harmony  with  those 
around  him  ?  What,  on  the  other  hand,  could  be  more 
prejudicial  to  quiet  and  self-possession  than  that  dis- 
turbing, harassing  conflict  with  things  which  makes  us 
lose  control  of  ourselves  ? 

The  whole  of  this  doctrine  was  summed  up  in  two 
famous  aphorisms  :  "  Virtue  is  one  in  JUself."  and 
"  Virtue  can  be  taught." 

o 

By  the  oneness  of  virtue,  Socrates  did  not  mean, 
after  the  fashion  of  the  mystics,  the  elimination  of  all 
particular  virtues  in  favour  of  some  transcendent 
perfection.  He  simply  meant  that  all  virtues  have  one 
common  root,  to  wit,  the  science  of  good,  as  he  under- 
stood it.  To  the  wise  man,  the  diversity  of  virtues 
held  in  honour  amongst  men  is  nothing  but  the 
multiplicity  of  the  aspects  shown  forth  by  the  one 
sovereign  virtue,  according  to  the  various  objects  to 
which  it  applies.  Thus,  virtue  was  neither  absolutely 
one  nor  altogether  multiple  :  it  was  unity  in  multiplicity, 
self-possession  and  the  science  of  good  realised  in  the 
virtues  sanctioned  by  tradition. 

Socrates  claimed  that  virtue  is  taught,  but  he  by 
no  means  meant  thereby  that  it  is  taught  by  purely 
theoretical  teaching  or  speculation,  like  the  doctrines 
of  the  physiologists.  Nor,  in  his  opinion,  is  it  taught 
by  practice  alone,  as  the  Sophists  had  imagined. 
Virtue  is  taught,  said  Socrates,  by  instruction  combined 
with  exercise  or  practice  (fidd^cn^  and  /ueXen?).  All  the 


SOCRATES  67 

texts  dealing  with  this  doctrine J  clearly  show  that 
Socrates  invariably  employs  these  two  words  together. 
This  is  the  natural  outcome  of  the  intimate  union  of  a 
theoretical  and  a  practical  element  in  the  very  science 
which  is  the  principle  of  wisdom. 

If  such  is  the  doctrine  of  Socrates  upon  particular 
good,  it  bears  the  impress  of  the  Socratic  dialectic,  as 
does  his  doctrine  of  the  good  in  general.  Scrupulous 
respect  for  tradition  and  for  the  laws  of  one's  country 
is  in  conformity  with  this  method,  which  places  the 
starting-point  of  knowledge  not  in  pure  reason,  but 
rather  in  general  ideas.  The  philosopher,  without 
contradicting  himself,  could  not  turn  against  these 
ideas  the  very  principles  he  extracted  from  them. 

On  the  other  hand,  the  dialectician  must  go  back  as 
far  as  possible  into  antiquity  when  seeking  the  general 
principles  implied  in  human  discourse.  Now,  in  the 
accomplishment  of  this  task,  Socrates  comes  to  regard 
the  essence  of  virtue  as  existing  not  in  external  acts 
that  conform  with  legality  but  in  self-possession  and 
the  science  of  good,  which  form  the  common,  permanent 
substratum  of  these  acts.  Self-possession  and  the  science 
of  good  bear  the  same  relation  to  good  actions  that 
definition  does  to  the  class  of  objects  to  be  defined. 

In  short,  the  special  sense  in  which  Socrates  teaches 
that  virtue  is  one  and  can  be  taught,  exactly  answers  to 
the  nature  of  the  general  in  Socratic  dialectic.  This 
"  general,"  indeed,  has  by  no  means  a  distinct  existence, 
it  is  only  what  is  continually  assumed  in  human 
discourse  ;  and,  since  it  is  drawn  from  the  common 
ideas  relating  to  social  and  private  life,  it  possesses, 
of  necessity,  both  a  practical  and  a  theoretical  nature. 

1  Mem.  m.  9.  2  ;  iv.  i.  3  ;  i.  2.  19.     Cf.  Laches,  190  E. 


68  STUDIES  IN  PHILOSOPHY 


Thus  do  Socratic  dialectic  and  ethic  interpenetrate 
and  determine  each  other.  yThe  idea  of  moral  things 
as  an  object  of  science,  leads  Socrates  to  invent  a 
scientific  method  applicable  to  such  an  object.)  On  the 
other  hand,  the  use  of  this  method  reacts  on  the  object 
itself,  giving  it  a  new  aspect.  From  the  elaboration  of 
the  form  with  a  view  to  the  object  arose  the  theory  of 
practical  induction  and  definition  ;  from  the  elaboration 
of  the  object  by  means  of  the  form  arose  the  doctrine 
of  virtue,  dwelling  in  the  free  and  deliberate  observance 
of  positive  laws  and  maxims. 

\  The  expression  "  moral  science  "  thus  would  seem 
to  characterize  the  invention  of  Socrates  exactly  and 
fully,  provided  we  mean  by  these  words,  not  morals 
founded  on  the  science  of  things  in  general,  but  rather 
an  effort  of  the  human  mind  to  build  up  a  science 
without  leaving  the  circle  of  moral  facts  themselves, 
and  confining  itself  to  the  fertilisation  of  moral  ex- 
perience by  an  appropriate  mode  of  reflection. 

Here,  indeed,  is  the  centre  of  Socrates'  doctrine,  the 
principal  mobile  of  his  thought. 

It  is  because  he  institutes  a  new  order  of  investiga- 
tion that  he  rejects  and  dismisses  the  investigations  of 
his  predecessors.  All  innovators  possess  this  disdain  of 
the  past  :  it  forms  part  of  their  faith  in  their  own 
mission. 

Because  his  conception  of  science  is  exclusively 
calculated  with  a  view  to  the  reasoned  knowledge  of 
human  things,  he  says,  along  with  Protagoras,  that 
science  does  not  attain  to  things  divine.  Stricter  in 
his  reasoning,  however,  he  has  not  the  impertinence  to 


SOCRATES  69 

suppress  a  given  object,  under  the  'pretext  that  our 
intelligence  cannot  grasp  it  :  on  the  contrary,  he 
acknowledges  the  limits  of  our  faculties  as  soon  as  he 
discovers  their  powers  ;  and,  faithful  to  his  country's 
religion,  he  trusts  to  the  gods  in  regard  to  everything 
beyond  the  reach  of  human  understanding. 

The  belief  of  Socrates  in  an  Apollonian  mission  and 
in  the  supernatural  warnings  of  a  protecting  divinity 
can  be  perfectly  reconciled  with  this  doctrine,  which 
both  respects  the  domain  of  the  gods  and  takes 
possession  of  that  of  men. 

flThat  it  was  the  ambition  of  Socrates  to  restore  the 
political  fortunes  of  his  city  by  a  moral  reformVwas 
only  natural  and  legitimate  for  one  able  to  distinguish 
the  principles  of  virtue  and  of  success  in  human  things, 
and  whose  very  philosophy  gave  him  a  fresh  motive  for 
gratitude  and  attachment  to  his  country. 

Finally,  that  Socrates  submitted  to  death  rather 
than  renounce  the  testing  of  the  Athenians  for  the 
purpose  of  convincing  them  of  their  ignorance,  is,  as 
he  says  himself,  the  logical  consequence  of  a  doctrine 
which  looks  upon  self-examination  as  the  principle  and 
condition  of  all  things  good,  and  expects  the  gods  to 
complete  what  human  wisdom  began. 

Of  Socrates'  many  preoccupations,  the  idea  of  setting 
up  morals  as  a  science  is  the  principal  one  ;  for  it  alone 
brings  harmony  and  light  into  this  apparently  strange, 
contradictory  character.  It  alone  explains  how  Socrates 
is  both  a  believer  and  a  free  thinker  ;  positive  and 
speculative  ;  a  man  of  his  own  age  and  country,  ever 
disposed  to  adapt  himself  to  his  environment,  and  yet 
one  who  retired  within  himself,  was  ever  master  of 
himself,  obstinately  jealous  of  his  freedom  and  independ- 
ence ;  an  aristocrat  attached  to  the  past,  contemptuous 


yo  STUDIES  IN  PHILOSOPHY 

towards  popular  caprice,  and  at  the  same  time  a 
revolutionist,  demanding  that  the  functions  of  the  State 
be  given  to  the  best  instructed  citizens  ;  in  a  word,  to 
sum  up  everything  perhaps,  it  alone  explains  how  he 
was  both  a  philosopher  and  a  man  of  action. 

The  idea  of  Socrates  is  not  only  novel  and  original, 
it  has  occupied  a  prominent  place  in  the  intellectual 
and  moral  history  of  mankind.  This  r6le  has  been  a 
double  one  :  showing  itself  both  in  the  order  of  the 
practical  and  in  that  of  the  theoretical  sciences. 

In  vain  did  Socrates  scrupulously  confine  himself  to 
the  study  of  human  things  ;  the  productiveness  of  his 
method  in  this  domain,  and  its  conformity  with  the 
Greek  genius,  quickly  caused  it  to  be  regarded  as 
applicable  to  all  objects,  physical  and  metaphysical. 
Plato  and  Aristotle  set  forth  the  principle  of  Socrates  : 
"  The  only  science  is  that  of  the  general,"  as  including 
not  merely  the  science  of  human  things  but  also 
universal  science. 

The  syllogism,  or  deductive  reasoning  in  qualitative 
matter,  the  final  definitive  form  of  the  Socratic  method, 
was  regarded  as  the  expression  of  the  connection 
between  things  in  nature  herself.  From  Aristotle  this 
method  passed  on  to  the  Schoolmen,  who  misinter- 
preted it,  substituting  for  the  living  discourse  of  men 
which  the  Greeks  had  taken  as  the  starting-point  of 
their  discussions,  the  mute,  rigid  text  of  some  particular 
book,  which  was  looked  upon  as  being  truth  itself. 
Nevertheless,  positive  science  gradually  developed. 
On  attaining  to  self-consciousness,  so  to  speak,  it 
declared,  with  Bacon,  that  syllogistic  science  was  nothing 
but  a  science  of  words  ;  and  with  Descartes,  that  the 
general  essences  of  the  Socratics  were  only  empty 


SOCRATES  71 

fiction,  that  science  had  as  its  object  not  quality  or 
the  general,  but  quantity  or  the  relations  of  dimension. 
The  progress  of  science  has  proclaimed  Descartes  to 
be  more  and  more  in  the  right,  and  one  is  nowadays 
tempted  to  ask  oneself  whether  the  Socratic  principle  : 
"  The  only  science  is  that  of  the  general,"  when  applied, 
as  it  has  been,  to  the  investigation  of  the  laws  of  nature, 
has  not  rather  bewildered  and  unsettled  the  human 
mind  than  helped  it. 

Even  were  such  to  be  the  case,  Socrates,  who  de- 
nounced all  investigation  into  moral  causes,  and  claimed 
only  to  build  up  moral  science,  would  not  be  responsible 
for  it.  This  extension  of  the  Socratic  method,  how- 
ever, was  by  no  means  an  aberration  of  the  human 
mind.  Before  knowing  things  in  themselves,  they 
must  be  known  in  their  relations  to  us,  and  it  is  this 
indispensable  provisional  knowledge  that  we  obtain 
from  Socratic  induction  and  definition.  It  may  be  that 
in  all  things  the  element  of  quantity  is  the  ultimate 
object  for  which  science  ought  to  look.  But  it  could 
not  attain  to  this  all  at  once  :  it  must  first  define  the 
qualities  which  form  its  support.  In  every  department 
of  knowledge,  classification  and  induction  must  precede 
the  application  of  mathematical  analysis. 

Anyhow,  the  Schoolmen  with  their  syllogistic 
science,  even  Plato  and  Aristotle,  in  so  far  as  they 
place  being,  strictly  so  called,  in  forms  expressed  by 
our  concepts,  are  not  the  true  successors  of  Socrates. 
Those  he  would  have  recognised  as  such,  are  the 
philosophers  who,  taking  as  their  starting-point  the 
observance  of  the  moral  facts  of  human  nature,  have 
endeavoured  to  set  up  morals  as  a  distinct  and  self- 
sufficient  science.  tThe  purest  and  finest  fruit  of  the 
Socratic  method  consists  of  the  Nicomachean  Ethics, 


72  STUDIES  IN  PHILOSOPHY 

in  which,  without  appealing  to  the  physical  sciences  or 
demanding  of  metaphysics  anything  else  than  an  ardent 
flow  of  the  mind  and  general  views  on  finality  and 
activity,  Aristotle  condensed  in  a  series  of  maxims 
the  very  thought  of  those  who  have  experience  of  life 
think  vaguely  regarding  the  conditions  of  virtue  and 
happiness. 

Nor  is  the  influence  of  Socrates,  along  this  line  of 
investigation  confined  to  antiquity.  When  the  Christian 
religion,  after  proving  adequate  to  the  moral  needs  of 
men  for  fifteen  centuries,  began  to  lose  its  power 
over  their  souls,  the  Socratic  study  of  man  was  restored 
to  favour.  They  were  not  content  with  finding  the  secret 
springs  of  human  actions  in  any  particular  case,  after 
the  fashion  of  the  moralists.  Morals  was  proclaimed 
anew  as  a  distinct  and  separate  science,  with  an  object 
and  a  method  of  its  own.  So  great  an  advance  was 
made  in  this  direction  that  a  daring  system  of  philo- 
sophy, that  of  Kant  and  Fichte,  not  content  with  claim- 
ing a  place  for  moral  science,  began  by  making  a  clean 
sweep  of  the  whole  of  metaphysics,  in  order  that  morals 
might  establish  itself,  unchecked,  in  its  own  fashion  ; 
nor  would  this  philosophy  acknowledge  that  theoretical 
reason  had  any  other  rights  than  those  admitted  by 
moral  science,  thus  organised.  And  soon  afterwards, 
just  as  in  former  times  Plato  and  Aristotle  had  built  up 
a  metaphysical  philosophy  on  the  basis  of  Socratic 
morals,  we  find  Fichte,  Schelling,  and  Hegel  founding  a 
new  philosophy  of  the  absolute  on  the  morals  of  Kant. 

Moral  science,  though  for  a  brief  space  compromised 
by  the  excess  of  its  claims,  now  that  it  has  been  restored 
within  the  limits  marked  out  for  it  by  Socrates,  has 
acquired  fresh  precision  and  vitality  at  the  present  time. 
Even  nowadays,  there  are  many  who  consider  that  the 


SOCRATES  73 

time  has  not  yet  come — if  it  is  ever  to  come — for  morals 
to  assume  the  same  scientific  form  as  physics  or  even 
the  natural  sciences,  and  yet  they  consider  that  it 
admits  of  something  else  than  the  particularities,  in 
which  the  moralist  confines  himself,  or  the  oratorical 
developments  that  suffice  for  the  man  of  action.  The 
truth  in  this  matter  would  seem  to  be,  even  nowadays, 
that  morals  has  a  distinct  domain,  i.e.  the  sum  total  of 
the  moral  facts  of  human  nature,  a  method  proper  to 
itself,  to  wit,  qualitative  induction  and  definition,  and 
that,  by  modestly  confining  itself  to  its  own  domain 
and  scrupulously  adapting  its  means  of  investigation  to 
the  object  under  study,  it  can  attain,  more  certainly 
than  by  any  other  means,  to  the  twofold  end  it  has  in 
view  :  the  knowledge  and  the  direction  of  human 
activity.  ^The  man  whose  ideas  are  most  instinct  with 
life  in  contemporary  society!  is  Socrates. 


ARISTOTLE 

Tb  irpwrov  ov  anrepfia  CCTTLV,  dAXa  TO  reAaov. 

ARISTOT.  Met.  xii.  7.  1073  a  i. 

IF  it  be  true  that  the  genius  of  a  people  is  sometimes 
incarnated  in  certain  men,  and  that  these  mighty,  com- 
prehensive minds  form,  as  it  were,  the  act  and  perfection 
in  which  a  whole  world  of  virtualities  finds  its  goal  and 
completion,  then  Aristotle,  more  than  any  one,  was  such 
a  man  :  in  him  the  philosophic  genius  of  Greece  found 
its  universal,  its  perfect  expression.  It  is  therefore 
something  more  than  the  thought  of  a  single  individual, 
far-reaching  and  profound  though  it  be,  that  we  now 
summon  forth  ;  it  is  the  spirit  of  Greece  itself,  which 
has  reached  the  highest  pitch  of  its  intellectual  great- 
ness. It  will  be  conformable  to  the  analytical  tempera- 
ment of  the  philosopher  with  whom  we  are  now 
dealing,  and  also  practically  indispensable  to  set  up 
numerous  divisions  in  so  vast  a  subject,  and  consider 
its  different  parts  one  after  another. 

I. — BIOGRAPHY  * 

Aristotle    was    born    at    Stageira,    a    Greek    Ionian 
colony  of  Thrace,  situated  on  the  coast  of  the  peninsula 

1  The  ancient  writers  who  deal  with  the  life  of  Aristotle  are  the  follow- 
ing :  (i)  Diogenes  Laertius,  v.  1-35  ;  (2)  Denys  of  Halicarnassus,  letter  to 
Ammaeus,  1.5;  (3)  the  anonymous  author  of  a  biography  of  Aristotle, 

74 


ARISTOTLE  75 

of  Chalcidice,  in  the  year  384  B.C.  He  died,  aged 
sixty-two,  at  Chalcis,  in  Euboea. 

His  father,  Nicomachus,  was  a  doctor,  as  also  were 
his  ancestors.  They  traced  their  descent  back  to 
Machaon,  son  of  Aesculapius  ;  and,  like  many  others, 
were  called  Asclepiads.  Nicomachus  was  physician  to 
the  king  of  Macedonia,  Amyntas  II.,  Philip's  father. 
This  circumstance  may  have  brought  it  about  that 
Aristotle  was  summoned  to  the  court  of  the  king  of 
Macedonia  to  undertake  the  education  of  Alexander. 
It  is  probable  that,  as  an  Asclepiad,  Aristotle  was 
instructed  in  anatomy  at  an  early  age. 

When  about  seventeen  years  old,  he  lost  his  parents. 
Being  now  independent  an£  in  possession  of  a  large 
fortune,  he  was  attracted  to  Athens.  He  went  to  this 
city  the  following  year.  Plato,  who  had  founded  his 
school  there  about  387  B.C.,  was  then  absent ;  he  had 
started  for  Syracuse,  368  B.C.,  left  that  town  three  years 
later,  and  returned  about  360  B.C.  Aristotle  joined 
Plato's  pupils,  remaining  with  them  for  twenty  years, 
until  the  master's  death.  Here  we  find  refuted  the 
story  of  a  quarrel,  which  was  alleged  to  have  arisen 
between  the  master  and  the  disciple  long  before  the 
death  of  Plato,  and  to  have  been  caused  by  Aristotle's 
ingratitude  and  lack  of  consideration.  It  is  said  that 
Plato,  having  remarked  Aristotle's  zeal  and  keenness  of 
mind,  called  him  "  the  reader,"  and  "  intelligence."  In 

published  by  Menage  in  the  second  volume  of  his  edition  of  Diogenes 
Laertius  ;  (4)  the  Pseudo  -  Ammonius  ;  (5)  the  Pseudo  -  Hesychius  ; 
(6)  Suidas,  under  the  article,  'Apto-roT-A^s.  These  texts  may  almost  all  be 
found  in  vol.  i.  of  the  edition  of  Aristotle's  works  undertaken  by  Buhle, 
between  1791  and  1800.  The  relative  importance  of  these  different 
sources  cannot  be  determined  a  priori.  All  that  is  possible  is  the  separate 
examination  of  each  hint  or  indication  from  the  standpoint  of  its  internal 
and  external  probability. 


76  STUDIES  IN  PHILOSOPHY 

all  probability  he  studied  not  only  Platonism  at  Athens, 
but  also  the  other  systems  then  in  vogue. 

Long  previous  to  the  death  of  Plato,  he  gave  proof 
of  his  independence  of  thought  and  action.  Quite 
possibly,  as  a  member  of  the  Platonic  school,  he  had 
already  taught  on  his  own  account.  At  all  events,  he 
began  to  write  at  that  period,  and  though  his  early 
works  were  Platonic  in  form  and  substance,  none  the  less 
did  they  contain,  even  then,  objections  to  the  theories 
of  ideas  along  with  the  affirmation  of  the  eternity  of  the 
world.  He  tells  us  that  it  is  with  regret,  and  because 
of  his  zeal  for  the  superior  interests  of  truth,  that  he 
thus  opposed  his  master.  Moreover,  he  set  an  example 
of  respect  for  the  genius  of  Plato.  In  a  poem  which 
has  come  down  to  us,  he  celebrates  his  master  as  one 
whom  the  wicked  have  no  right  to  praise,  and  who 
showed,  both  by  his  life  and  his  teachings,  how  a  good 
man  is  also  a  happy  man. 

The  death  of  Plato  (347  B.C.)  begins  a  new  period 
in  the  life  of  Aristotle.  He  left  Athens,  accompanied 
by  Xenocrates,  and  went  to  Atarnea,  in  Mysia,  to  his 
friend  and  fellow-disciple,  Hermias,  the  ruler  of  that 
town,  whose  sister,  or  niece,  Pythias,  he  subsequently 
married.  Later  on,  he  married  a  woman  named  Her- 
pyllis.  After  the  fall  and  death  of  Hermias  (345  B.C.) 
Aristotle  went  to  Mytilene.  From  there  he  would  seem 
to  have  returned  to  Athens  and  opened  the  school  of 
rhetoric,  in  which  he  set  up  as  an  opponent  to  Isocrates. 
In  342  B.C.,  he  responded  favourably  to  the  summons 
of  Philip,  king  of  Macedonia,  who  requested  him  to 
undertake  the  education  of  his  son  Alexander,  at  that 
time  about  fourteen  years  of  age.  He  remained  at  the 
court  of  Macedonia  until  Alexander  undertook  his 
expedition  into  Asia  (334  B.C.).  Without  losing  him- 


ARISTOTLE  77 

self  in  pursuit  of  an  ideal  too  far  removed  from  the 
conditions  of  practical  life,  Aristotle  appears  to  have 
instilled  generous  qualities  in  the  mind  of  his  pupil. 
Throughout  his  life  Alexander  retained  feelings  of 
respect  and  love  for  his  master,  though  after  the  death 
of  Callisthenes,  Aristotle's  nephew,  in  325  B.C.,  all 
relations  between  the  two  were  discontinued. 

In  335  B.C.  or  334  B.C.,  Aristotle  returned  to  Athens, 
and  at  Lycaeum  opened  what  was  called  the  Peripatetic 
School,  probably  on  account  of  the  master's  habit  of 
walking  about  with  his  disciples  as  they  talked  of  science 
and  philosophy.  In  the  mornings,  relates  Aulus-Gellius, 
Aristotle  gave,  to  a  chosen  body  of  hearers,  acroamatic, 
or  esoteric,  instruction,  dealing  with  the  most  difficult 
portions  of  philosophy,  mainly  dialectic  and  the  philo- 
sophy of  nature.  In  the  evenings  he  gave  exoteric 
instruction  to  all  who  offered  themselves,  dealing  with 
rhetoric,  topics  and  politics.  His  teaching  took  the  form 
both  of  classes  and  lectures  ;  and  his  school,  like  that 
of  Plato,  was  a  band  of  friends  who  assembled  on  fixed 
days,  and  took  their  meals  in  common. 

Wealthy  himself,  and  able  to  rely  on  the  assistance 
of  the  king,  Aristotle  was  in  a  position  to  obtain  all  the 
scientific  resources  the  society  of  the  times  could  offer. 
It  is  said  that  Alexander  sent  him  eight  hundred  talents 
to  enable  him  to  complete  his  Historia  animalium.  It 
is  even  related  that  he  placed  at  his  disposal  millions  of 
men,  whose  duty  it  was  to  seek  out  animals  of  every 
kind,  especially  fishes,  to  maintain  in  perfect  order 
aviaries  and  gardens  filled  with  animals,  and  to  keep 
the  philosopher  informed  on  such  observations  and  dis- 
coveries as  were  calculated  to  advance  science.  These 
are,  doubtless,  mere  inventions,  though  facts  were  at  the 
bottom  of  them.  Certainly  Aristotle  gathered  together 


78  STUDIES  IN  PHILOSOPHY 

all  the  documents  of  every  kind  it  was  possible  for  him 
to  obtain.  He  was  the  first  to  form  a  large  collection 
of  books. 

Although  Aristotle  had  broken  off  all  relations  with 
Alexander  in  325  B.C.,  none  the  less  did  the  king's 
death,  two  years  afterwards,  prove  an  occasion  of  peril 
for  him.  When  the  Lamian  war  broke  out,  he  was 
looked  upon  as  a  friend  of  the  kings  of  Macedonia  and 
Antipater,  and  was  prosecuted  on  the  charge  of  atheism. 
He  left  Athens,  so  that  the  Athenians,  as  he  said,  might 
not  a  second  time  be  guilty  as  regards  philosophy. 
He  fled  to  Chalcis,  in  Euboea,  where  he  fell  sick  and 
died  in  the  summer  of  322  B.C.,  a  few  months  before 
Demosthenes,  who  was  born  in  the  same  year  as  himself. 
He  was  sixty-two  years  of  age. 

Though  early  attacked  by  his  political  and  scientific 
opponents,  he  would  appear  from  his  writings  to  have 
been  of  a  noble,  humane,  and  loyal  nature,  and  we  are 
acquainted  with  no  actual  proved  fact  to  the  contrary. 
His  life  bears  the  impress  of  moral  philosophic  dignity. 
Aristotle  was  both  a  creative  and  a  universal  genius,  and 
an  indefatigable  worker.  He  is  devoid  of  the  ardent 
buoyancy  of  Plato.  With  mind  bent  on  the  reality 
presented  to  him,  whatever  bears  no  relation  thereto  he 
looks  upon  as  fantastic  ;  he  does  not  bury  himself  in  the 
facts  of  the  sensible  world,  however,  but  is  always  look- 
ing for  the  intelligible.  In  all  things  he  recommends 
moderation,  the  golden  mean.  A  moderate  fortune, 
government  by  the  moderate  classes :  such,  to  his  mind, 
is  the  best  condition  both  for  the  individual  and  for 
society. 

We  are  told  that  he  was  short  and  thin,  with  small 
eyes  and  an  ironical  expression  playing  about  his  mouth. 
By  Pythias,  his  first  wife,  he  left  a  daughter  of  the  same 


ARISTOTLE  79 

name  ;  and  by  his  second  wife,  Herpyllis  of  Stageira, 
a  son  Nicomachus,  whose  name  we  find  in  the  Nico- 
machean  Ethics.  In  his  will  he  speaks  affectionately  of 
both  his  wives  and  of  his  two  brothers  and  their  children  ; 
he  also  refers  in  sympathetic  terms  to  his  friends  and 
distant  relatives. 


II. — ARISTOTLE'S  WRITINGS 

The  story  of  the  preservation  of  Aristotle's  works  is 
but  little  known.  According  to  Strabo  and  Plutarch, 
the  writings  of  Aristotle  and  Theophrastus,  after  the 
latter's  death,  fell  into  the  hands  of  Neleus,  who  took 
them  to  his  home  in  Skepsis,  Mysia.  There  they  would 
appear  to  have  been  hidden  away  in  a  cellar,  where  they 
were  discovered  by  Apellicon,  in  the  time  of  Sylla.  The 
latter  is  reported  to  have  had  them  transferred  to  Rome. 
Whatever  degree  of  truth  there  may  be  in  these  anec- 
dotes, the  texts  that  had  been  preserved  were  revised  and 
classified  in  the  first  century  before  Christ  by  Andronicus, 
a  Peripatetic  philosopher  of  Rhodes,  who  published  a 
complete  edition  about  60-50  B.C.  It  is  this  text  of 
Aristotle,  more  or  less  remodelled,  that  we  now  possess. 
In  all  probability  our  collection  contains  everything 
authentic  that  existed  in  the  time  of  Andronicus,  and 
we  have  good  grounds  for  regarding  as  apocryphal  the 
works  mentioned  by  Diogenes  Laertius,  that  are  absent 
from  this  collection.  Most  likely,  however,  all  that  is 
contained  in  the  so-called  Andronicus  edition,  is  not  by 
Aristotle  ;  even  the  authentic  works  themselves  are  not 
free  from  additions  and  changes.  There  have  also  come 
down  to  us  the  titles  of  works  that  are  certainly  authentic, 
and  yet  are  lacking  in  our  collection,  having  apparently 
been  lost  at  the  time  of  Andronicus.  All  the  same,  it 


8o  STUDIES  IN  PHILOSOPHY 

would  appear  that  the  most  important  works  on  Aris- 
totelian philosophy  and  science  have  been  preserved. 

Which  of  the  works  we  possess  are  to  be  laid  aside 
as  unauthentic  ?  In  many  cases  the  question  cannot  be 
answered  with  any  degree  of  precision  or  certainty.  The 
following  are  the  results  reached  by  Zeller  in  his  Philo- 
sophic der  Griechen,  3rd  vol.  3rd  edition.  The  authen- 
ticity of  the  following  works  is  either  inadmissible  or 
very  doubtful  : — De  Xenophane,  Zenone  et  Gorgia ;  De 
animalium  motu ;  De  plantis ;  De  coloribus ;  De  audi- 
bilibus ;  De  mirabilibus  auditis  •  Physiognomonica ;  Me- 
chanica  problemata  ;  De  indivisibilibus  lineis  ;  De 
mundo  ;  De  respiratione ;  De  virtutibus  et  vitiis  ; 
Oeconomica  ;  Rhetorica  ad  Alexandrum.  The  Eudemian 
Ethics  and  the  Great  Ethics  are  alterations  of  the  Nico- 
machean  Ethics.  Such  fragments  of  letters  as  we  possess 
have  undergone  considerable  additions  and  changes. 

The  works  left  by  Aristotle  may  in  all  probability  be 
placed  in  the  three  following  categories  : — 

I  st.  Books  of  instruction  and  science  properly  so 
called  :  summaries  and  treatises  of  which  he  made  use 
in  his  classes.  He  did  not  publish  them,  but  merely 
imparted  their  teachings  to  his  pupils. 

2nd.  Published  writings  :  intended  for  the  masses 
of  the  people.  They  were  written,  we  are  told,  with 
considerable  fluency  and  charm,  and  were  partly  in  the 
form  of  dialogues. 

Using  Aristotle's  own  terms,  the  unpublished  writ- 
ings have  been  called  acroamatic  or  acroatic,  and  the 
published  ones  exoteric.  These  expressions  clearly 
answer  to  a  fundamental  distinction  in  Aristotle's 
philosophy.  In  his.  mind,  there  are  two  modes  of 
instruction,  proportioned  to  the  two  degrees  of  know- 
ledge. That  which  is  cognizable  as  necessary  and 


ARISTOTLE  81 

absolutely  certain  is  a  matter  of  demonstration  strictly 
so  called  ;  that  which  is  cognizable  as  being  only  likely 
is  a  matter  of  dialectic.  In  his  classes  Aristotle  taught 
complete  science  ;  he  gave  demonstrations  ;  the  pupil 
had  nothing  to  do  but  to  listen.  Apart  from  these  classes, 
however,  Aristotle  directed  dialectic  conversations,  in 
which  reasoning  from  probabilities,  and  from  considera- 
tions more  or  less  foreign  to  the  subject  in  question,  was 
carried  on  ;  to  these  conversations  others  were  admitted 
as  well  as  pupils.  Such  is  the  significance  of  the  words 
acroamatic  and  exoteric,  used  with  reference  to  the 
teaching  of  Aristotle.  He  himself  does  not  apply 
them  to  his  works,  though  such  application  may  well 
be  made. 

3rd.  To  these  two  categories  must  be  added  a  third, 
viz.,  notes  intended  for  the  personal  use  of  Aristotle. 
These  latter  writings  may  be  called  hypomnematic. 

Last  of  all,  Aristotle  left  behind  him  speeches,  letters 
and  poems.  Of  these  three  classes,  only  the  first  have 
come  down  to  us,  and  a  few  fragments  of  the  second 
and  third.  Amongst  the  lost  works,  the  most  important 
are,  in  the  first  category  :  the  Treatise  on  Plants,  Anatomy, 
Astrological  Theorems ;  in  the  second  :  the  Dialogues, 
and  the  History  of  Rhetoric ;  in  the  third :  extracts  from 
some  works  of  Plato,  and  writings  on  the  Pythagoreans 
and  other  philosophers.  In  this  third  category,  evidently, 
we  must  place  the  Constitutions  (HoXtretai),  in  which 
were  to  be  found  all  kinds  of  information  about  158 
Greek  and  foreign  cities,  a  lost  collection  of  which  we 
possess  many  very  interesting  extracts.  The  treatise 
entitled  The  Constitution  of  the  Athenians  was  recently 
discovered  on  a  papyrus,  and  published  in  1891. 

We  may  classify  as  follows  the  scientific,  properly 
so  called,  or  unpublished  writings,  in  our  possession, 

G 


82  STUDIES  IN  PHILOSOPHY 

representing,  in  a  probably  complete  manner  as  regards 
essentials,  the  philosophical  work  of  Aristotle  : 

i  st.  Works  on  logic,  collected  at  the  Byzantine 
period  only  under  the  name  of  opyavov  :  Karyyopiat 
(categories),  partially  added  to  and  altered  ;  Hepl 
eppyveias  (on  speech  or  propositions)  —  this  appears 
to  be  the  work  of  a  Peripatetic  of  the  third  century 
before  Christ ;  'AvaXim/ca  Trporepa  (Prior  Analytics), 
dealing  with  syllogism  ;  'AvaXvTiKa  wrepa  (Posterior 
Analytics),  dealing  with  demonstration;  Tom/cd  (Topics), 
dealing  with  dialectic,  or  reasoning  in  probabilities. 
The  ninth  book  of  this  work  is  usually  given  as  a 
special  work,  entitled  Hepl  a-ofaa-Tiicwv  eKeyxwv  (On 
Sophistical  Refutations). 

2nd.  Works  on  natural  philosophy  :  Qvaiicr)  a/cpoacr^ 
(Physics),  in  eight  books,  the  seventh  of  which,  though 
edited  from  Aristotelian  notes,  does  not  appear  to 
have  been  written  by  Aristotle ;  Hepl  ryeveaeax;  KOI 
<f>0opas  (On  Generation  and  Destruction)  ;  Hepl 
ovpavov  (On  the  Heavens)  ;  MerewpoKoyt/cd  (Meteoro- 
logies) ;  Hepl  ^/ru^5  (On  Soul),  and  divers  treatises 
referring  thereto,  entitled  Parva  Naturalia ;  Hepl 
ra  %wa  iaroplai  (Animal  History),  in  ten  books,  a 
work  that  has  undergone  considerable  changes  and 
the  tenth  book  of  which  is  not  authentic  ;  Hepl 
fjbopiwv  (The  Parts  of  Animals)  ;  Hepl  Tropelas 
(The  Motor  Organs  of  Animals)  ;  ITe^l  tyw 
(On  the  Generation  of  Animals),  a  work  that  has  been 
considerably  changed. 

3rd.  So-called  metaphysical  works,  dealing  with  what 
Aristotle  calls  first  philosophy  (TT/JCOTT;  ^tXoo-o^ta)  : 
the  work  called  Metaphysics,  in  fourteen  books,  is  a 
collection  made,  in  all  probability,  shortly  after  the 
death  of  Aristotle  ;  it  comprises  all  that  his  papers 


ARISTOTLE  83 

contained  referring  to  first  philosophy.  These  writings 
owe  their  present  name  (Ta  /tera  ra  <j>v<ritca)  to  their 
position  after  physics  in  the  edition  of  Andronicus. 
The  substance  of  these  writings  is  comprised  in  Books 
i.,  iii.,  iv.,  vi.  to  ix.,  x.  (numbers  of  the  Berlin  edition). 
Book  ii.  and  Book  xi.,  from  chapter  viii.,  1065,  a,  26, 
are  unauthentic. 

4th.  Works  on  the  practical  sciences  :  'HOnca 
Nt/co/ia^em  (Morals  addressed  to  Nicomachus,  or  Nico- 
machean  Ethics)  ;  IIo\m/ca  (Politics),  an  unfinished 
work.  According  to  Zeller,  Books  vii.  and  viii.  of 
the  Politics  ought  most  probably  to  be  inserted  between 
Books  iii.  and  iv.  ;  Te^v^  pyropiicij  (Rhetoric)  ;  Hepl 
TroiriTucfy  (On  Poetry). 

With  regard  to  the  didactic  works,  the  question  of 
chronology  is  only  of  moderate  importance.  Indeed, 
they  were  all  written  during  the  last  twelve  years  of 
the  philosopher's  life  (335-322),  they  make  references 
to  each  other,  and,  in  their  ensemble,  offer  us  the  com- 
pleted system,  without  any  proof  of  progress.  So  far 
as  can  be  judged  by  the  paltry  indications  that  may  be 
obtained  from  historical  testimony  and  the  examination 
of  the  works  in  themselves,  Aristotle  first  wrote  the 
works  on  logic  (except  the  notes  from  which  the  liepl 
epWveLas  were  compiled,  and  which  appeared  after  the 
Hepl  ^v%^?).  Then  the  writings  on  natural  history 
appeared,  followed  by  the  physiological  and  psycho- 
logical works  and  those  relating  to  the  practical 
sciences  ;  last  of  all,  most  probably  and  in  any  case 
subsequent  to  the  physics,  the  collection  called  meta- 
physics. Thus  Aristotle  appears  to  have  proceeded 
from  the  abstract  to  the  concrete,  and,  in  the  domain 
of  the  concrete,  from  changing  being  to  immutable 
being. 


84  STUDIES  IN. PHILOSOPHY 

III. — THE  ENSEMBLE  OF  ARISTOTLE'S  WORK 

As  indicated  by  the  very  title  of  his  writings, 
universality  is  the  first  characteristic  of  Aristotle's 
work.  Theory  and  practice,  metaphysics  and  the 
science  of  observation,  erudition  and  speculation,  his 
philosophy  includes  everything.  It  is,  or  would  like 
to  be,  knowledge  in  its  totality.  The  idea  of  science, 
considered  as  the  loftiest  object  of  activity,  stands  out 
in  Aristotle  as  more  precise  than  in  Plato  and  more 
general  than  in  Anaxagoras  and  Democritus.  It  is 
not  the  curiosity  of  a  scholar,  it  is  the  ambition  to  enter 
into  the  very  essence  and  cause  of  things.  Without 
exception,  everything  that  is,  even  what  appears  mean 
and  insignificant,  calls  in  this  sense  for  the  philosopher's 
investigations.  He  knows  he  will  find  the  divine  and 
the  intelligible  in  all  the  productions  of  nature,  even 
those  that  are  apparently  the  humblest. 

It  was  thus  that  he  approached  everything  accessible 
to  human  intelligence  ;  and,  provided  with  all  the  positive 
knowledge  it  was  at  that  period  possible  to  acquire,  a 
philosopher  of  penetrating  intuition  and  strict  reasoning 
power,  he  either  created  or  constituted  most  of  the 
sciences  which  the  genius  of  mankind  was  subsequently 
destined  to  develop.  The  list  of  the  sciences  he  thus 
organised  is  but  the  list  of  those  he  himself  studied  : 
the  history  of  philosophy,  logic,  metaphysics,  general 
physics,  biology,  botany,  ethics,  politics,  archaeology, 
literary  history,  philology,  grammar,  rhetoric,  poetics 
and  the  philosophy  of  art.  In  each  of  these  sciences 
Aristotle  is  at  home  ;  for  each  of  them  he  lays  down 
special  and  appropriate  principles.  A  pure  ethicist  when 
dealing  with  justice  and  friendship,  he  is  a  professional 
naturalist  when  dealing  with  zoology. 


ARISTOTLE  85 

Are  we  then  to  conclude  that  Aristotle  comprises 
many  human  beings  in  himself,  so  to  speak  ;  is  his 
vast  work  nothing  but  the  juxtaposition  of  the  most 
diverse  labours,  such  as  might  result  from  the  collabora- 
tion of  many  learned  men  ?  Such  an  appreciation 
would  certainly  be  a  superficial  one.  First  and  foremost, 
there  is  community  of  spirit  and  method  between  the 
different  works  of  Aristotle.  This  common  substratum 
might  be  defined  as  a  harmonious  blend  of  idealism, 
observation  and  logical  formalism.  Aristotle  always 
seeks  for  the  idea  in  the  fact ;  for  the  necessary  and 
the  perfect  in  the  contingent  and  the  imperfect ;  every- 
where he  endeavours  to  substitute  fixed  conceptions 
and  definitions  for  the  fleeting  data  of  sensible  observa- 
tion. Nor  is  this  all  ;  according  to  him,  the  different 
parts  of  knowledge  hold  a  fixed  relation  to  one  another, 
and  this  relation  he  very  clearly  defines.  Speaking 
generally,  the  superior  is  known  only  after  the  inferior, 
and  that  only  by  the  help  of  the  knowledge  of  this 
inferior  ;  at  the  same  time,  however,  the  true  cause  and 
raison  d'etre  of  the  inferior  is  to  be  found  in  the  superior. 
For  instance,  the  soul  can  be  known  only  after  the 
body,  which  is  its  basis  and  the  condition  of  its  exist- 
ence. But  the  body  exists  only  for  the  soul ;  from 
this  latter  it  obtains  the  regulated  movement  which  con- 
stitutes its  being.  This  principle  of  Aristotle's  will 
assist  us  in  classifying  the  many  forms  of  his  philo- 
sophical activity. 

IV. — CLASSIFICATION   OF  THE  SCIENCES 

Without  attaining  to  precision  or  even  permanence 
in  detail,  Aristotle  was  none  the  less  the  first  to  con- 
ceive of  science  from  an  encyclopaedic  point  of  view 


86  STUDIES  IN  PHILOSOPHY 

and  to  endeavour  to  discover  a  principle  for  the  com- 
plete classification  of  knowledge. 

In  the  first  place,  science  stands  clearly  out  from 
the  very  things  to  which  it  relates.  It  consists  of  the 
conception  of  things  as  necessary,  and  admits  of  different 
degrees  according  as  the  object  under  consideration 
itself  admits  of  necessity  or  only  of  probability. 

Science,  in  its  ensemble,  follows  a  double  line  of 
direction,  according  as  the  human  mind  adopts  as  its 
starting-point  that  which  is  first  from  its  own  point  of 
view,  or  that  which  is  first  absolutely.  These  two 
steps  are  the  very  opposites  of  each  other  :  for  facts 
are  what  is  first  to  us,  and,  in  the  internal  order  of 
nature,  facts  are  what  exists  in  the  last  resort ;  and  vice 
versa,  what  is  first  in  itself  consists  of  principles,  and 
principles  are  the  last  thing  to  which  we  can  attain. 

Philosophy,  in  the  broad  sense  of  the  word,  is  science 
in  general.  In  the  first  place,  it  comprises  first  philo- 
sophy or  the  science  of  unconditioned  principles  ;  in 
the  second  place,  the  totality  of  the  particular  sciences, 
the  chief  of  which  are  :  mathematics,  physics,  ethics 
and  poetics.  Philosophy  is  one,  thanks  to  first  philo- 
sophy, which  is  the  common  reservoir  whence  all  par- 
ticular sciences  draw  their  principles. 

This  division,  although  fundamental,  does  not  always 
reappear  in  Aristotle's  classifications  of  the  sciences. 
In  certain  places  he  divides  the  propositions,  after  the 
fashion  of  the  Platonists,  into  ethical,  physical  and 
logical,  these  latter  comprising  the  very  propositions 
that  refer  to  first  philosophy. 

More  frequently  he  divides  the  sciences  into  theo- 
retical, practical  (or  relating  to  action)  and  poetical 
(or  relating  to  production  by  means  of  matter)  ; 
placing,  from  the  logical  and  absolute  point  of  view, 


ARISTOTLE  87 

theory  before  practice,  and  practice  before  poetics. 
Then  he  subdivides  the  theoretical  sciences  into  theo- 
logy, mathematics  and  physics.  Theology  may  be 
brought  under  first  philosophy,  of  which  it  forms  the 
summit.  Mathematics  deals  with  essences  still  stable 
though  not  separable  from  matter,  except  by  abstraction. 
Physics  deals  with  sensible — i.e.  movable  and  perishable 
— substances.  The  practical  sciences,  or  sciences  of 
human  things,  are  subdivided,  if  we  proceed  from 
potency  to  act,  i.e.  from  that  which  is  first  for  us  to 
that  which  is  first  in  itself,  into  ethics >  economics  and 
politics.  In  fact,  economics  is  often  given  by  Aristotle 
as  included  in  politics.  Rhetoric  is  more  particularly 
set  forth  as  an  auxiliary  science  to  politics.  Poetics 
includes  all  the  arts,  the  most  important  of  them  being 
poetry  and  music.  No  mention  is  made  of  logic  in 
this  classification,  doubtless  because  the  latter  embraces 
only  the  sciences  dealing  with  realities,  whereas  logic 
deals  with  concepts. 

V. — METHOD  AND  THE  POINT  OF  VIEW 

The  object  Aristotle  has  in  view  is  essentially  theo- 
retical. To  know  in  order  to  know,  to  understand,  to 
adjust  things  to  the  intelligence  :  such  is  the  end  of  all 
his  efforts. 

All  men,  he  says,  have  a  natural  desire  to  know. 
We  love  science  quite  apart  from  any  advantage  to  be 
gained  thereby.  Wisdom  is  independent  of  utility  ; 
in  fact,  the  greater  it  is.  the  less  useful  it  is.  The 

o  - 

highest  science  is  that  of  the  goal  or  end,  in  view  of 
which,  beings  exist.  This  science  alone  is  truly  free, 
because  it  alone  exists  solely  in  view  of  knowledge 
itself.  It  is  the  least  necessary  of  all  sciences,  and 


88  STUDIES  IN  PHILOSOPHY 

therefore  the  most  excellent.  Science  enables  us  to 
become  acquainted  with  the  intelligible  reasons  of  things. 
The  ignorant  man,  who  all  the  same  observes,  is 
astonished  that  things  are  as  they  are,  and  this  very 
astonishment  is  the  beginning  of  science  :  the  wise  man 
would  be  astonished  were  things  otherwise  than  as  he 
knows  them. 

How  does  Aristotle  proceed  in  order  to  acquire 
science,  thus  understood  ?  Aristotle  is  neither  the 
dogmatic  idealist  that  Bacon  supposes,  building  up  the 
world  with  nothing  but  the  categories  of  language,  nor 
the  empiric  that  many  moderns  see  in  him.  •  He  is  both 
an  observer  and  a  constructor :  speaking  generally,  he 
closely  allies  and  combines  the  scrupulous  study  of  facts 
with  the  effort  to  make  them  intelligible.  For  him, 
facts  are  the  starting-point,  but  he  does  not  stop  there  : 
he  tries  to  distil  from  them  the  rational  truths  he  knows 
beforehand  to  be  contained  therein.  The  end  he  has 
in  view  is  the  knowledge  of  things  in  demonstrative 
form,  i.e.  in  the  form  of  a  deduction  in  which  the 
properties  of  the  thing  are  known  by  its  very  essence. 

Most  frequently,  and  especially  when  dealing  with 
metaphysical  or  moral  matters,  before  entering  upon 
the  study  of  things  in  themselves,  he  investigates  and 
discusses  all  the  opinions  of  others  thereon.  This  is 
the  dialectic  method  ;  drawing  its  arguments  not  from 
the  essence  of  the  thing,  but  rather  from  the  admissions 
of  one's  interlocutor,  it  does  not  go  beyond  proba- 
bility. In  using  this  method,  Aristotle  frequently  begins 
with  popular  conceptions  :  he  finds  a  philosophical 
meaning  in  them  and  utilises  it  in  constructing  his 
theory.  He  also  starts  with  language  which,  for  him, 
is  a  sort  of  intermediary  between  things  and  reason. 
He  pays  special  attention  to  the  doctrines  of  his  pre- 


ARISTOTLE  89 

decessors,  carefully  going  over  all  the  opinions  they  have 
upheld  ;  and  even  when  rejecting  these  opinions,  he  tries 
to  find  out  the  reason  they  were  held  and  the  degree 
of  truth  in  them.  His  philosophical  dissertations  are 
generally  composed  as  follows:  ist,  he  determines  the 
object  of  investigation,  so  as  not  to  be  exposed  to  mis- 
understanding, as  is  the  case  with  Plato ;  2nd,  he 
enumerates  and  estimates  the  indications  and  opinions 
held  on  the  matter  in  hand  ;  3rd,  he  investigates  and 
examines  as  completely  as  possible  the  difficulties  or 
airopiai  offered  by  the  question  asked  ;  4th,  considering 
things  in  themselves,  and  utilising,  in  his  reasonings,  the 
results  of  the  foregoing  discussions,  he  seeks  for  the 
solution  of  the  problem  in  the  determination  of  the  one 
eternal  essence  of  the  object  in  question. 

VI. — ARISTOTLE,  THE  HISTORIAN 

We  see  from  the  preceding  that  Aristotle  is  a  his- 
torian above  all  else.  He  began  by  learning  as  much 
as  possible.  According  to  report,  Plato  called  him 
the  reader.  But  history  was  not  a  final  end  for  him, 
although  he  manifested  extraordinary  curiosity  regarding 
facts ;  it  was,  however,  an  indispensable  means  to  an  end. 
It  supplied  the  mind  with  materials  without  which  it 
would  have  nothing  to  work  upon.  Aristotle  gave 
himself  up  to  profoundly  historical  studies  in  every 
domain  of  science. 

As  regards  the  history  of  philosophy,  he  wrote  mainly 
on  Platonism  and  the  Pythagoreans.  The  whole  of 
the  first  book  of  the  Metaphysics  is  full  of  historical 
research:  it  is  a  summary  of  the  principles  set  forth 
from  the  time  of  Thales  to  that  of  Plato.  But  as  the 
object  he  has  in  view  is  dogmatic,  he  makes  previous 


9o  STUDIES  IN  PHILOSOPHY 

systems  fit  into  the  framework  of  his  own  philosophy. 
He  tries  to  find  their  perfect  form,  the  idea  within 
each,  their  end  and  completion  ;  he  is  determined  to 
understand  them  more  profoundly  than  even  their 
authors  did,  and  he  summarises  them  into  rules  created 
by  himself,  which  rules  are  used  as  stepping-stones  to 
his  own  system.  When  he  classifies  doctrines,  he  does 
so  according  to  the  resemblances  and  differences  they 
offer  from  his  own  point  of  view,  not  according  to  the 
influence  they  have  had  upon  one  another.  Thus,  the 
summary  contained  in  the  first  book  of  the  Metaphysics 
is  intended  to  prepare  the  ground  for  the  Aristotelian 
theory  of  the  four  causes.  Aristotle  shows  that,  before 
his  time,  the  material,  motive  and  formal  principles  were 
more  or  less  discerned  and  rightly  estimated,  but  that 
the  final  cause  was  spoken  of  as  though  by  accident,  as 
something  unessential.  Anaxagoras,  who  had  caught  a 
glimpse  of  the  final  cause,  stands  out,  says  our  author, 
as  a  sensible  man  amongst  men  who  speak  at  random. 
Chronological  investigations  have  little  to  do  with  these 
considerations.  Aristotle,  likewise,  troubles  himself 
little  with  the  relations  of  master  to  disciple.  He  notes 
the  services  rendered  by  each  of  his  predecessors  to 
philosophy  in  general,  as  he  conceives  it ;  he  points  out 
anything  of  a  lasting  nature  that  each  thinker  has  found, 
and  mentions  the  inventors  and  promoters  of  ideas  that 
have  played  some  part  in  the  development  of  science 
and  appear  to  him  deserving  of  examination.  In  a  word, 
making  no  attempt  to  find  out  the  historical  origins  of 
the  systems,  he  all  the  same  elicits  from  the  crude  mass 
of  facts,  the  logical  formation  of  definitive  philosophy. 

With  political  history  are  connected  the  famous 
TroTuretat  in  which  Aristotle  set  forth  the  constitutions 
of  158  Greek  and  foreign  cities.  This  collection 


ARISTOTLE  91 

of  treatises  belonged  to  what  we  call  archaeology  and 
the  history  of  civilisation.  In  them  were  to  be  found 
many  a  striking  national  custom,  and  even  the  proverbs 
and  popular  songs  of  different  peoples.  According 
to  certain  Greek  commentators,  the  order  of  the  con- 
tents was  alphabetical.  Diogenes  says  that  the  con- 
stitutions were  classified  according  as  they  resembled 
democracies,  oligarchies,  aristocracies  or  tyrannies.  We 
can  nowadays  form  some  idea  of  the  iro\Lrela^  thanks 
to  the  recently  discovered  treatise  on  the  Constitution  of 
the  Athenians.  The  first  part  of  this  treatise  is  an 
explanation  of  the  political  transformations  of  Athens 
from  its  historical  beginnings.  The  second  describes 
the  political  and  administrative  organisation  of  Athens 
about  the  time  of  the  Crown  trial  (330  B.C.). 

In  the  literary  order  of  things,  Aristotle  had  written 
the  history  of  rhetoric  and  poetry.  This  history,  which 
has  not  come  down  to  us,  was  greatly  praised  by  Cicero. 
"  Aristotle,"  he  said,  "  had  noted  down  all  the  precepts 
given  by  the  rhetors,  and  that,  too,  with  such  a  degree 
of  perfection  that  these  precepts  were  found  to  be  more 
clearly  set  forth  by  him  than  by  their  authors  them- 
selves ;  so  that  when  one  wished  to  become  acquainted 
with  them,  it  was  in  Aristotle's  works  that  search  was 
made." 

He  had  also  drawn  up  chronological  lists  of  dramatic 
performances  as  well  as  lists  of  the  victors  in  the  Olympic 
and  Pythian  games.  These  works  are  lost. 

As  may  be  seen,  Aristotle's  curiosity  is  insatiable  and 
embraces  every  department  of  nature.  Still,  he  is 
determined  to  know  and  understand,  not  to  amuse  him- 
self with  the  mere  statement  of  facts  :  history  for  him 
is  nothing  but  an  instrument  of  science,  and  a  fact  has 
no  value  except  as  the  vehicle  of  an  idea. 


92  STUDIES  IN  PHILOSOPHY 

VII.— LOGIC 

Aristotle  is  determined  to  become  acquainted  with 
facts,  not  only  as  regards  what  they  are,  but  as  regards 
what  they  ought  to  be  ;  he  wishes  to  resolve  the  con- 
tingent into  the  necessary.  First,  then,  he  has  to  find 
out  under  what  conditions  the  mind  conceives  some- 
thing as  necessary  ;  in  other  words,  he  has  first  to 
consider  science  in  its  form,  putting  on  one  side  its 
content  :  this  is  the  object  of  logic. 

Logic  is  the  determination  of  the  laws  of  reasoning 
and  of  the  conditions  of  science.  In  knowledge  Aristotle 
makes  a  distinction  between  form  and  matter,  he  regards 
form  as  possessed  of  an  existence  and  laws  of  its  own. 
Its  existence  lies  in  the  reality  of  stable  concepts,  or 
general,  single  ideas,  exactly  determined  both  as  regards 
their  comprehension  and  their  extension.  Its  funda- 
mental law  is  the  principle  of  contradiction  :  "  It  is 
impossible  for  one  and  the  same  attribute  to  belong 
and  not  to  belong  to  a  given  subject,  regarded  in  one  and 
the  same  connection."  Moreover,  according  to  Aristotle, 
there  is  proportion  as  well  as  agreement  between  thought 
and  being ;  consequently,  our  philosopher  does  not 
object  to  the  introduction  into  his  logic  of  many 
elements  of  a  metaphysical  nature. 

Aristotelian  logic  is  the  rational  analysis  of  the  condi- 
tions which  any  reasoning  must  satisfy  for  its  conclusion 
to  be  regarded  as  necessary.  The  thing  is  not  to  know 
how,  as  a  matter  of  fact,  we  reason  in  ordinary  life,  but 
rather  how  reasoning  must  be  built  up  in  order  that  the 
necessity  of  the  connection  it  establishes  may  appear 
immediately  and  irresistibly  evident.  This  is  why  the 
problem  of  the  psychological  analysis  of  natural  reason- 
ing, indicated  by  Locke,  could  be  substituted  for  that 


ARISTOTLE  93 

of  Aristotle  only  by  admitting  the  reduction  of  the 
necessary  to  the  contingent,  the  ideal  to  the  real, 
precept  to  fact,  and  art  to  nature. 

It  is  advisable  to  distinguish  between  :  ist,  the 
instruments  of  thought ;  2nd,  the  role  and  value  of 
these  instruments  in  the  constitution  of  science. 

The  instruments  of  thought  are  :  notions,  proposi- 
tions and  reasoning. 

The  general  heading  of  notions  includes  the  predic- 
ables,  the  categories  and  the  notions  of  logical  relations. 

The  predicableSy  which  Aristotle,  it  would  seem,  calls 
the  genera  of  problems,  are  the  universal  notions  that 
relate  to  the  general  modes  according  to  which  one 
thing  may  be  enunciated  with  reference  to  another. 
These  are  what  are  called  the  universals,  viz.  genus, 
species,  difference,  property  and  accident. 

The  categories  are  the  irreducible  genera  of  words, 
and  consequently  of  things,  for  classes  of  words  are  the 
classes  of  the  things  themselves.  These  are  the  ultimate 
genera.  The  categories  are  ten  in  number  :  ist,  essence, 
for  instance,  man,  horse  ;  2nd,  quantity  :  two  ells  long  ; 
3rd,  quality  :  white  ;  4th,  relation  :  double,  half ;  5th, 
place  :  at  school ;  6th,  time  :  yesterday  ;  yth,  position  : 
to  be  seated,  lying  ;  8th,  possession  :  to  be  shod,  armed  ; 
9th,  action  :  to  cut,  to  burn  ;  loth,  passivity  :  to  be  cut, 
burnt.  The  categories  are  divided  into  two  classes, 
essence  alone  forming  the  first,  and  the  nine  other 
categories  constituting  the  second. 

This  table  of  categories  seems  to  have  been  drawn 
up  empirically  by  comparison  of  the  words  with  one 
another.  It  differs  fundamentally  from  Kant's,  which 
sets  forth  the  different  ways  of  connecting,  a  priori  and 
necessarily,  the  various  elements  of  an  intuition  in  general, 


94  STUDIES  IN  PHILOSOPHY 

i.e.  of  bringing  this  scattered  matter  under  the  unity  of 
transcendental  apperception. 

The  different  logical  relations  of  terms  to  one  another 
are  identity  and  opposition,  the  latter  including  con- 
trariety, contradiction  and  the  relation  between 
deprivation  and  possession. 

The  general  principle  with  regard  to  opposition  is 
that  two  terms  opposed  to  each  other  always  depend  on 
one  and  the  same  science. 

Propositions  result  from  the  union  of  concepts. 
They  are  affirmative  or  negative,  universal  or  particular. 
They  alone  admit  of  truth  or  error,  whereas  isolated 
concepts  are  neither  true  nor  false.  The  result  is  not 
the  same,  when  two  judgments  are  contradictory  to 
each  other,  as  when  they  are  simply  contrary.  Two 
contrary  judgments  cannot  both  be  true,  though  they 
may  be  false,  whereas  one  of  two  contradictory  judg- 
ments is  of  necessity  true  and  the  other  false  :  this 
results  from  the  principle  of  excluded  middle,  a  par- 
ticular expression  of  the  principle  of  contradiction. 

Propositions  admit  of  conversions  or  inversions  of 
subject  and  predicate,  the  rules  of  which  are  determined 
by  Aristotle. 

Reasoning  consists  essentially  of  syllogism.  The 
theory  of  syllogism  and  of  demonstration,  or  perfect 
syllogism,  is  called  by  Aristotle  analytics.  Aristotle 
claims  to  have  invented  it.  He  affirms  that,  previous 
to  his  time,  there  existed  nothing  on  this  subject,  that 
he  had  not  merely  to  improve  but  to  invent,  and  that 
he  attained  his  end  by  dint  of  laborious  attempts. 
Kant  said  regarding  the  theory  of  the  syllogism  that, 
ever  since  the  days  of  Aristotle,  it  had  not  moved  a 
step,  either  backwards  or  forwards. 

The  syllogism  is  a  process  of  reasoning  in  which, 


ARISTOTLE  95 

certain  things  being  posited,  something  different 
necessarily  results.  The  property  of  the  syllogism  is 
that  it  makes  evident  the  necessity  of  the  conjunction. 
This  result  is  obtained  by  the  use  of  elements  adapted 
to  an  exact  application  of  the  principle  of  contradiction. 
These  elements  are  terms  regarded  as  holding  to  each 
other  the  relation  of  the  part  to  the  whole.  Granted 
that  A  contains  B  and  B  contains  C,  it  necessarily 
follows,  in  accordance  with  the  principle  of  contradiction, 
that  A  contains  C.  This  is  the  type  of  the  syllogism, 
and  the  three  terms  it  implies  are  therefore  called 
major,  middle  and  minor.  This  relation  of  extent  is 
regarded  by  Aristotle  as  equivalent  to  the  relation 
between  general  and  particular.  The  genus  is  a  kind 
of  definite  circle,  containing  the  various  species. 

The  syllogism  is  perfect  or  imperfect,  according  as 
it  conforms  immediately  to  the  type  we  have  just  in- 
dicated, or  becomes  conformable  thereto  only  by  the 
aid  of  transformations  or  reductions. 

The  origin  of  this  theory  may  be  found  in  mathe- 
matics. It  consists  in  an  adaptation  to  the  qualitative 
notions  of  the  relations  of  dimension.  It  was  natural 
that  Aristotle  should  seek,  in  an  analogical  imitation  of 
mathematics,  for  the  means  of  demonstrating  necessarily 
in  qualitative  matter  ;  since  it  was  acknowledged  by  all 
that  mathematics  realised  that  necessity  in  the  con- 
catenation of  the  terms,  which  he  had  in  view.  In  the 
syllogism  the  instrument  of  necessary  connection  is  the 
middle  term. 

Of  the  particular  cases  of  syllogism,  the  most  im- 
portant is  induction,  or  the  reasoning  which  proceeds 
from  particular  to  general.  The  following  is  an  in- 
stance of  this  reasoning  :  "  The  man,  the  horse  and  the 
mule  live  long.  Now  the  man,  the  horse  and  the  mule  are 


96  STUDIES  IN  PHILOSOPHY 

animals  devoid  of  gall.  Therefore,  all  animals  devoid 
of  gall  live  long."  The  condition  of  the  legitimacy  of 
the  conclusion  lies  in  the  convertibility  of  the  minor 
premise.  Here,  for  instance,  for  the  proposition  : 
"  The  man,  the  horse  and  the  mule  are  animals  devoid 
of  gall,"  we  should  have  to  be  permitted  to  substitute  : 
"  All  the  animals  devoid  of  gall  are  man,  the  horse  and 
the  mule."  The  legitimacy  of  this  substitution  is  no 
longer  a  matter  of  logic.  In  fact,  the  series  of  animals 
devoid  of  gall  is  an  infinite  one.  But  the  essence  of 
the  animal  devoid  of  gall  is  entirely  in  each  animal 
devoid  of  gall.  The  question  is  to  discern  this 
essence,  to  find  the  type  of  the  animal  devoid  of  gall, 
so  as  to  distinguish  the  characteristics  belonging  to 
animals  devoid  of  gall,  in  this  particular  condition  of 
being  devoid  of  gall,  and  separate  them  from  the 
characteristics  belonging  to  them  independently  of  this 
condition.  To  effect  this,  we  consider  a  certain  number 
of  animals  devoid  of  gall,  compare  them  with  one 
another,  find  out  what  they  have  in  common,  and  so 
what  there  is  in  them  that  is  essential  and  necessary. 
In  other  words,  we  consider  the  beings  of  nature  not 
only  with  the  senses,  but  with  the  vovs — the  seat  of  the 
essences — which  is  capable  of  finding  and  recognising 
them  in  the  data  of  the  senses. 

Aristotle's  induction  thus  aims  at  the  classification 
of  beings  and  facts,  and  also  at  a  natural  classification. 
In  so  far  as  it  is  applied  in  distinguishing  necessary 
relations  from  contingent  ones,  it  makes  prediction 
possible,  and  thus  supplies  us  with  true  laws,  in  the 
modern  sense  of  the  word.  This  possibility  of  pre- 
diction, however,  is  restricted  to  the  facts  that  proceed 
immediately  from  a  determined  essence  ;  it  does  not 
extend  to  the  facts  that  result  from  the  mingling  of 


ARISTOTLE  97 

several  essences.  There  is  no  necessary  reason  for 
the  mingling  of  the  essences  ;  this  is  something  purely 
contingent.  The  genera,  according  to  Aristotle,  are 
radically  separated  from  one  another  ;  each  of  them  is 
an  absolute.  In  this  doctrine  of  the  independence  of 
genera,  the  Aristotelian  theory  of  induction  is  opposed 
both  to  Cartesianism,  which  reduces  physical  laws  to 
mathematical  determinations,  the  heterogeneous  to  the 
homogeneous ;  and  also  to  evolutionism,  which  re- 
cognises the  present  existence  of  species,  though 
attributing  to  them  a  natural  genesis  in  the  past, 
starting  from  one  common  origin. 

Syllogism,  properly  so  called,  and  induction  are 
to  each  other,  says  Aristotle,  as  the  order  of  nature 
and  that  of  human  knowledge.  In  itself,  syllogism  is 
the  more  intelligible  ;  to  us,  induction  is  the  more 
distinct.  A  syllogism  starts  from  the  general.  Now, 
it  is  impossible  for  us  to  have  knowledge  of  the 
general  except  by  induction.  Not  that  general  prin- 
ciples rest  on  sensation  and  induction  as  their  founda- 
tion ;  it  is  rather  that  induction  discovers  these  principles 
for  us  and  supplies  us  with  the  intelligible  elements 
which  the  i/oO?  acknowledges  to  be  both  necessary  and 
true. 

Such  are  the  instruments  of  science.  How,  by 
means  of  them,  is  science  formed  ? 

Science  is  the  knowledge  of  things  in  so  far  as  they 
are  necessary.  A  thing  is  known  scientifically  when 
we  know  that  it  could  not  be  otherwise  than  it  is. 
Now,  this  knowledge  is  realised  when  we  succeed  in 
connecting  the  given  thing  with  its  cause. 

In  nature  there  are  three  kinds  of  connections  :  ist, 
conjunctions  that  are  always  realised,  for  instance  :  the 

H 


9  8  STUDIES  IN  PHILOSOPHY 

relations  of  astronomical  phenomena  ;  2nd,  conjunctions 
that  are  usually  realised,  for  instance  :  the  relations  of 
physical  things  to  one  another,  and,  even  more  so,  of 
moral  things  ;  3rd,  chance,  i.e.  the  coincidences  that  are 
but  seldom,  or  never,  reproduced.  The  first  kind  of 
connection  admits  of  perfect  science  ;  the  second,  of 
imperfect  science,  limited  to  probability  ;  the  third  is 
outside  of  the  domain  of  science.  There  is  no  science 
of  what  is  passing  away. 

Neither  opinion  nor  sensation  can  produce  science, 
for  as  they  are  both  incapable  of  perfect  determination 
and  finity,  they  cannot  grasp  the  finite  and  immovable. 
Platonic  dialectic,  too,  is  powerless  to  afford  us  science, 
for  as  it  consists  of  questions  and  answers,  it  relies  only 
on  the  consent  of  the  opponent,  not  on  truth  in  itself. 
Starting  from  hypothesis,  it  does  not  go  beyond  a 
purely  formal  and  logical  inference.  It  is  by  demon- 
stration that  we  arrive  at  science.  Apodeictic,  or  the 
theory  of  demonstration,  differs  essentially  from 
dialectic. 

Demonstration  is  effected  by  direct  syllogism  of  the 
first  figure.  Reductio  ad  absurdum  and  syllogisms  of 
the  second  and  third  figures  are  not  yet  demonstration, 
which  has  its  starting-point  in  a  principle  that  is  not 
only  granted  by  the  opponent,  but  is  necessary  in 
itself.  This  is  how  the  mathematician  reasons. 

Demonstration  comprises  three  elements  :  ist,  the 
subject ;  2nd,  the  predicate,  which  has  to  be  linked  to 
the  subject  by  a  bond  of  necessity  ;  3rd,  the  general 
principles  on  which  demonstration  is  based.  These 
latter  are  the  principle  of  contradiction  and  its  de- 
rivatives. Though  indispensable,  they  are  empty 
and  insufficient  in  themselves.  It  is  in  the  nature 
of  the  subject  that  the  basis  of  demonstration  lies. 


ARISTOTLE  99 

There  are,  in  effect,  principles  proper  to  the  subject, 
for  instance  the  continuous,  inherent  in  extent ;  and 
the  discontinuous,  inherent  in  number  :  it  is  these 
special  principles  that  have  a  content  and  are  productive. 
On  these  principles  it  is  good  to  rely,  and,  in  deduction, 
we  should  never  pass  from  one  genus  to  another  unless 
the  one  is  properly  subordinated  to  the  other.  Thus, 
geometry  could  not  be  explained  by  arithmetic ;  it 
would  be  impossible  to  adapt  to  dimensions  of  ex- 
tension demonstrations  proper  to  number.  When 
this  rule  is  violated,  we  have  for  our  guidance  none  but 
the  principles  common  to  all  sciences  ;  and  so  the  con- 
nections established  are  known  only  as  accidental  and 
contingent,  not  as  essential  and  necessary  :  we  have 
been  proceeding  by  analogy,  not  by  demonstration.  The 
impossibility  here  seen  by  Aristotle  was  at  a  later  date 
removed  by  Descartes  and  Leibnitz. 

Proper  principles  cannot  be  proved  like  common 
ones.  To  claim  to  demonstrate  everything  would  be 
to  condemn  oneself  either  to  progress  ad  infinitum  or 
to  the  argument  in  a  circle.  Thus  each  science  has  its 
special  irreducible  principles. 

Whence  come  these  principles?  They  are  neither 
innate,  nor  received  purely  and  simply  from  without. 
There  is  within  us  a  disposition  to  conceive  them  ; 
and,  as  the  result  of  experience,  this  disposition  passes 
into  action.  It  is  in  this,  after  all,  that  induction  con- 
sists, and  so  it  is  by  induction  that  we  know  the  first 
principles  proper  to  each  science. 

Demonstration  implies  definition.  There  must  be 
undemonstrable  definitions  :  otherwise  we  should  pro- 
ceed ad  infinitum.  There  is  no  definition,  either  of  the 
individual  or  of  the  accident — or  the  indeterminate 
general — but  only  of  intermediate  species  between  the 


ioo  STUDIES  IN  PHILOSOPHY 

general  and  the  individual.  Definition  is  effected  by 
indicating  the  next  genus  and  the  specific  differences. 
In  order  to  constitute  a  definition,  we  must  proceed 
from  the  particular  to  the  general  and  verify  this  induc- 
tion by  a  deduction  proceeding  from  genus  to  species. 

To  sum  up,  a  thing  is  known  as  necessary  when 
connected,  by  deduction,  with  a  specific  essence. 

Below  apodeictic,  which  teaches  how  one  comes  to 
know  a  thing  as  necessary,  stands  dialectic,  or  the  logic 
of  the  probable  :  we  find  it  set  forth  in  the  Topics. 
The  domain  of  dialectic  is  opinion,  a  mode  of  know- 
ledge admitting  of  truth  or  of  falsehood.  The 
dialectician  takes,  as  his  starting-point,  not  definitions 
necessary  in  themselves,  but  opinions  or  theses  pro- 
pounded either  by  philosophers  or  by  common  sense  ; 
he  tries  to  discover  which  of  these  divers  opinions  is 
the  most  probable.  Proceeding  by  means  of  questions 
and  answers,  he  contradictorily  examines  the  yes  and 
the  no  regarding  each  subject.  Thus,  he  arranges  his 
questions  in  such  fashion  as  to  present  first  a  thesis, 
then  an  antithesis  ;  afterwards  he  discusses  both  pro- 
positions. This  discussion  consists  in  examining  the 
difficulties  that  arise,  when  we  wish  to  apply  the  pro- 
position to  particular  instances.  The  dialectician 
reasons  syllogistically,  though  he  starts  with  the  pro- 
bable. The  probable,  taken  as  the  given,  is  in  reality 
the  purely  generic  essence,  not  yet  determined  by  the 
specific  difference.  Only  by  the  addition  of  the  specific 
principle  to  the  generic  principle  could  the  conclusion 
be  made  necessary.  The  specific  principles,  however, 
cannot  be  deduced  from  the  generic  ones,  for  every 
genus  admits  alike  of  different  species. 

The  rble  of  dialectic  is  important  :  it  is  the  only 
possible  mode  of  reasoning  in  things  which  do  not 


ARISTOTLE  101 

admit  of  necessary  definitions.  And  in  the  search  after 
necessary  truths  themselves,  dialectic  is  the  indispensable 
preliminary  of  demonstration. 

What  dialectic  is  in  logic,  rhetoric  is  in  morals. 
If  the  former  seeks  after  the  probable,  the  object  of 
the  latter  is  to  commend  it  to  acceptance.  And  so 
rhetoric  and  dialectic  go  well  together,  or  rather,  as 
practice  is  to  theory  what  the  particular  is  to  the  general, 
so  rhetoric  is  a  part  of  dialectic.  The  mode  of  reason- 
ing proper  to  rhetoric  is  the  enthymeme,  a  syllogism  in 
which  one  of  the  three  propositions  is  left  unexpressed, 
and  the  reasons  are  not  obtained  from  the  essence  of 
things,  but  from  probabilities  and  signs.  The  main 
element  of  the  enthymeme  that  rhetoric  uses,  is  analogy, 
or  the  induction  which  proceeds  from  particular  to 
particular. 

Finally,  a  distinction  must  be  made  between  dialectic 
and  eristic.  Whereas  the  former  has  to  deal  with 
things  that  are  general  and  ordinary,  without  being 
necessary,  the  latter  deals  with  pure  accident,  and  that 
deliberately.  Eristic  contents  itself  with  a  probability 
that  is  accepted  by  the  hearer ;  consequently  eristic 
reasonings  are  pure  sophisms.  Aristotle  minutely 
exposes  and  describes  them. 

Below  things  that  always  happen,  which  depend  on 
an  essence  both  generic  and  specific  and  are  capable  of 
being  known  as  necessary,  even  below  things  that 
usually  happen,  which  depend  on  a  simply  generic 
essence  and  are  capable  of  being  known  as  probable, 
there  are  those  that  happen  accidentally,  apart  from 
any  rule  at  all.  As  things  that  usually  happen  result 
from  the  mingling  of  species,  so  isolated  phenomena 
result  from  the  mingling  of  genera  ;  but  whereas  that 
which  is  not  determinable  by  species  is  still  determin- 


102  STUDIES  IN  PHILOSOPHY 

able,  to  some  extent,  by  genus,  the  common  substratum 
of  several  species,  that  which  is  not  even  determinable 
by  genus  is  no  longer  determinable  at  all,  since,  above 
genera,  there  are  none  but  universal  principles,  which, 
as  they  apply  to  everything,  determine  nothing.  There 
is  no  science,  then,  of  hazard,  as  such,  the  meet- 
ing with  the  two  genera.  Only  the  elements  of  which 
the  fortuitous  phenomenon  consists  can  be  known  as 
necessary  or  possible,  in  so  far  as  they  are  connected 
with  their  respective  specific  or  generic  essences  :  the 
union  of  these  elements,  which,  properly  speaking, 
constitutes  the  fortuitous  phenomenon,  is  without 
reason,  because  genera,  as  such,  are  without  mutual 
connection. 

Aristotelian  logic  held  undisputed  sway  down  to  the 
time  of  Bacon  and  Descartes.  From  the  beginning  of 
modern  philosophy,  it  has  been  attacked  and  battered 
on  every  side  ;  either  reproached  for  being  the  logic  of 
exposition,  and  not  that  of  invention,  or  else  regarded 
as  artificial  and  illegitimate.  Discussion  bears  mainly 
on  the  value  of  the  concept  or  general  idea,  the  basis  of 
the  theory.  The  empirics,  in  particular,  to  whom  ideas 
are  only  traces  of  sensation,  estimate  the  value  of 
generalities  by  the  number  of  ascertained  facts  they 
represent ;  they  maintain  that,  as  the  truth  of  the 
major  premise  of  a  syllogism  implies  that  of  the  con- 
clusion, the  syllogism  is  necessarily  an  argument  in  a 
circle. 

Here  the  thing  to  discover  is  whether  a  concept 
is  anything  else  than  a  collective  idea,  or  a  static  or 
dynamic  unity,  valid  for  an  indefinite  series  of  past, 
present  and  future  facts.  But  even  should  the  Aris- 
totelian concept  not  exactly  coincide  with  the  nature 
of  things,  as  would  be  the  case  were  continuity  the 


ARISTOTLE  103 

fundamental  law  of  being,  the  logic  of  Aristotle  would 
none  the  less  retain  real  value.  Not  only  would  it 
subsist  as  an  analysis  of  the  conditions  of  ideal  know- 
ledge for  the  human  mind,  but  it  would  be  legitimate 
in  proportion  as  there  exist  species  in  nature.  Now, 
these  do  exist,  if  not  in  an  eternal,  primitive  fashion, 
perhaps,  at  all  events  in  actuality  and  at  the  present 
time.  Superior  beings,  especially,  form  relatively  stable 
groups.  Even  though  continuity  were  the  fundamental 
law,  none  the  less  would  it  be  necessary  to  recognise  in 
nature  a  tendency  to  discontinuity  and  specification. 
Aristotelian  logic  would  answer  to  that  part  or  side  of 
nature  which  is  governed  by  the  law  of  specification. 
Deprived  of  the  metaphysical  and  absolute  value  its 
founder  attributed  to  it,  it  would  retain  a  relative  and 
experimental  value. 

VIII. — METAPHYSICS 

Whereas  each  science  considers  some  particular 
species  of  beings  ;  physics,  for  instance,  considers  being 
in  so  far  as  it  has  matter  and  motion  ;  mathematics,  the 
form  of  mobile  being  in  so  far  as  it  is  isolated  by 
abstraction  from  the  matter  in  which  it  is  realised  ; 
first  philosophy,  as  Aristotle  calls  it,  considers  being,  in 
so  far  as  it  is  being,  TO  ov,  y  ov,  and  in  this  way  tries  to 
discover  its  principles. 

Aristotelian  metaphysics  has  been  set  up  as  opposed  to 
Platonic  philosophy.  Thus,  we  find  Aristotle  beginning 
his  exposition  by  a  criticism  of  his  master.  Plato,  he 
says,  seeks  both  the  object  of  science  and  being,  in 
so  far  as  it  is  being,  in  the  general  essences  conceived 
of  as  existing  apart,  outside  of  things  and  also  outside 
of  one  another.  Now,  here,  the  true  is  confounded  with 


io4  STUDIES  IN  PHILOSOPHY 

the  false.  Plato  clearly  saw  that  the  general  alone  can 
be  an  object  of  science  and  that  the  sensible  world  as 
such  cannot,  therefore,  be  known  scientifically.  But  he 
was  mistaken  in  thinking  that  genera  can  exist  apart,  that 
they  are  themselves  principles  and  substances.  Genera 
•exist  only  in  individuals.  We  get  entangled  in  in- 
extricable difficulties  if  we  insist  that  they  exist  per  se. 
Under  this  hypothesis,  what  will  be  the  relation  of 
things  to  their  respective  genera  ?  Will  it  be  one 
of  participation  ?  Then  how  can  this  participation 
be  conceived  ?  Besides,  how  many  substantial  genera 
will  there  be  ?  How  can  the  idea,  the  one  substance, 
be  met  with  in  an  infinite  number  of  individuals  ?  If 
the  general  idea  is  substance,  either  there  are  no 
individuals  or  there  is  only  one.  In  addition,  the 
general  cannot  be  principle  and  substance,  because  it 
is  devoid  of  force  and  cannot  exist  per  se.  The  general 
is  always  an  attribute,  a  predicate  :  substance,  on  the 
other  hand,  is  a  subject,  a  thing  existing  apart.  There- 
fore, it  is  quite  true  that  the  general  alone  is  an  object 
of  science  ;  substance,  on  the  contrary,  can  be  only 
individual. 

Here,  however,  a  difficulty  arises.  If,  on  the  one 
hand,  all  science  rests  on  the  general,  and,  on  the  other, 
substance  can  be  only  something  individual,  how  can 
there  be  a  science  of  substance  ?  Does  not  our  theory 
end  in  the  following  result  :  a  science  whose  object  is 
not  in  being  ;  a  being  which  cannot  be  an  object  of 
science  ? 

To  solve  this  difficulty,  we  must  enlarge  our  notion 
of  science.  All  science  does  not  rest  on  the  general. 
Science  has  two  modes,  two  degrees.  There  is  science 
in  potency  and  there  is  science  in  act.  The  former  has 
the  general  for  its  object,  but  it  is  not  so  with  the 


ARISTOTLE  105 

latter,  which  has  for  its  object  the  perfectly  determined 
being,  the  individual. 

In  this  doctrine  we  find  the  central  idea  of  Aris- 
totelianism.  The  general  is  not  the  constitutive  prin- 
ciple of  being,  it  is  nothing  but  the  matter  thereof. 
Though  determined  in  one  direction,  it  is  indeterminate 
in  another  :  every  general  type  may  be  realised  in 
divers  ways.  A  real  being,  a  substance,  is  a  completed 
being,  which,  in  every  respect,  is  this  and  not  that  : 
consequently,  in  any  real  being  whatsoever,  there  is 
something  more  than  in  any  general  idea.  The  entire 
science  of  the  general  could  not  build  up  the  in- 
dividuality of  Socrates.  There  are  necessarily  two 
things  outside  of  this  abstract  science  :  accidents,  be- 
cause they  are  below  the  general ;  individuals,  because 
they  are  above  it.  The  knowledge  of  individuals  is 
effected  by  an  intuition,  which,  immediately,  grasps  the 
x  substantial  unity  that  could  not  be  deduced. 

This  irreducibleness  of  the  individual  to  the  general 
will  be  seen  through  the  philosophy  of  Aristotle.  By 
virtue  of  this  principle,  abstract  speculation  will  be  power- 
less to  enable  us  to  know  nature  ;  to  do  this,  experience 
will  be  necessary.  And  in  the  moral  order  of  things,  laws 
will  be  inadequate  to  bring  about  the  reign  of  justice  ; 
the  magistrate  must  be  brought  in,  empowered  by  law 
to  apply  general  rules  to  the  endless  diversity  of 
individual  cases. 

What  are  the  principles  of  being  ?  Being,  which  is 
given  to  us,  is  subject  to  a  process  of  becoming.  Now, 
becoming,  in  so  far  as  it  exists,  implies  principles  that 
have  not  been  generated  :  a  halt  must  necessarily  be 
made  in  the  retrogression  towards  causes,  when  we 
have  to  find  out  what  are  the  integral  elements  of 
'present  existence. 


io6  STUDIES  IN  PHILOSOPHY 

What  are  the  principles  required  in  order  to  explain 
becoming  ?  They  are  four  in  number  :  ist,  a  sub- 
stratum or  matter,  the  scene  of  change,  i.e.  of  the 
substitution  of  one  mode  of  being  for  another  ;  2nd, 
a  form  of  determination  ;  3rd,  a  motor  cause  ;  4th,  an 
aim  or  end.  For  instance,  the  principles  of  a  house  are  : 
the  timber,  as  matter  or  material,  the  idea  of  the  house, 
as  form,  the  architect,  as  motor  cause,  and  the  dwelling 
to  be  realised,  as  object  or  end. 

These  four  principles,  in  turn,  may  be  reduced  to 
two  :  matter  and  form.  In  fact,  the  motor  cause  is 
nothing  more  than  form  in  an  already  realised  subject  : 
thus,  the  motor  cause  of  the  house  is  the  idea  of  the 
house,  as  conceived  by  the  architect.  And  the  final 
cause  is  also  the  form,  for  the  final  cause  of  each  single 
thing  really  consists  of  the  perfection  or  form  towards 
which  it  is  tending. 

And  so  matter  and  form  are  definitely  the  two  non- 
generated  principles  that  are  necessary  and  sufficient  to 
explain  becoming.  Matter  is  the  substratum.  It  is 
neither  this  nor  that ;  it  is  capable  of  becoming  this  or 
that.  Form  is  that  which  makes  of  matter  a  deter- 
minate (roSe  rt)  and  real  thing.  It  is  the  perfection, 
activity  or  soul  of  the  thing.  As  Aristotle  interprets 
it,  the  word  form  has  quite  a  different  meaning  from 
ours.  For  instance,  in  Aristotle's  phraseology,  a 
sculptured  hand  possesses  the  figure  and  not  the  form 
of  a  hand,  because  it  cannot  perform  such  functions  as 
are  proper  to  the  hand. 

There  is  a  scale  of  existences  from  lowest  matter, 
which  has  no  form  at  all,  to  highest  form,  which  is 
devoid  of  matter.  Absolutely  indeterminate  matter  is 
non-existent.  Form  without  matter  is  outside  of 
nature.  All  the  beings  of  nature  are  compounds  of 


ARISTOTLE  107 

matter  and  form.  The  opposition  of  matter  and  form 
is  relative.  That  which  is  matter  from  one  point  of 
view  is  form  from  another.  Timber  is  matter  in 
relation  to  the  house,  and  form  in  relation  to  uncut 
wood.  The  soul  is  form  with  regard  to  the  body, 
matter  the  intelligence. 

Aristotle  does  not  content  himself  with  this  reduction 
of  the  four  principles  to  matter  and  form  ;  he  attempts 
to  bring  together  these  two  principles  themselves.  To 
effect  this,  he  brings  them  within  the  scope  of  potency 
and  act.  For  him,  matter  is  not  mere  receptivity,  as  it 
is  for  Plato  :  it  has  a  propensity  to  receive  form,  it 
desires  form.  The  latter  is  not  something  hetero- 
geneous as  regards  matter  :  it  is  its  natural  completion. 
Matter  is  potency,  potency  that  is  capable  of  two 
determinate  contraries.  The  logical  mechanism  of  the 
substitution  of  forms  in  inert  matter  thus  resolves  itself 
into  a  metaphysical  dynamism.  There  is  an  inner 
action  in  the  transition  from  potency  to  act.  This  is 
no  longer  a  juxtaposition  or  separation  of  inert,  pre- 
existing elements  :  it  is  a  spontaneous  creation  of  being 
and  perfection.  If  a  force  of  determinate  quantity, 
says  Aristotle,  is  needed  to  produce  a  certain  effect,  the 
half  of  this  force,  applied  separately,  does  not  produce 
this  effect  at  all.  Were  it  not  so,  given  a  ship  which 
several  men,  with  united  effort,  set  in  motion,  a  single 
man  would  be  able  to  communicate  a  certain  amount 
of  motion  to  the  ship,  but  this,  as  we  know,  is  contrary 
to  experience.  Any  particular  part,  which  produces 
motion  when  united  with  the  whole,  if  taken  separately 
and  acting  alone,  becomes  altogether  powerless.  Truth 
to  tell,  the  part  has  no  existence  as  a  part  in  what  is 
really  a  whole  :  a  part  exists  only  potentially  in  the 
whole  from  which  it  may  be  taken. 


io8  STUDIES  IN  PHILOSOPHY 

As  we  see,  the  Aristotelian  concept  of  potency  and 
act  is  very  empirical.  Aristotle  takes  it  for  granted 
that  the  effort  of  a  single  man  produces  no  result  on  a 
ship,  because  he  does  not  know  that  the  work  which 
does  not  become  manifest  in  the  form  of  movement,  at 
all  events,  generates  heat.  None  the  less  is  the  push  of 
a  single  man  really  ineffective,  so  far  as  actual  removal 
from  one  place  to  another  is  concerned.  Even  at  the 
present  time,  there  is  a  school  of  chemists  who  reason 
like  Aristotle,  and  do  not  regard  hydrogen  and  oxygen 
as  existing  in  water  in  act,  but,  relying  on  experience, 
these  scientists  say  that  hydrogen  and  oxygen  exist  in 
water  in  potency,  in  this  sense  that,  if  water  is  subjected 
to  certain  conditions,  hydrogen  and  oxygen  may  be 
obtained. 

To  sum  up,  becoming,  according  to  Aristotle,  origin- 
ates neither  in  absolute  being  nor  in  absolute  non-being ; 
it  originates  in  being  in  potency,  midway  between  being 
and  non-being. 

From  this  being  in  potency,  or  matter,  proceeds  all 
that  is  indetermination  and  imperfection  in  the  world. 
Matter  is  the  principle  of  brute  necessity  or  dvdy/cr), 
which  is  mechanical  and  blind  causality,  in  contrast  with 
the  motor  cause  which  acts  with  a  view  to  an  end.  If 
such  necessity  exists,  it  is  because  nature  is  compelled 
to  employ  material  causes  in  its  creations.  Now,  in  a 
sense,  matter  is  resistant  to  form.  That  is  why  the 
creations  of  nature  are  invariably  imperfect ;  there  are 
even  produced  many  things  that  are  devoid  of  purpose, 
in  so  far  as  they  come  into  being  by  the  action  of 
mechanical  forces  only.  Slaves,  for  instance,  whose 
actions  are  regulated  often,  nevertheless,  act  on  their 
own  account,  quite  apart  from  regulations.  Matter 
is  the  principle  of  the  contingence  of  future  events. 


ARISTOTLE  109 

As  regards  the  future,  the  position  of  a  determinate 
alternative  is  alone  necessary  :  the  realisation  of  either 
term  of  this  alternative  is  indeterminate.  From  matter 
proceeds  hazard.  In  any  given  being,  those  phenomena 
are  fortuitous  which  do  not  spring  from  the  essence  of 
that  being,  but  are  the  result  either  of  its  imperfection 
or  of  the  influx  of  extraneous  causes.  Hazard  manifests 
itself  by  the  rarity  of  the  event.  The  fortuitous  event 
is  mechanically  necessary,  though  necessary  only  from 
this  point  of  view  :  in  relation  to  finality  it  is  indeter- 
minable and  uncognisable.  Matter  is  the  cause  of  the 
imperfection  of  beings  as  well  as  the  cause  of  evil. 
It  is  likewise  the  cause  of  the  multiplicity  of  species, 
for,  in  all  their  infinite  variety,  the  beings  of  nature 
are  only  more  or  less  complete  realisations  of  one  and 
the  same  type  supplied  by  the  form.  Animals  are  only 
incomplete  men,  arrested  at  a  certain  stage  of  their 
natural  development.  From  the  presence  of  matter  in 
natural  things,  it  follows  that  these  things  cannot  be  the 
object  of  perfect  science,  i.e.  they  cannot  be  known  as 
fully  determined.  In  itself,  the  material  element  of 
things  does  not  admit  of  science. 

Such  are  the  proximate  causes  of  being  when  sub- 
mitted to  a  process  of  becoming.  We  could  not  have 
a  full  explanation  of  this  being,  however,  were  we  to 
confine  ourselves  to  a  consideration  of  its  elements. 
Being  in  process  of  becoming  finds  its  ultimate  explana- 
tion only  in  an  eternal  being. 

The  existence  of  God  is  already  proved,  in  a  popular 
way,  by  the  gradual  perfection  of  beings  and  the  finality 
that  reigns  throughout  nature.  Scientifically  it  is  proved 
by  the  analysis  of  the  conditions  of  motion.  This  is 
what  is  called  the  argument  of  the  prime  mover. 


no  STUDIES  IN  PHILOSOPHY 

Motion  is  change,  the  relation  of  matter  to  form. 
In  this  sense,  the  motion  of  the  world  is  eternal.  In- 
deed, time  is  necessarily  eternal ;  now,  without  motion  or 
change,  time  could  not  exist.  But  motion  implies  both 
something  movable  and  a  moving  principle.  Motion, 
then,  in  so  far  as  it  is  eternal,  presupposes  something 
eternally  movable  and  an  immovable  first  mover. 
The  "  eternally  movable "  moves  in  a  circle  ;  this  is 
the  first  heaven,  the  heaven  of  the  fixed  stars.  The 
immovable  first  mover  is  what  men  call  God. 

This  proof  may  be  generalised  as  follows.  The 
actual  is  always  previous  to  the  potential.  The  first, 
in  the  absolute,  is  not  the  germ,  but  rather  the  com- 
pleted being.  Besides,  actuation  could  not  take  place 
were  not  pure  act  already  in  existence.  God  is  this 
pure  act. 

In  a  word,  demonstration  of  the  existence  of  God 
is  based  on  the  following  dual  principle  :  ist,  act,  from 
the  point  of  view  of  the  absolute  nature  of  things,  is 
anterior  to  potency  ;  2nd,  the  conditioned  presupposes 
the  unconditioned. 

What  is  God?  His  nature  is  determined  by  his 
rtle  as  first  mover.  God  is  pure  act,  i.e.  he  is  exempt 
from  indetermination,  imperfection  and  change.  He  is 
both  immovable  and  immutable.  He  is  thought  which 
has  thought — and  nothing  else — as  its  object  (rj  vorja-is 
z/oTjcreo)?  vori<n<^.  He  sees  not  the  world,  for  when 
we  are  dealing  with  imperfect  things,  not  to  see 
them  is  better  than  to  see  them  :  the  dignity  of  an 
intelligence  is  gauged  by  the  perfection  of  its  object. 
He  is  eternal,  all-excelling  life,  and  therefore  supremely 
happy. 

To  this  thought  which  thinks  itself  the  world  is 
suspended,  as  a  thought  which  does  not  think  itself 


ARISTOTLE  1 1 1 

and  tends  to  do  so.  This  is  how  God  moves  the 
world.  What  is  desired  and  thought  moves  without 
oneself  moving.  It  is  the  intelligible  that  determines 
intelligence,  not  intelligence  that  determines  the  intel- 
ligible. Now,  God  is  the  supremely  desirable  and  the 
supremely  intelligible.  God,  therefore,  moves  the 
world  as  final  cause,  without  himself  moving.  God  is 
not  the  ultimate  product  of  the  world's  development ; 
logically,  he  is  anterior  to  the  world.  Nor  is  he 
immanent  in  the  world,  as  order  is  immanent  in  an 
army  :  he  is  out  of  the  world,  as  a  general  is  distinct 
from  his  army. 

The  immediate  effect  of  divine  action  is  the  rotatory 
motion  of  the  whole  universe,  which  gives  rise  to 
the  motions  or  changes  of  perishable  things.  The 
world  is  one,  because  God  is  one.  Because  God  is 
intelligent,  the  world  is  a  harmonious  whole,  a  well- 
composed  poem.  Everything  therein  is  arranged  with 
a  view  to  a  single  end.  The  relation  of  the  various 
beings  to  the  whole  is  all  the  closer  from  the  fact  that 
these  beings  are  higher  in  the  scale  of  nature  ;  just  as,  in 
a  well-ordered  house,  the  actions  of  free  men  are  more 
regulated  than  those  of  slaves.  God,  moreover,  to 
whom  the  world  is  as  though  it  did  not  exist,  intervenes 
in  no  single  detail  of  his  own  events. 

This  theology  is  an  abstract  monotheism.  All  the 
beings  and  facts  of  nature  are  wholly  referred  to  natural 
causes.  It  is  only  nature,  regarded  in  its  entirety,  that 
is  made  contingent  on  divinity.  There  is  neither  special 
providence  nor  supernatural  reward  in  another  life.  The 
only  thing  in  popular  religion  that  Aristotle  admits  to 
be  true  is  the  general  belief  in  divinity  and  in  the 
divine  nature  of  the  sky  and  the  stars.  To  his  mind, 
the  rest  consists  of  nothing  but  mythical  additions,  the 


ii2  STUDIES  IN  PHILOSOPHY 

explanation  of  which  a  philosopher  finds  either  in  the 
tendency  of  men  towards  anthropomorphic  conceptions, 
or  in  the  calculations  of  politicians. 

IX. — GENERAL  PHYSICS 

The  object  of  first  philosophy  was  immovable  and 
incorporeal  being  ;  the  object  of  physics,  or  second 
philosophy,  is  movable  and  corporeal  being,  in  so  far 
as  the  latter  has  within  itself  the  principle  of  its  motion. 
<&vo-i,<s  is  spontaneous  motion,  in  opposition  to  that  which 
results  from  compulsion. 

Does  nature  exist  as  such  ?  Is  there,  in  the  universe, 
an  internal  principle  of  motion,  a  tendency  to  an  end  ? 

According  to  Aristotle,  it  is  the  fundamental  principle 
of  physics  that  God  and  nature  do  nothing  in  vain  ;  that 
nature  always  tends  towards  something  better  ;  that,  as 
far  as  possible,  it  always  brings  to  pass  what  is  to  be 
the  most  beautiful.  The  existence  of  finality  in  the 
universe  is  proved  by  observation.  In  the  smallest  as 
in  the  largest  things,  if  we  take  notice,  we  find  there 
is  reason,  perfection,  divinity.  Nature  converts  even 
its  own  imperfections  to  good. 

But  if  order  and  harmony  exist  throughout  the 
universe,  does  it  follow  that  the  universe  is  the  product 
of  a  <£vo-£9  properly  so  called,  a  divine  creative  power  ? 
Is  not  there  some  other  possible  explanation  of  this 
order  and  harmony  ?  Why,  for  instance,  should  we 
not  say  :  Jupiter  does  not  send  the  rain  in  order  to 
make  the  corn  grow  ;  the  corn  grows  because  it  rains. 
Necessity  makes  the  rain  fall,  and  when  this  pheno- 
menon takes  place,  the  wheat  profits  thereby.  Necessity 
likewise  makes  the  organs  of  animals,  and  of  these  they 
make  use.  Whereas  everything  appears  to  take  place 


ARISTOTLE  113 

with  a  view  to  an  end,  it  is  really  only  things  that 
survive,  because  they  happen  to  have  been  constituted 
by  chance  so  as  to  conform  with  their  conditions  of 
existence.  And  those  things  which  did  not  happen  to 
be  constituted  in  that  way,  die  out,  and  have  always 
died  out,  as,  according  to  Empedocles,  happened  in 
the  case  of  the  oxen  with  human  faces. 

A  vain  explanation,  replies  Aristotle,  for  the  organs 
of  animals  and  the  majority  of  the  beings  with  which 
nature  brings  us  in  contact  are  what  they  are,  to  wit, 
harmonious  compounds,  either  in  every  case,  or  in  most, 
at  all  events.  Now,  it  is  never  so  with  things  produced 
by  chance ;  here,  fortunate  occurrences  are  never  any- 
thing but  exceptions. 

But,  we  shall  be  told,  there  exist  monsters. 
Monsters  are  but  incomplete  pieces  of  work,  the 
results  of  effort  which  is  incapable  of  realising  the 
perfect  type.  Nature,  as  well  as  art,  may  make 
mistakes,  by  reason  of  the  obstruction  which  the  very 
matter,  on  which  it  is  working,  sets  up  against  it. 

Finally,  will  the  objection  be  raised  that  we  do  not 
see  the  mover  deliberating  and  choosing  ?  That  matters 
little,  for  art  does  not  deliberate  either ;  it  acts 
intelligently,  without  giving  account  of  what  it  does. 

Nature,  then,  is  a  cause,  and  a  cause  that  acts 
with  a  view  to  an  end.  It  must,  however,  be  recog- 
nised that  it  is  not  the  only  cause  in  the  universe. 
Its  action  is  only  possible  owing  to  the  co-operation 
of  the  material  or  mechanical  cause,  which,  though 
yielding  to  its  attraction,  never  allows  itself  to  submit 
completely.  .  Along  with  finality,  then,  we  find  every- 
where throughout  the  universe,  a  certain  proportion  of 
brute  necessity  and  chance. 

This  explains  why,  on  the  one  hand,  the  principle 


n4  STUDIES  IN  PHILOSOPHY 

of  the  best  may  legitimately  be  employed  in  explaining 
the  things  of  nature  ;  though,  on  the  other  hand,  the 
things  of  nature  can  never  come  within  the  domain  of 
perfect  science,  wherein  everything  seems  wholly  deter- 
mined from  the  point  of  view  of  the  intelligence. 
The  science  of  nature  is  always  imperfect  in  some 
direction  ;  it  admits  of  degrees,  as  do  the  parts  of 
nature  itself.  In  accordance  with  these  principles,  the 
cause  of  natural  things  may  be  found,  either  in  their 
matter,  or  in  their  form  or  destination.  And,  as  far  as 
possible,  the  teleological  explanation  should  complete 
the  mechanical  one,  which,  however  finished  it  be, 
leaves  things  indeterminate  in  the  sight  of  reason. 
Such  is  the  method  Aristotle  is  to  pursue  in  his 
investigations  into  natural  things. 

Motion  or  change  is  the  realization  of  a  possible. 
There  are  four  kinds  of  change  :  ist,  substantial  change, 
which  consists  in  being  born  and  in  perishing.  This  is 
motion  which  proceeds  from  relative  non -being  to 
being,  and  from  the  latter  to  the  former.  There  is 
no  such  thing  as  absolute  generation  and  destruction. 
Individuals  alone  are  born  and  die  :  genera  remain. 
2nd,  quantitative  change  :  increase  or  diminution  ;  3rd, 
qualitative  change,  or  the  transition  from  one  sub- 
stance to  another  ;  4th,  spatial  change,  or  displacement. 

All  modes  of  change  are  conditioned  by  motion  in 
space.  Aristotle  makes  a  profound  study  of  this 
motion.  He  brings  against  the  arguments  of  the 
Eleatics,  who  deny  the  possibility  of  motion,  the 
doctrine  that  the  infinite  exists  only  in  potency,  not  in 
act.  The  infinite  consists  only  in  the  possibility  of  an 
indefinite  increase  of  numbers  and  in  the  indefinite 
divisibility  of  dimensions  :  it  cannot  be  the  given. 


ARISTOTLE  115 

When,  therefore,  we  reason  about  the  real,  we  should 
presuppose  only  finite  quantities. 

As  regards  space,  Aristotle  investigates  the  nature 
of  place.  The  place  of  a  body  is  not  something  in 
itself,  it  is  the  interior  limit  of  the  surrounding  body. 
It  is  like  a  motionless  vase  which  contains  the  body. 
Consequently,  all  bodies  are  not  in  a  place,  but  only 
those  enclosed  in  other  bodies.  The  sky,  the  universal 
container,  is  not  itself  in  a  place.  Space,  or  rather  the 
extent  of  the  world,  is  limited. 

Time  is  the  number  of  motion  as  regards  before  and 
after.  It  is  limitless  in  both  directions. 

Continuousness  is  the  characteristic  of  time  and  space. 
It  is  divisible  ad  infinitum,  though  in  dimensions  that 
are  themselves  continuous  :  not,  as  Zeno  supposed, 
in  indivisible  points.  All  dimension  is  divisible  into 
dimensions.  Moreover,  continuousness  is  an  imperfect 
notion,  and  relates  to  sensible  things,  for  it  is  divisible 
ad  infinitum,  and  consequently  is  indeterminate  as 
regards  the  number  of  its  elements. 

From  these  principles  Aristotle  concludes  that,  out- 
side the  world,  there  is  neither  space  nor  time,  that  the 
vacuum  of  the  atomists  is  inconceivable,  that  all  motion 
takes  place  in  the  plenum  by  a  process  of  substitution, 
and  that  time,  which  is  a  number,  presupposes,  as  does 
every  number,  a  soul  which  counts  its  units.  Motion 
in  space,  the  condition  of  all  other  motions,  is  the  only 
one  that  is  continuous.  And  circular  motion  is  the 
only  kind  that  is  capable  of  being  both  one  and  con- 
tinuous, without  beginning  or  end. 

Aristotle  does  not  regard  it  as  possible  to  explain  all 
changes  by  motion  in  space  alone.  He  looks  upon 
qualities  as  realities,  and  admits  that  qualitative  change 
is  incapable  of  being  reduced  to  motion  in  space.  This 


n6  STUDIES  IN  PHILOSOPHY 

theory  he  sets  up  in  opposition  to  the  mechanism  of 
Democritus  and  the  mathematisme  of  Plato.  He  raises 
two  objections  to  the  doctrines  of  these  philosophers  : 
1st,  Democritus  and  Plato  reduce  dimensions  to  in- 
divisible points  :  now,  all  dimension  is  divisible  ad 
infinitum  ;  2nd,  however  we  set  about  it,  it  is  impossible 
to  extract  quality  from  quantity,  pure  and  simple. 

It  is  for  this  reason  that  Aristotle  lays  down  the 
principle  of  a  qualitative  distinction  in  substances. 

And,  just  as  there  is  a  qualitative  nature,  there  is 
likewise  a  qualitative  transformation.  One  substance, 
acted  upon  by  another,  becomes  modified  in  its  inner 
nature.  This  phenomenon  is  possible  when  two  bodies 
are  partly  alike  and  partly  unlike,  that  is  to  say,  when 
two  substances  are  opposed  to  each  other  within  one 
and  the  same  genus,  and  it  is  possible  only  in  this  case. 
The  changing  of  one  of  these  substances  into  the  other 
is  no  mere  mechanical  displacement,  in  which  the 
elements  remain  identical  throughout  the  change  in  the 
compound  substance  :  it  is  really  the  formation  of  a 
new  substance,  fundamentally  different  from  the  former. 
The  given  substance  bears  the  same  relation  to  the 
substance  resulting  from  the  qualitative  change  that 
potency  bears  to  act. 

X. — MATHEMATICS 

Mathematics  considers  relations  of  dimension, 
quantity,  and  continuousness,  neglecting  the  other 
physical  qualities.  Thus,  it  deals  with  things  that  are 
immovable  without  existing  apart,  essences  intermediary 
between  the  world  and  God.  By  a  process  of  abstrac- 
tion the  mathematician  isolates  form  from  matter,  in 
sensible  things. 


ARISTOTLE  117 

Mathematics  is  either  pure  or  applied.  Geometry 
and  arithmetic  constitute  pure  mathematics.  Mathe- 
matics may  be  applied  either  to  the  practical  arts, 
geodesy,  for  instance,  or  to  natural  sciences,  such  as 
optics,  mechanics,  harmonics,  or  astrology.  In  the 
latter  case  the  question  of  fact  is  the  business  of  the 
physicist,  the  why  or  wherefore  is  the  business  of  the 
mathematician. 

Mathematics  makes  use  of  the  notions  of  the  good 
and  the  beautiful,  because  order,  symmetry,  and  deter- 
mination, all  of  which  are  pre-eminently  mathematical, 
are  some  of  the  most  important  elements  in  the  good 
and  the  beautiful. 

None  of  Aristotle's  mathematical  works  have  come 
down  to  us.  His  principal  ones  were  treatises  on 
mathematics,  unity,  optics,  and  astronomy.  In  the 
works  we  possess  we  frequently  come  across  examples 
taken  from  mathematics. 

XI. — COSMOLOGY 

From  the  eternity  of  form  and  matter  follows  the 
perpetuity  of  motion,  as  well  as  that  of  the  existence  of 
the  world.  Species  in  themselves  are  eternal,  and  there 
have  always  been  men  :  individuals  alone  are  born  and 
perish.  As  the  world  is  eternal,  the  science  of  the 
world  is  not  a  cosmogony,  but  rather  a  cosmology. 
Aristotle  has  not  the  formation  of  the  universe  to 
explain,  but  only  its  system. 

The  world  is  one,  finite  and  well  regulated.  It  is  a 
work  of  art,  as  beautiful  and  good  as  the  resistance  of 
the  material  element  permits.  It  has  a  perfect  form, 
the  spherical,  the  only  one,  moreover,  that  enables  the 
whole  to  move  without  causing  a  vacuum  outside  itself. 


n8  STUDIES  IN  PHILOSOPHY 

It  consists  of  two  unequal  halves:  ist,  the  supra- 
lunar  or  celestial  world,  the  vault  to  which  the  fixed 
stars  are  attached  ;  2nd,  the  infralunar  or  terrestrial 
world. 

The  celestial  world  is  animated  with  a  rotatory 
motion  produced  directly  by  God.  The  imperishable 
nature  of  the  stars  and  the  unchangeable  regularity  of 
their  motions  prove  that  they  differ,  as  regards  matter, 
from  terrestrial  things,  which  are  subject  to  continual 
change.  The  matter  composing  the  stars  is  ether,  or 
the  fifth  element  (quint-essence),  the  body  that  is  with- 
out a  contrary,  and  is  therefore  incorruptible,  admitting 
of  no  other  change  than  that  of  place,  and  no  other 
motion  than  a  circular  one.  The  other  elements,  on 
the  contrary,  being  formed  of  terrestrial  bodies,  are 
corruptible  and  admit  of  motion  from  below,  upwards, 
and  from  above,  downwards,  that  is  to  say,  from  centre 
to  circumference  and  from  circumference  to  centre. 
The  heaven  of  the  fixed  stars  is  the  abode  of  being,  of 
perfect  life,  and  of  unchangeable  order.  The  stars  are 
beings  which  are  not  subject  to  old  age,  beings  that  live 
a  life  of  happiness  whilst  exercising  eternal  and  inde- 
fatigable activity.  They  are  far  more  divine  than  man. 
Our  ancestors  had  a  vague  intuition  of  the  truth  when 
they  regarded  the  stars  as  being  gods. 

Within  the  heaven  of  the  fixed  stars  is  the  region  of 
the  planets,  including,  says  Aristotle,  the  sun  and  the 
moon,  as  well  as  the  five  planets  known  to  the  ancients. 
In  the  middle  of  the  world  is  the  earth,  spherical  in 
form.  The  heaven  of  the  planets  is  made  of  a  sub- 
stance that  is  less  and  less  pure  in  inverse  ratio  to  its 
distance  from  the  heaven  of  the  fixed  stars.  In  contra- 
distinction to  the  first  heaven,  which  is  a  single  sphere 
bearing  all  the  stars,  the  heaven  of  the  planets  consists 


ARISTOTLE  119 

of  a  multiplicity  of  spheres,  for  the  movements  of  the 
planets,  being  relatively  irregular,  presuppose  a  multi- 
plicity of  movers  whose  actions  combine  with  one 
another. 

Beings  other  than  the  fixed  stars  are  made  of  the 
four  elements.  Each  element  has  a  motion  of  its  own, 
the  rectilineal  march  towards  the  place  natural  to  it. 
Hence  we  obtain  weight  and  lightness.  Weight  is 
the  tendency  of  each  body  to  follow  its  own  direction. 
It  is  not  possible  to  say,  with  Democritus,  that  all 
motion  is  simply  the  result  of  impacts  ad  infinitum.  A 
halt  in  retrogression  must  be  made,  in  the  logical  order 
of  things,  at  all  events.  The  motion  that  results  from 
compulsion  presupposes  spontaneous  motion. 

It  is  the  property  of  the  terrestrial  element  to  in- 
cline towards  the  centre,  hence  the  position  of  the  earth, 
immovable  in  the  centre  of  the  universe.  The  earth  is 
spherical.  Its  elements  are  in  double  opposition — of 
weight  and  quality — to  one  another.  On  the  one 
hand,  they  are  heavy  or  light ;  on  the  other,  they  are 
hot  or  cold,  dry  or  moist.  The  result  of  this  opposi- 
tion is  that  the  elements  of  the  earth  are  constantly 
changing  into  one  another.  Heat  and  light  are 
generated  by  the  friction  to  which  the  air  is  subjected 
owing  to  the  extreme  velocity  of  the  celestial  spheres. 
By  reason  of  the  inclination  of  the  ecliptic,  light  and 
heat  are  produced  in  different  degrees  in  different 
regions  of  the  earth  and  at  different  times  of  the  year. 
This  is  the  origin  of  the  circulus  of  generation  and 
destruction,  that  image  of  eternity  in  perishable  nature. 
Action  proceeds  from  periphery  to  centre  ;  the  heaven 
of  the  fixed  stars  representing  highest  form,  the  earth 
representing  lowest  matter.  The  various  mineral  and 
organised  bodies  are  formed  by  the  mutual  action  of  the 


120  STUDIES  IN  PHILOSOPHY 

two  active  potencies,  heat  and  cold,  and  of  the  two 
passive  potencies,  moisture  and  dryness. 

Terrestrial  beings  form  a  hierarchy,  extending  from 
the  being  which  is  nearest  to  brute  matter  up  to  the 
male  human  being.  Each  lower  form  is  the  basis  of 
the  higher  ones  ;  each  higher  form  the  relative  comple- 
tion of  the  lower  ones.  The  principal  stages  in  the 
hierarchy  are  represented  by  lifeless  bodies,  plants, 
animals,  and  man. 

XII. — ASTRONOMY 

Aristotle  made  much  study  of  astronomy.  Simpli- 
cius,  so  Porphyry  tells  us,  relates  that,  with  a  view  to 
investigations  in  this  science,  he  instructed  Callisthenes 
to  collect  the  astronomical  observations  made  by  the 
Chaldaeans  in  Babylon,  especially  those  that  dated  back 
nineteen  hundred  years  before  the  time  of  Alexander. 
Aristotle  himself  tells  us  that  he  utilised  the  observa- 
tions of  the  Egyptians  and  Babylonians,  dating  back  to 
a  very  distant  age.  He  wrote  an  'Ao-rpovofurcov,  which 
is  lost. 

All  the  celestial  beings,  according  to  Aristotle,  are 
spherical.  The  first  heaven,  that  of  the  fixed  stars,  is  a 
sphere.  The  planets  are  moved  by  spheres  ;  the  earth 
is  spherical. 

All  simple  motion  is  rotation  round  an  axis.  The 
heaven  of  the  fixed  stars  has  only  one  motion.  The 
heaven  of  the  planets  (Saturn,  Jupiter,  Venus,  Mercury, 
Mars,  the  Sun,  the  Moon)  has  several  for  each  planet. 
The  earth  is  without  motion. 

Aristotle  maintains  the  doctrine  of  the  sphericity  of 
the  earth,  and  gives  the  correct  explanation  of  the 
phases  of  the  moon. 


ARISTOTLE  121 

He  worked  with  the  astronomer  Callipus  in  com- 
pleting and  rectifying  the  theory  of  the  spheres  formed 
by  Eudoxus,  the  first  astronomer  of  his  day,  as  well  as 
the  theory  of  Callipus  himself.  His  theory  may  be 
summed  up  as  follows  : — 

We  must  admit,  said  Aristotle,  along  with  Plato — 
who  in  this  matter  followed  the  lead  of  Eudoxus  and 
Callipus — the  number  of  spheres  and  their  mode  of 
motion,  necessary  for  the  explanation  of  the  revolutions 
of  the  planets,  as  they  appear  under  observation,  with 
no  other  elements  than  uniform  rotatory  motions. 
Presenting  the  problem  in  this  way,  Eudoxus  inferred 
that  there  were  twenty-six  spheres,  Callipus  thirty- 
three.  Aristotle  accepts  the  latter  figure.  But  since, 
in  his  philosophy,  the  exterior  spheres  are  to  the 
interior  what  form  is  to  matter,  he  is  obliged  to  add 
antagonistic  spheres,  in  order  that  each  exterior  sphere 
may  not  communicate  its  motion  to  all  the  spheres 
interior  to  itself,  as  does  the  sphere  of  the  fixed  stars. 
For  each  planet,  then,  there  are  as  many  antagonistic 
spheres  as  are  needed  to  counteract  the  action  of  the 
exterior  planetary  spheres.  The  supplementary  spheres 
are  twenty-two  in  number,  and  these,  added  on  to  the 
thirty-three  of  Callipus,  make  fifty-five  spheres.  But  if 
we  consider  that  the  sun  and  the  moon,  being  far  away 
from  the  rest  of  the  planets,  have  no  need  of  antagonistic 
spheres,  the  total  number  of  the  spheres  will  be  reduced 
to  forty -seven.  This,  says  Aristotle,  is  probable 
enough. 

To  each  of  these  spheres  motion  must  be  communi- 
cated, as  it  was  to  the  first  heaven,  by  an  incorporeal 
substance,  a  spirit,  or  a  god.  The  constellations,  the 
object  and  end  of  the  motions  of  the  spheres,  are 
moreover,  for  that  reason,  their  true  causes.  Conse- 


122  STUDIES  IN  PHILOSOPHY 

quently  the  constellations  are  animated  beings,  endowed 
with  reason  and  superior  to  man. 

XIII. — METEOROLOGY 

Meteorology  had  been  much  studied  since  the  time 
of  Thales.  Aristotle  profited  by  the  labours  of  his 
predecessors,  though  he  also  made  original  investigations 
in  this  science  along  the  lines  of  his  own  philosophy. 

Meteorological  phenomena  are  the  result,  he  says,  of 
the  action  of  the  four  elements  upon  one  another.  In 
accordance  with  the  nature  of  these  elements,  the  results 
of  their  mutual  action  are  less  determined  and  obey  less 
strict  laws  than  the  phenomena  that  take  place  in  the  first 
element :  the  ether.  It  is  for  this  reason  that  Aristotle, 
when  considering  meteors,  seeks  after  explanations 
that  are  mainly  of  a  mechanical  and  empirical  nature. 
He  attributes  a  preponderating  influence  to  heat.  In 
this  way  he  explains  comets,  the  Milky  Way,  clouds, 
fogs,  winds,  the  relations  between  seas  and  continents 
and  the  formation  of  the  sea.  His  explanations  often 
testify  to  exact  observation  and  skilful  reasoning. 
Winds,  for  instance,  are  explained  by  the  motion  of 
vapours,  as  a  result  of  their  differences  of  tempera- 
ture. Earthquakes  are  due  to  the  action  of  subterranean 
gases.  The  rainbow  is  but  a  phenomenon  caused  by 
reflection  :  in  the  sun's  light,  the  spray  composing  the 
clouds  acts  as  a  mirror. 

These  investigations  are  purely  theoretical :  Aristotle 
does  not  dream  of  using  them  for  the  purpose  of  pre- 
dicting phenomena. 

XIV. — MINERALOGY 

Minerals  are  homogeneous  bodies  which  remain  so, 
without  becoming  organised  into  individuals  consisting 


ARISTOTLE  123 

of  different  parts.  These  bodies  are  formed  by  cold 
and  heat,  combining  or  disintegrating — in  so  far  as 
they  are  active  properties — the  moist  and  the  dry,  which 
play  the  rtk  of  passive  properties. 

XV. — GENERAL  BIOLOGY 

Biology  forms  a  considerable  portion  of  Aristotle's 
scientific  work.  He  probably  utilised  many  of  the 
works  of  his  predecessors,  mainly  those  of  Democritus, 
but  he  went  so  far  beyond  the  rest  that  he  stands  out 
as  the  true  founder  of  biology  in  Greece.  He  works 
mainly  by  observation,  the  determination  of  phenomena 
being  made  to  precede  the  investigation  of  causes.  To 
simple  observation  he  would  appear  to  have  added 
dissection.  He  proceeds  from  anatomy  to  physiology 
and,  speaking  generally,  regards  biology  as  the  ground- 
work of  physics,  basing  it  on  a  knowledge  of  the 
four  elements.  He  deals  not  only  with  every  conceiv- 
able problem  of  his  own  times,  but  also  with  almost  all 
the  problems  that  engage  the  attention  of  modern 
scientists.  The  solutions  he  offers  are,  for  the  most 
part,  carefully  set  forth  ;  and,  considering  the  state  of 
knowledge  at  the  time,  his  reasonings  are  correct 
and  ingenious.  It  must  be  confessed,  however,  that  his 
explanations  are  frequently  arbitrary  or  rather  meta- 
physical ;  at  times,  even,  he  appears  to  have  given 
demonstrative  value  to  mere  legends. 

Life  is  motion.  Now,  all  motion  presupposes  a 
form  that  moves  and  matter  that  is  moved.  The  form 
is  the  soul ;  matter  the  body.  The  soul  is  neither  body 
nor  without  body.  It  moves  without  moving  ;  it  is 
immovable,  not  self-moving,  as  Plato  imagined.  As 
being  the  form  of  the  body,  it  is  its  goal ;  the  body 


i24  STUDIES  IN  PHILOSOPHY 

is  nothing  but  the  instrument  of  the  soul,  and  its 
structure  is  guided  by  this  destination.  Aristotle 
correctly  defines  the  soul  as  the  first  entelechy  of  an 
organic  physical  body.  This  means  that  the  soul  is 
the  permanent  force  which  moves  the  body  and  deter- 
mines its  constitution. 

It  is  natural  for  the  finality  of  nature  to  appear  in 
living  beings  more  clearly  than  anywhere  else,  because 
everything,  in  them,  at  the  outset,  is  calculated  with 
a  view  to  the  soul.  But  just  as  form  only  gradually 
overcomes  the  resistance  of  matter,  so  there  are  degrees 
in  the  psychic  life,  and  these  are  essentially  three  in 
number  :  nutritiveness,  sensibility,  and  intelligence. 
Nutritiveness  is  the  fundamental  quality  of  living 
beings  ;  from  it  proceed  vital  development  and  death. 
It  exists  both  in  plants  and  in  animals.  The  latter 
possess  sensibility  in  addition.  Man,  a  superior  animal, 
possesses  all  three  :  nutritiveness,  sensibility,  and 
intelligence. 

Aristotelian  biology  deals  principally  with  animals. 
The  body  of  an  animal  consists  of  homoeomerous 
substances  :  a  mixture  of  elementary  substances.  The 
immediate  matter  of  the  soul  is  breath  (Trvevpa),  the 
principle  of  vital  heat,  a  body  akin  to  ether,  along  with 
which  the  soul  is  transmitted,  in  the  semen,  from  father 
to  child.  The  principal  seat  of  heat  is  the  central 
organ,  that  is  to  say,  the  heart,  in  animals  endowed  with 
blood.  In  the  heart  the  blood  is  cooked,  after  being 
formed  of  the  nutritive  substances  introduced  by  the 
veins,  and  blood,  as  final,  definite  nourishment,  feeds 
and  sustains  the  body.  It  becomes  flesh  and  bone,  nail 
and  horn,  etc.  The  nutritive  power  of  foods  is  not 
the  result  of  their  containing  particles  of  flesh,  bone, 
and  marrow,  which  would  go  to  unite  directly  with  like 


ARISTOTLE  125 

substances  existing  in  the  body,  it  is  rather  owing  to  the 
food  being  cooked  several  times  that  it  reaches  a  state 
enabling  it  to  be  assimilated  by  the  organism.  Though 
very  precise  on  the  matter  of  assimilation,  Aristotle 
would  appear  never  to  have  thought  of  disassimilation. 

XVI. — BOTANY 

Aristotle's  works  on  botany  are  lost,  but  he  certainly 
gave  an  impetus  to  the  investigations  made  on  plants 
in  his  school ;  he  seems  to  have  largely  contributed  to 
the  creation  of  scientific  botany. 

XVII. — ANIMAL  ANATOMY  AND  PHYSIOLOGY 

A  distinction  must  be  made  between  general 
anatomy  and  physiology  and  comparative  anatomy  and 
physiology. 

The  parts  of  the  animal  organism  are  of  two  kinds  : 
the  homogeneous,  such  as  the  tissues  ;  and  the  hetero- 
geneous, such  as  the  organs.  Each  organ  has  a 
function,  the  tongue,  hand,  etc.  The  tissues  have 
properties.  Aristotle  studies  first  the  homogeneous, 
then  the  heterogeneous  parts. 

The  homogeneous  parts  are :  ist,  veins,  bones, 
cartilages,  nails,  hair,  horn,  etc.  ;  2nd,  fat,  grease, 
blood,  marrow,  flesh,  milk,  semen,  membranes.  In 
many  cases,  Aristotle's  explanations  regarding  these 
parts  are  finalistic,  and  derive  the  nature  of  the  part 
from  its  function.  For  instance,  he  says  the  incisors 
appear  before  the  molars,  because  food  must  first  be 
cut  up  or  torn  to  be  in  a  fit  state  to  be  ground. 

The  anatomical  study  of  the  heterogeneous  parts  is 
not  distinct  from  their  physiological  study. 


126  STUDIES  IN  PHILOSOPHY 

The  first  of  all  the  organs  is  the  heart.  Aristotle 
has  no  notion  of  the  circulation  of  the  blood,  as  we 
understand  the  word,  nor  does  he  say  anything  of  the 
two  kinds  of  blood  ;  he  acknowledges,  however,  that  it 
is  carried  throughout  the  body  by  the  veins,  as  by 
canals.  The  heart  is  the  centre  of  the  living  being,  the 
seat  of  the  forhiation  of  the  blood,  and  the  source  of  its 
heat.  All  animals  possess  a  heart  and  blood,  or  substi- 
tutes for  these  primary  conditions  of  life.  Those 
animals  that  can  be  divided  or  cut  up  without  the  parts 
immediately  ceasing  to  live,  are  not  simple  animals,  but 
rather  aggregates  of  animals.  The  degree  of  unity  is 
the  standard  of  the  perfection  of  the  being.  No 
mutilated  animal  recovers  from  its  injuries  as  does  the 
plant,  in  which  the  life  principle  is  dispersed  throughout 
the  entire  being. 

The  other  heterogeneous  parts  are  :  the  diaphragm, 
the  sense  organs,  the  organs  of  motion,  the  encephalon, 
the  lungs,  the  abdominal  viscera,  and  the  sex  organs. 

Aristotle  enlarges  on  the  senses.  Sensation  consists 
in  being  moved,  in  experiencing  some  change.  There  are 
two  kinds  of  senses  :  the  mediate,  which  act  through  the 
medium  of  the  atmosphere,  as  sight,  hearing,  and  smell ; 
and  the  immediate,  which  act  by  contact,  as  touch  and 
taste  ;  the  latter  being  more  important  for  the  preserva- 
tion of  the  individual.  The  mediate  senses  estimate 
either  differences  in  the  nature  of  objects,  or  else 
distances  ;  consequently  we  must  make  a  distinction 
between  their  acuteness  and  their  sphere  of  action. 

The  eye  is  not  a  mere  mirror  :  the  presence  of  an 
image  would  not  suffice  to  produce  vision  :  there  is 
required  a  psychic  property,  which  a  mere  mirror  does 
not  possess.  The  inmost  recesses  of  the  eye  not  only 
reflect  the  image,  they  have  the  property  of  seeing  as  well. 


ARISTOTLE  127 

Indirectly,  hearing  is  the  most  intellectual  of  all  the 
senses,  for  it  enables  ideas  to  be  communicated  by  means 
of  language.  Speech  is  nothing  but  a  sequence  of  sounds 
that  have  entered  the  ear  ;  it  is  one  and  the  same  motion 
diffused  from  ear  to  throat. 

Touch  differs  from  the  rest  of  the  senses  in  that  the 
latter  supply  us  with  oppositions  or  contrasts  of  a  single 
kind  only,  whereas  touch  enables  us  to  distinguish  hot 
and  cold,  dry  and  moist,  hard  and  soft. 

Aristotle  is  acquainted  with  no  other  organs  of 
motion  than  the  tendons,  and  these  he  calls  nerves. 
He  tries  to  discover  the  principle  thereof,  not  in  the 
limbs  themselves,  but  in  a  central  organ  of  motion. 
The  principle  of  motion  is  the  heart,  or,  in  the  case  of 
animals  that  have  none,  the  corresponding  organ. 
Motions  are  of  two  kinds,  voluntary  and  involuntary. 
The  beating  of  the  heart,  for  instance,  belongs  to  the 
second  type  of  motion. 

As  the  heart  is  a  calorific  organ,  so  the  encephalon 
and  the  lungs  are  refrigerant  organs. 

Of  the  abdominal  organs,  Aristotle  carefully  studies 
the  stomach,  giving  remarkably  correct  descriptions  as 
regards  ruminants,  birds  and  the  organs  of  sex  upon  which 
his  observations  are  frequently  very  apt  and  successful. 
His  investigations  lead  him  to  discuss  the  part  played 
by  both  sexes  in  the  production  of  the  new  being. 

He  also  applies  himself  to  the  question  of  heredity. 
He  rejects  pangenesis  (which  states  that  the  parents 
contribute  germs  resembling  themselves),  alleging  that 
there  are  products  which  do  not  resemble  their  parents  : 
for  instance,  caterpillars  born  of  butterflies.  According 
to  Aristotle,  the  material  that  goes  to  the  formation  of 
the  new  being  is  made  up  of  substances  different  from 
that  of  the  parents  themselves.  There  is  a  male  seminal 


128  STUDIES  IN  PHILOSOPHY 

fluid,  the  sperm,  and  a  female  one,  the  menstrua.  From 
the  blending  of  these  two  elements,  as  from  the  union 
of  form  with  matter,  results  the  germ.  Thus,  from  the 
man  there  is  born  the  soul,  and  from  the  woman  the 
body,  of  the  child  resulting  from  their  union. 

The  difference  in  the  sexes  may  be  reduced  to  a 
difference  in  degree.  In  the  woman  food  has  not 
received  so  complete  an  elaboration  as  in  man,  the 
creative  power  has  not  finished  its  work. 

In  like  manner  Aristotle  explains  instances  of  tera- 
tology. Monstrosities  are  only  greater  or  less  dis- 
similarities, the  result  of  excess  or  defect.  They  deviate 
from  the  ordinary  course  of  events,  though  having  their 
basis  in  natural  forces. 

In  the  same  spirit  Aristotle  dealt  with  embryogeny. 
Interpreting  the  results  of  his  delicate  observations  in 
accordance  with  the  principles  of  his  philosophy,  he  admits 
that  the  development  of  the  germ  is  an  epitome  of  the 
general  progress  of  life  in  nature.  First,  the  life  of  the 
germ  is  comparable  to  vegetable  life  ;  afterwards,  the 
embryo  is  in  a  state  that  may  be  compared  to  sleep  : 
sleep,  however,  from  which  there  is  no  awakening.  The 
foetus  becomes  animal  when  it  acquires  feeling  ;  then 
only  is  it  capable  of  genuine  sleep.  The  order  in  which 
the  organs  appear  is  determined  by  their  utility  and  by 
the  share  they  have  in  the  formation  of  the  other  organs. 
Thus,  the  heart  is  the  first  organ  to  be  developed. 

In  Aristotle  we  find  numerous  aphorisms  and  biolo- 
gical considerations  resulting  from  what  we  call  com- 
parative anatomy  and  physiology.  He  makes  a  careful 
study  of  organic  resemblances  and  differences.  Organs 
may  resemble  one  another  in  form.  Organs  apparently 
different  may  be  only  more  or  less  complete  develop- 
ments of  one  and  the  same  type,  so  that,  at  bottom, 


ARISTOTLE  129 

excess  or  deficiency  really  constitutes  the  whole  differ- 
ence. There  may  be  resemblance  by  analogy ;  for 
instance,  the  feather  is  to  the  bird  what  the  scale  is  to 
the  fish.  There  is  the  same  relation  between  the  bones 
of  land  animals  and  those  of  fishes,  between  nails  and 
horns,  etc.  Different  species  may  have  the  same  organs 
diversely  situated.  Different  organs  may  perform  the 
same  function. 

Aristotle  determines  numerous  organic  correlations. 
For  instance,  all  animals  have  blood,  or  its  equivalent. 
Animals  with  no  feet  at  all,  and  those  with  two  and  four 
feet,  possess  blood  ;  in  those  with  more  than  four  feet 
lymph  takes  the  place  of  blood.  In  ruminants  there  is 
a  correlation  between  the  possession  of  horns  and  the 
lack  of  canine  teeth.  The  lateral  movements  of  the 
lower  jaw  are  found  only  in  such  animals  as  grind  their 
food.  All  truly  viviparous  animals  breathe  in  air,  etc. 

The  law  regarding  division  of  work  is  clearly  formu- 
lated. Nature,  says  Aristotle,  if  there  is  nothing  to 
hinder,  always  employs  two  special  and  distinct  organs 
for  two  different  functions.  When  this  cannot  be  done, 
the  same  instrument  is  used  for  several  purposes  ;  though 
it  is  preferable  that  the  same  organ  should  not  be  used 
for  several  functions. 

The  influences  of  environment  are  shown  to  con- 
tribute to  the  determination  of  animal  forms.  In  hot 
climates,  says  Aristotle,  it  is  principally  animals  which 
are  cold  by  nature,  such  as  serpents,  lizards  and 
those  covered  with  scales,  that  grow  to  considerable 
dimensions. 

Aristotle  also  studied  physiognomy,  or  the  relation 
which  the  physical  bears  to  the  moral.  In  all  probability 
the  Physiognomonica  is  not  an  authentic  work,  though 
doubtless  it  owes  its  origin  to  his  teaching.  In  the 

K 


130  STUDIES  IN  PHILOSOPHY 

Historia  animalium  we  find  him  trying  to  find  out  to 
what  moral  differences  the  physical  differences  in  the 
human  face  correspond. 

According  to  our  philosopher,  the  species,  properly 
so  called,  are  stable  and  separated  from  one  another. 
Along  with  the  absolute,  however,  Aristotle  recognises 
the  existence  of  the  contingent.  Consequently,  there 
is  a  certain  freedom  of  action  in  nature,  and  organic 
forms  and  faculties  admit  of  restricted  variability.  An 
apparently  insignificant  difference,  found  in  small  parts, 
may  suffice  to  produce  considerable  differences  in  the 
ensemble  of  the  animal's  body.  For  instance,  only  a 
small  portion  of  an  animal's  body  is  removed  by  castra- 
tion, and  yet  this  removal  changes  its  nature,  bringing 
it  into  closer  resemblance  with  the  other  sex.  When 
the  animal  is  in  the  embryonic  state,  a  very  slight  differ- 
ence will  cause  it  to  be  either  a  male  or  a  female.  The 
difference  between  the  terrestrial  and  the  aquatic  animal 
results  from  the  different  arrangement  of  small  parts. 
In  a  word,  says  Aristotle,  in  nature  there  is  unity  of 
composition  and  progressive  continuity.  Man  himself, 
who,  as  far  as  we  know,  is  at  the  top  of  the  ladder,  is 
only  separated  from  the  animals,  physically  speaking, 
by  more  or  less  pronounced  differences.  The  transition 
from  one  kingdom  to  another  is  imperceptible.  Thus, 
in  the  sea  we  find  beings  at  a  stage  intermediate  between 
animals  and  plants,  e.g.  sponges.  The  principal  types 
and  stages  of  growth,  as  it  were,  are  none  the  less 
determined  and  mutually  irreducible. 

XVIII.— ZOOLOGY 

Aristotle  was  the  first  classifying  zoologist.     Truth 
to  tell,  he  does  not  appear  to  have  had  any  intention  to 


ARISTOTLE  131 

set  up  a  zoological  classification,  and  his  attempts  in 
this  direction  are  offered  only  as  examples.  Nor  did  he 
make  any  sharply-drawn  distinction  of  animals,  distri- 
buting them  in  a  hierarchy  of  genera  and  species  ;  he 
merely  assigned  limits  to  the  principal  groups.  He 
clearly  saw,  however,  that  the  criterion  of  species  is 
obtained  through  reproduction — interfecundity.  He 
regards  as  of  the  same  species  only  such  animals  as 
spring  from  common  parents.  His  classification  aims 
at  being  natural,  that  is  to  say,  it  tends  to  bring  together 
those  animals  that  have  a  fundamental  resemblance  to 
one  another.  Here,  as  elsewhere,  his  object  is  to 
distinguish  essence  from  accident. 

The  first  division  is  that  between  animals  that  have 
blood  (our  vertebrates)  and  those  that  have  no  blood 
(our  invertebrates).  The  divisions  between  sanguineous 
animals  are  mainly  based  on  embryogeny  and  a  considera- 
tion of  the  element  in  which  they  live.  Sanguineous 
animals  are  divided  into  true  vivipara,  ovovivipara  and 
ovipara.  Animals  devoid  of  blood  are  divided  into 
mollusks  (corresponding  to  our  cephalopoda),  Crustacea, 
testacea  (corresponding  to  our  mollusks,  with  the  excep- 
tion of  the  cephalopoda)  and  insects. 

In  his  description  of  the  species — he  mentions  about 
four  hundred  of  them — Aristotle  shows  that  he  pos- 
sesses extensive  knowledge.  Amongst  other  things,  he 
deals  with  the  mental  and  moral  faculties  of  animals. 
Bees  he  calls  the  wise — the  well-behaved — ones. 

As  regards  the  first  origin  of  man  and  of  the  other 
sanguineous  animals,  he  is  inclined  to  think  that  they 
proceed  from  a  sort  of  scolex  (head  of  the  tapeworm),  or 
else  from  a  perfect  egg,  in  which  only  a  portion  becomes 
the  germ,  developing  at  the  expense  of  the  rest.  He 
considers  the  spontaneous  production  of  a  perfect  egg 


1 32  STUDIES  IN  PHILOSOPHY 

as  not  at  all  likely,  since  we  never  meet  with  an  instance 
of  it.  Testacea  and  worms,  on  the  contrary,  have 
spontaneous  birth. 

XIX. — PSYCHOLOGY 

That  which  differentiates  man  from  the  rest  of  the 
animal  kingdom  is  the  1/01)9,  which,  in  him,  is  united 
to  the  animal  soul.  He  possesses  faculties  common 
to  himself  and  the  animals,  and  faculties  peculiar  to 
himself.  In  common  with  the  animals,  man  has  sensa- 
tion and  the  faculties  derived  therefrom. 

Sensation  is  the  change  effected  in  the  mind  by  a 
sensible  object,  as  by  a  contrary,  through  the  agency 
of  the  body,  and  consisting  in  the  form  of  the  object 
that  is  sensed  being  communicated  to  the  subject  that 
senses.  Thus,  sensation  is  the  common  act  of  a 
sensible  objec^and  a  sensing  subject. 

Each  sense  gives  us  exclusive  information  regarding 
the  properties  of  those  things  with  which  it  specially 
deals  ;  what  it  tells  us  of  these  properties  is  always 
true.  General  properties  are  known  by  the  sensorium 
commune,  in  which  all  sensible  impressions  meet.  Here, 
too,  sensations  are  compared  and  related  to  objects  as 
causes,  and  to  ourselves  as  conscious  subjects.  The 
organ  of  the  sensorium  commune  is  the  heart.  Its  data 
may  be  either  true  or  false. 

Sensation  is  the  basis  of  animal  psychic  life.  Both 
from  the  theoretical  and  the  practical  standpoint  it  is 
capable  of  a  development  which  brings  several  other 
faculties  into  being. 

When  motion  in  the  sense-organ  continues  beyond 
the  duration  of  the  sensation,  extends  to  the  central 
organ,  and  there  causes  a  new  appearance  of  the  sensible 


ARISTOTLE  133 

image,  we  have  imagination.  The  products  of  this 
faculty  may  be  either  true  or  false.  When  an  image 
is  recognised  as  the  reproduction  of  a  past  perception, 
we  have  memory.  Aristotle  adds  to  the  study  of  these 
faculties  investigations  as  to  the  nature  of  sleep,  death 
and  dreams,  from  the  psychological  point  of  view. 

Looked  upon  from  the  practical  standpoint  of  good 
and  evil,  sensation  admits  of  development  along  the 
lines  just  mentioned.  From  the  sole  fact  that  an 
animal  is  endowed  with  sensation,  it  is  capable  of 
pleasure  and  pain.  When  its  activity  is  unchecked, 
we  have  pleasure  ;  in  the  contrary  event,  pain.  Plea- 
sure and  pain,  in  beings  fully  susceptible  to  them,  are 
really  judgments  upon  the  true  value  of  things.  Con- 
sequently, beings  capable  of  pleasure  and  pain  have 
desire,  which  is  nothing  but  the  seeking  after  what  is 
agreeable.  They  also  have  passions. 

All  these  functions  already  appertain  to  animals, 
though  they  are  realised  to  perfection  only  in  man. 
Man  possesses  intelligence  in  addition  to  the  rest. 
Hitherto  we  have  seen  that  there  has  been  continual 
development  and  progress.  Between  the  animal  soul 
and  the  i/oi)?,  however,  there  is  a  break  of  continuity. 
The  vovs  is  the  knowledge  of  first  principles.  It  has 
no  birth,  but  is  eternal.  Exempt  from  passivity,  it 
exists  in  act.  Being  without  organ,  it  is  not  the  result 
of  the  development  of  sensation,  but  comes  from 
without,  and  is  separable. 

Human  intelligence,  however,  is  not  merely  this  com- 
plete, immovable  z/ov?.  It  learns,  becomes  acquainted 
with  perishable  things,  things  capable  of  being  as  they 
are  or  otherwise.  The  z/ofc,  therefore,  in  man  blends 
with  the  soul  :  there  is  a  lower  intellect,  intermediary 
between  the  absolute  vovs  and  the  animal  soul.  This 


i34  STUDIES  IN  PHILOSOPHY 


intellect  may  be  called  vovs  Tra^ri/co?,  passive  intellect, 
in  opposition  to  1/01)9  airadr)?,  or  active  intellect.  This 
lower  i/oi)?  is  the  subject,  but  not  the  object  ;  perishable 
things  are  its  object.  Depending  on  the  body,  it 
perishes  with  the  body.  There  are  rudiments  of  this 
passive  intellect  in  certain  animals,  e.g.  in  bees,  but 
only  in  man  is  it  fully  developed. 

The  I/oik  7ra077Tt/co5  has  two  kinds  of  functions, 
theoretical  and  practical. 

From  the  theoretical  point  of  view,  the  1/01)5  7ra^7/Tt/co5, 
at  first,  is  1/01)5  only  in  potency.  It  is  a  tabula  rasa  on 
which  nothing  has  yet  been  written.  The  z/ov5  ira&prMcof 
thinks  only  by  the  aid  of  images,  and  under  the 
influence  of  the  higher  1/01)5.  It  thus  deduces  from 
sensation  the  general  contained  therein,  and  which 
sensation  reaches  only  by  accident  :  it  gradually  becomes 
determined  by  reason  of  these  general  essences.  Per- 
fect science,  however,  belongs  only  to  the  z/o£>5  OewprjriKos, 
the  higher  1/01)5,  which,  starting  from  causes,  proceeds 
a  priori. 

The  i/ou5,  as  regards  its  practical  use,  has  no  prin- 
ciples of  its  own  :  practice  consisting  only  of  the 
application  of  theoretical  ideas.  This  realisation  comes 
about  in  two  ways  :  ist,  by  production  (Troielv)  ;  2nd, 
by  action  (Trpdrreiv). 

With  regard  to  action,  Aristotle  offers  a  theory  of 
will,  the  spring  of  action.  Will  is  the  combination  of 
intellect  and  desire.  As  desire,  it  posits  ends  to  be 
realised  ;  as  intellect,  it  determines  the  means  that 
correspond  to  these  ends.  \The  objects  of  will  are 
determined  with  reference  to  two  principal  ends  :  the 
good  and  the  possible. 

Free-will  is  connected  with  the  existence  of  will. 
In  beings  devoid  of  reason,  desire  can  only  spring  from 


ARISTOTLE  135 

sensation.  In  man,  it  may  be  engendered  either  by 
sensation  or  by  reason.  Engendered  by  sensation,  it  is 
appetite  ;  engendered  by  reason,  it  is  will.  Between 
appetite  and  will  we  have  free-will  :  the  faculty  of  self- 
determination.  Virtue  and  vice  depend  on  ourselves  ; 
each  man  is  the  principle  of  his  own  actions.  The  reality 
of  free-will  is  proved  by  moral  imputability,  which 
legislation,  praise  and  blame,  exhortation  and  prohibi- 
tion imply.  .The  essence  of  free-will  is  spontaneity^  in 
more  precise  language,  that  spontaneity  which  mani- 
fests preference  ;  for  children  and  animals  show  con- 
siderable spontaneity,  but  man  alone  is  truly  free,  for 
he  alone  is  capable  of  choice. 

\J 

XX. — MORALS 

In  the  case  of  beings  without  intelligence,  ends 
are  attained  immediately  and  of  necessity.  Man  has 
a  loftier  end,  which  is  not  only  realised  by  the  sole 
operation  of  natural  forces,  but  also  by  using  his  free- 
dom. The  problem  is  to  find  out  how  to  organise 
one's  life  in  order  to  realise  the  human  idea,  to  act 
according  to  one's  own  essence,  and  not  from  necessity 
or  chance.  Hence  the  idea  of  practical  philosophy  : 
the  philosophy  of  human  affairs.  The  aim  of  this 
philosophy  is  to  find  out  what  are  the  end  and  the 
means  of  that  activity  which  is  proper  to  mankind. 

Practical  philosophy  comprises  three  parts,  corre- 
sponding to  the  three  spheres  of  action  that  open  out 
to  man  :  ist,  ethics  >  or  the  rules  of  individual  life  ;  2nd, 
economics,  or  the  rules  of  family  life  ;  and  3rd,  politics, 
or  the  rules  of  social  life.  In  chronological  order, 
ethics  precedes  economics  which  itself  precedes  politics. 
In  the  order  of  nature  and  perfection,  the  relation  is 


136  STUDIES  IN  PHILOSOPHY 

inverted.  Politics,  indeed,  is  the  completion  of  econo- 
mics, which  itself  determines  human  activity  with 
greater  precision  than  ethics,  pure  and  simple. 

We  will  begin  with  ethics  or  morals.  Morals  may 
be  divided  into  general  and  particular  morals. 

In  Aristotle  morals  does  not  bear  the  same  relation 
to  physics  as  in  Plato.  The  good  is  not  transcendent ; 
nature  is  not  hostile  or  simply  passive  when  brought 
in  contact  with  the  ideal.  As  form  exists  in  potency 
in  matter,  so  nature  is  inclined  to  virtue,  which  is  only 
the  normal  development  of  natural  tendencies.  We 
may  not  be  born  virtuous,  but  of  ourselves  we  tend 
to  become  so  :  culture  and  art  are  the  completion  of 
nature.  Moreover,  we  must  distinguish  between  good 
in  itself  and  good  for  mankind.  The  good  which  is 
taken  into  consideration  by  morals  is  not  good  in  itself, 
but  only  so  far  as  it  deals  with  human  nature. 

What  is  moral  good  ?  (  Since  all  action  has  an  object, 
there  must  be  a  supreme  object,  and  this  can  only  be 
that  good  which  is  superior  to  all  other  good,  the  best. 
What  is  this  best  ?  The  general  impression  is  that  it 
is  happiness,  but  there  is  no  agreement  as  to  the  defini- 
tion of  happiness.  We  must  try  to  find  out  in  what 
it  really  consists. 

For  every  living  being,  good  consists  in  the  perfec- 
tion or  full  realisation  of  the  activity  peculiar  to  itself. 
Such  is  the  distinctive  mark  of  true  happiness.  This 
happiness,  then,  cannot  be  said  to  be  either  in  the 
enjoyment  of  the  senses,  which  is  common  to  man  and 
animal,  nor  in  pleasure,  which  is  not  an  end  in  itself 
but  is  pursued  only  with  a  view  to  happiness,  nor  in 
honour,  which  does  not  lie  within  our  power  and  comes 
from  without.  Perhaps  even  virtue  alone  does  not 


ARISTOTLE  137 

afford  happiness,  for  we  could  not  designate  as  happy 
a  virtuous  man,  hindered  in  his  activity  or  suffering 
acutely.  Happiness  consists  of  the  constant  exercise 
of  our  strictly  human,  i.e.  intellectual,  faculties.  Happi- 
ness is  action  guided  by  reason,  in  circumstances 
favourable  to  that  action. 

If  such  be  the  case,  the  element  that  constitutes 
happiness  is  doubtless  virtue  or  the  self-realisation  of 
the  higher  part  of  the  soul :  virtue  plays  the  part  of 
form  and  principle  as  regards  happiness.  But  happiness 
has  also,  as  material  to  work  upon  or  condition  of 
existence,  the  possession  of  external  forms  of  good  : 
health,  beauty,  birth,  fortune,  children  and  friends  ; 
although  it  is  true  that  even  the  greatest  of  misfortunes 
cannot  make  a  virtuous  man  really  miserable. 

Pleasure,  regarded  as  an  end,  is  not  an  integral 
element  of  happiness ;  since,  however,  it  naturally 
accompanies  action,  being  its  complement,  it  is  closely 
allied  to  virtue.  Pleasure  is  inherent  to  action  as 
vigour  is  inherent  to  youth.  It  is  the  consciousness 
of  activity.  The  value  of  pleasure  may  thus  be  gauged 
by  that  of  the  activity  it  accompanies.  Virtue  carries 
with  it  a  special  kind  of  satisfaction,  necessarily  pos- 
sessed by  the  virtuous  man.  Pleasures  are  admissible 
in  so  far  as  they  spring  from  virtue  or  can  be  reconciled 
therewith.  Coarse  or  violent  pleasures,  which  disturb 
the  soul,  ought  to  be  spurned.  In  a  word,  pleasure 
has  its  place  in  happiness  not  as  an  end,  but  ratKer 
as  a  result. 

Finally,  happiness  implies  leisure,  one  condition  of 
activity.  This  latter,  indeed,  needs  relaxation  ;  it  is 
not,  however,  leisure  that  is  the  end  of  work,  but  work 
that  is  the  end  of  leisure.  Leisure  should  be  devoted 
to  art,  science,  and  above  all,  philosophy. 


138  STUDIES  IN  PHILOSOPHY 

And  what  is  virtue,  the  principle  of  happiness? 
What  are  the  principal  virtues?  Virtue  is  a  habit 
whose  characteristic  is  the  complete  realisation  ot  the 
powers  of  man.  Now,  human  nature  is  two-fold,  to 
wifTTntellectual  and  rnoraL  The  intellectual  element 
has  thenecessary  for its""object,  and  is  immovable  ;  the 
moral  element,  in  so  far  as  it  is  connected  with  the  con- 
tingent, desires  and  acts.  Thus  there  are  two  kinds  of 
virtues  :  the  dianoetic  or  intellectual,  and  the  ethical  or 
moral. 

The  dianoetic  virtues  are  the  higher  of  the  two 
kinds  ;  they  can  only  be  acquired  by  instruction,  not 
by  an  effort  of  the  will.  The  virtue  that  affords  the 
greatest  felicity  is  science  or  contemplation.  This  is 
the  noblest  of  all  human  occupations,  for  the  vovs,  its 
organ  or  instrument,  is  the  most  divine  of  all  things. 
It  is  the  most  disinterested  activity,  the  one  that  causes 
least  fatigue,  and  most  readily  admits  of  continuity. 
And  it  is  the  calmest,  the  one  that  best  suffices  unto 
itself.  It  is  by  science  that  man  draws  nearest  to  divinity. 
Therefore  we  must  not  follow  the  advice  of  those  who 
maintain  that  we  should  have  only  human  feelings 
because  we  are  men,  and  only  aspire  after  the  destiny 
of  a  mortal  creature  because  we  are  mortal.  As  far 
as  in  us -lies,  we  should  do  our  best  to  make  ourselves 
worthy  of  immortality. 

Supreme  felicity,  however,  joined  to  the  possession 
of  perfect  science,  falls  but  seldom  to  the  lot  of  man. 
It  is  the  ethical  or  moral  virtues  that  are  truly  congenial 
to  him  and  adapted  to  his  condition  as  spirit  joined  to 
a  body.  Ethical  virtue  is  a  mental  habit  or  disposi- 
tion which  tends,  in  all  things,  to  choose  the  golden 
mean  suitable  to  our  nature,  and  is  determined  by  the 
practical  judgment  of  the  intelligent  man. 


ARISTOTLE  139 

It  is  a  habit,  a  mode  of  the  will.  Socrates,  who 
made  a  science  of  it,  forgot  that,  in  considering  virtue, 
we  have  nothing  to  do  with  the  knowledge  of  moral 
rules,  but  only  with  their  realisation.  Moreover,  to 
constitute  virtue,  there  is  needed  not  only  a  present 
determination  of  the  will,  but  rather  a  habit,  a  lasting 
mode  thereof. 

Again,  all  virtue  is  a  mean  between  two  vices,  and 
this  mean  varies  in  different  individuals.  Virtue  in 
a  man  is  different  from  virtue  in  a  woman,  a  child 
or  a  slave.  Time  and  circumstance  must  likewise  be 
taken  into  account.  Thus,  courage  is  the  mean  between 
rashness  and  cowardice ;  magnanimity  is  the  mean 
between  insolence  and  baseness,  and  so  on. 

Finally,  it  is  the  good  man  who  is  the  rule  and 
standard  of  the  good  in  each  particular  instance. 
Indeed,  abstract  rules  determine  only  what  is  good  in 
a  general  way.  In  each  instance  that  offers  itself  there 
is  something  unique  which  these  rules  neither  could 
nor  must  have  foreseen.  The  living,  universal  judg- 
ment of  the  highly  gifted  man  makes  up  for  their 
insufficiency. 

Aristotle  studies  in  detail  the  different  virtues,  both 
dianoe"tic  and  moral. 

The  dianogtic  virtues  are  the  perfect  habits  of  the 
intelligent  part  of  the  soul.  Now,  the  intellect  is  of 
two  kinds :  scientific  and  logistic.  The  virtues  of 
the  scientific  intellect  are  :  ist,  the  1/01)9,  which  knows 
the  principles  of  things ;  2nd,  science,  which,  from 
these  principles,  deduces  particular  truths.  The  union 
of  the  z/o£/s  and  science  constitutes  wisdom  (<ro<£ia). 
The  virtues  of  the  logistic  intellect  are  :  the  art  or 
capacity  of  producing  with  a  view  to  an  end  ;  2nd, 
judgment,  or  practical  intelligence. 


140  STUDIES  IN  PHILOSOPHY 

The  moral  virtues  are  as  numerous  as  the  different 
relations  in  human  life.  Since  the  number  of  these 
relations  is  indeterminate,  no  complete  list  of  the  moral 
virtues  is  possible  ;  a  fortiori^  these  virtues  cannot  be 
reduced  to  a  single  principle,  as  Plato  insists  upon. 
Aristotle  investigates  the  most  important  of  the  moral 
virtues.  His  dissertations  are  very  remarkable,  abound- 
ing in  keen  psychological  and  moral  observations.  His 
analyses  of  justice  and  friendship  are  particularly  deserv- 
ing of  mention. 

Justice,  he  says,  is  the  restoration  of  true  or  pro- 
portional equality  in  social  life.  Equity  is  more  perfect 
than  justice,  for  whereas  the  latter  takes  actions  into 
consideration  only  from  a  general  and  abstract  point  of 
view,  equity  takes  account  of  the  particular  element  in 
each  separate  action.  It  is  the  completion  of  justice, 
demanded  by  reason,  since  the  law  cannot  provide  for 
every  individual  case.  It  is  concrete,  actual  justice 
superposed  on  abstract,  and  still  indeterminate  justice. 

Friendship  is  supreme  justice,  delicate  and  perfect, 
wherein  a  blind,  dead  rule  is  entirely  replaced  by  the 
living  intelligence  of  the  good  man.  Friendship  has 
three  sources :  pleasure,  interest  and  virtue.  Virtue 
alone  creates  firm  and  lasting  friendships. 

XXI. — ECONOMICS 

Man,  in  family  life,  attains  to  a  degree  of  perfection 
superior  to  that  of  which  individual  life  admits.  The 
family  is  a  natural  society.  It  comprises  three  kinds  of 
relation  :  that  between  man  and  wife,  that  between 
parents  and  children,  and  that  between  master  and  slave. 

The  family  relation  between  man  and  wife  is  a  moral 
one,  based  on  friendship  and  mutual  service.  The  wife 


ARISTOTLE  141 

has  her  own  will,  her  own  virtue,  different  from  the 
man's  :  she  ought  to  be  treated  not  as  a  slave,  but  as 
a  free  person.  Still,  as  the  wife  is  less  perfect  than  the 
man,  the  latter  ought  to  have  authority  over  her.  The 
family  is  an  aristocracy  or  community  of  free  beings,  to 
whom  different  attributions  are  assigned.  The  wife, 
man's  free  companion,  ought  to  have  in  the  home  her 
own  sphere  of  influence,  with  which  man  does  not 
interfere. 

The  relation  between  parents  and  children  is  that 
between  a  king  and  his  subjects.  Parents  and  children 
form  a  monarchy.  As  regards  his  father,  the  child  has 
no  rights  whatsoever,  for  he  is  still  a  part  of  the  father ; 
it  is  the  father's  duty,  however,  to  watch  over  his 
child's  best  interests,  for  the  child  also  has  a  will  and 
a  virtue  of  his  own,  imperfect  though  they  be.  The 
father  should  transmit  his  own  perfection  to  his  son, 
and  the  latter  appropriate  to  himself  the  former's  per- 
fection. 

Aristotle  makes  a  special  study  of  slavery,  showing 
its  necessity  and  justifiableness,  and  determining  the  way 
in  which  slaves  ought  to  be  treated.  Slavery  is  neces- 
sary, for  the  home  has  need  of  living  and  intelligent 
workers.  And  slavery  is  justifiable.  Given,  indeed,  a 
being  fit  only  for  bodily  labour,  such  a  being  is  the 
justifiable  possession  of  one  who  is  capable  of  intellectual 
activity  ;  the  relation  of  the  former  to  the  latter  being 
that  of  matter  to  form.  Now,  such  a  relation  actually 
exists  between  the  Barbarians  and  the  Greeks.  Thus, 
the  free  man  is  owner  of  the  slave.  None  the  less 
ought  he  to  look  upon  the  slave  as  a  human  being,  and 
treat  him  as  such. 


1 42  STUDIES  IN  PHILOSOPHY 

XXIL— POLITICS 

Aristotle's   politics   deals  :    ist,   with    the   State   in 
general ;  2nd,  with  the  Constitutions. 

Politics  is  the  end  and  completion  of  economics,  as 
the  latter  is  the  proximate  end  of  morals.  The  indi- 
vidual, of  himself,  cannot  attain  to  virtue  and  happiness. 
Now,  the  tendency  towards  social  life  lies  in  the  very 
nature  of  man.  This  kind  of  life,  which  is  one  of  the 
conditions  of  human  existence,  is  likewise  a  means  of 
moral  improvement.  Politics,  which  sets  forth  the  ideal 
and  the  rules  relating  to  human  communities,  is  thus 
intimately  linked  with  morals  :  it  is  the  whole,  whereof 
morals  and  economics  are  but  parts ;  the  act,  of  which 
they  are  the  potency.  Politics  is  the  true  name  of  all 
practical  science.  Philosophy  should  set  forth  the 
ideal  of  politics ;  but  just  as  morals,  in  its  application, 
takes  individuals  into  account,  so  applied  politics  will 
take  circumstances  into  account. 

How  is  political  society  formed  ?  In  the  order  of  time, 
the  family  is  the  first  society  to  be  formed.  Then  we 
have  the  union  of  several  families,  or  the  K^^TJ.  Finally 
comes  the  State,  or  city  (-TroXt?)  :  the  highest  society  of 
all.  This  is  the  chronological  order ;  from  the  stand- 
point of  nature  and  truth,  however,  the  State  is  before 
individuals,  family  and  village,  as  the  whole  is  before  its 
parts  :  the  latter  having  in  the  former  their  final  cause 
and  loftiest  realisation. 

The  end  of  the  State  is  the  highest  that  can  be 
conceived,  for  the  State  is  the  most  perfect  expression 
of  the  social  tendency.  This  end  is  neither  the  mere 
satisfaction  of  physical  needs,  the  acquisition  of  wealth 
commerce,  nor  even  the  protection  of  the  citizens  by 
means  of  laws.  It  should  consist  in  the  happiness  of 


ARISTOTLE  143 

the  citizens.  It  is  the  mission  of  the  State  to  see  that 
its  citizens  possess,  first,  inner  good,  or  virtue,  and 
afterwards,  outer  good.  The  State  completes  the  pro- 
gress of  human  nature,  rising  from  potency  to  act. 

Although  in  agreement  with  Plato  as  regards  the 
final  good  of  politics,  Aristotle  is  none  the  less  led 
to  criticise  his  master  in  things  that  concern  the  rights 
and  duties  of  the  State.  He  opposes  the  Platonic 
doctrine  that  tends  to  dower  the  State  with  the  greatest 
possible  unity,  from  which  doctrine  resulted  the  necessity 
of  sacrificing  property  and  family  to  the  State.  Unity 
belongs  only  to  the  individual.  Already  the  family  has 
ceased  to  be  a  unit.  By  nature,  the  city  is  a  plurality,  and 
a  heterogeneous  one.  The  Platonic  theories  of  property 
and  the  family  cannot  be  admitted.  Not  only  are  they 
inapplicable  ;  they  even  misunderstand  both  the  tendency 
of  nature  and  the  interests  of  the  State.  Property  and 
the  family  are  by  no  means  artificial  products,  they  are 
the  objects  of  natural  tendencies.  Besides,  they  are 
useful  to  the  State,  procuring  for  it  advantages  it  could 
not  obtain  by  any  other  means.  The  State,  therefore, 
ought  to  regulate  property  and  the  family,  not  to  do 
away  with  them.  In  practice,  of  course,  Aristotle  often 
agrees  with  Plato,  whom  he  opposes  in  theory  ;  but  the 
conclusion  could  not  therefore  be  drawn  that  there  is 
no  difference  between  Platonic  and  Aristotelian  politics. 
The  importance  assigned  to  nature  in  the  latter  turns 
it  in  quite  another  direction. 

The  following,  then,  is  the  essential  tendency  of 
Aristotle's  politics.  As  supreme  good  lies  in  intellectual 
leisure,  the  useful  professions  are  incompatible  with  the 
title  of  citizen  :  farmers,  business  men,  workmen,  cannot 
be  members  of  the  city ;  of  an  ideal  one,  at  all  events. 
The  rtle  of  the  State  is  to  educate  its  citizens ;  its  efforts 


i44  STUDIES  IN  PHILOSOPHY 

are  directed  to  regulating  their  actions.  The  worst  of 
States  is  that  which  allows  every  man  to  live  as  he 
pleases.  The  State  regulates  the  age  and  the  season  for 
procreation,  fixes  the  number  of  the  population,  orders 
that  abortion  be  practised,  in  case  this  number  is  likely 
to  be  exceeded,  and  likewise  the  exposing  and  abandon- 
ing of  crippled  children.  Education  should  be  public, 
ever  keeping  in  view  the  good  of  the  intellect  through 
the  attention  bestowed  on  sensibility,  and  that  of 
the  soul  through  the  attention  bestowed  on  the  body. 
It  includes  grammar,  gymnastics,  music  and  drawing. 
In  all  things,  its  aim  is  to  form  the  moral  habits  of  the 
child.  It  is  essentially  liberal ;  such  arts  and  sciences 
as  are  of  a  mechanical  and  utilitarian  nature  being 
eliminated.  The  essential  virtue  of  the  State  is  justice, 
i.e.  the  order  by  virtue  of  which  each  member  of  the 
State  occupies  the  post  and  condition  of  life  suitable 
to  him,  and  is  entrusted  with  the  function  he  is  able  and 
worthy  to  exercise. 

The  maxim  by  which  the  Constitutions  ought  to  be 
regulated  is  as  follows  : — the  realisation  of  the  end  of 
the  State  presupposes  two  instruments :  laws  and  the 
magistracy.  The  true  sovereign,  the  only  ruler,  is  reason, 
order.  As  this  sovereign  or  ruler  is  invisible,  reason,  in 
practice,  must  be  represented  by  laws.  But  laws  are, 
of  necessity,  set  forth  in  general  formulae.  Now,  how- 
ever comprehensive  a  formula,  Tt  necessarily  allows  of 
an  infinity  of  particular  cases  escaping  through  its  toils. 
Hence  the  necessity  of  the  magistrate.  He  is  sovereign 
arbiter  whenever  the  law  is  unable  to  solve  a  difficulty, 
owing  to  the  impossibility  of  specifying  all  the  details 
of  the  case  under  general  regulations. 

Aristotle  does  not,  like  Plato,  lay  down  one  form  of 


ARISTOTLE  145 

government  as  being  good,  and  all  others  bad.  He  says 
that  the  Constitutions  ought  to  fit  in  with  the  character 
and  the  needs  of  the  nations  for  whom  they  are  framed ; 
that  the  one  which  is  worst  in  itself  may  be  the  best 
under  certain  circumstances.  He  also  examines  how 
bad  governments  may  be  utilised,  when  they  alone  are 
possible.  With  these  reservations,  he  classifies  the 
different  forms  of  government. 

There  are  three  kinds  of  government,  differing  in 
the  number  of  those  who  govern  :  power  may  be  in  the 
hands  either  of  one,  of  several,  or  of  the  majority  of  the 
nation.  Each  of  these  has  two  forms,  the  one  just,  the 
other  corrupt,  according  as  those  who  govern  have  in 
view  the  general  interest  or  their  own  private  interest. 
To  the  just  forms  of  government,  Aristotle  gives  the 
names  of  royalty,  aristocracy  and  polity ;  the  corrupt 
forms  he  calls  tyranny,  oligarchy  and  democracy. 

The  best  Twm  of  government  is  a  republic  which 
combines  order  with  freedom.  This  is  an  aristocracy. 
All  the  citizens  are  allowed  to  participate  in  public 
functions ;  only  those,  however,  are  citizens,  whose 
position  and  culture  enable  them  to  fulfil  civic  duties. 
All  corporal  toil,  especially  agriculture  and  the  various 
industrial  arts,  must  be  done  by  slaves  or  half- 
breeds. 

Lower  than  this  ideal  form  of  government  we  have 
forms  less  perfect,  though  justifiable  according  to  cir- 
cumstances. The  most  practical  of  these,  under  ordin- 
ary conditions,  is  a  temperate  republic,  a  mean  between 
democracy  and  oligarchy.  Democracy  is  characterised 
by  freedom  and  equality,  as  well  as  by  the  fact  that  the 
government  is  in  the  hands  of  the  majority  of  free  men 
and  of  the  poor.  In  an  oligarchy  the  government  is 
carried  on  by  a  minority  of  the  wealthy  and  the  noble. 


146  STUDIES  IN  PHILOSOPHY 

A  temperate  republic  bestows  power  on  the  middle 
classes.  It  is  the  political  equivalent  of  moral  virtue, 
which  is  the  mean  between  two  extremes. 

Evidently  Aristotle's  political  ideas  are  often  only 
the  putting  into  theory  of  the  facts  that  fall  under  his 
observation  ;  still,  it  would  be  an  exaggeration  to  see 
in  them  nothing  else.  Though  the  means  he  advocates 
are  frequently  the  result  of  a  necessarily  restricted  ex- 
perience, the  ends  he  has  in  view  are  determined  by 
reason  and  philosophy,  and  even  nowadays  Aristotle's 
politics  is  a  mine  of  information  for  statesmen  and 
historians. 

XXIII.— RHETORIC 

In  rhetoric,  Aristotle  tells  us,  he  had  nothing  to 
create,  for  this  science  had  been  developed  before  his 
time  by  Tisias,  Thrasymachus,  Theodorus  and  many 
others.  These  authors,  however,  confined  themselves 
to  the  particular,  never  going  beyond  the  empirical 
point  of  view.  To  Aristotle  belongs  the  idea  of 
scientific  rhetoric,  and  more  particularly  the  determina- 
tion of  a  close  connection  between  rhetoric  and  logic. 
Plato  had  unsuccessfully  endeavoured  to  base  rhetoric 
on  science.  Aristotle,  thanks  to  his  logical  theories, 
finds  in  dialectic,  as  distinguished  from  apodeictic,  the 
very  basis  of  rhetoric.  Rhetoric  is  the  application  of 
dialectic  to  politics,  i.e.  to  certain  practical  ends.  Logi- 
cally, dialectic  is  anterior  to  rhetoric ;  it  is  the  whole 
of  which  rhetoric  is  only  a  part.  In  the  order  of  time, 
rhetoric  is  anterior  to  dialectic  ;  but  in  the  order  of 
science,  it  is  the  contrary  that  holds  good. 

Rhetoric  teaches  persuasion  by  likely  reasons.  Thus, 
the  essential  part  of  rhetoric  is  the  doctrine  of  oratorical 


ARISTOTLE  147 

means.  These  are  of  three  kinds  :  ist,  those  referring 
to  the  object ;  2nd,  those  referring  to  the  speaker  ; 
3rd,  those  referring  to  the  listener. 

The  first  consist  in  making  affirmations  appear  true. 
They  are  based  on  proof.  Proof  is  thus  the  main 
element  in  rhetoric  ;  it  is  also  the  one  on  which  Aristotle 
insists  most.  As  dialectic  proves  by  means  of  syllogism 
and  induction,  so  rhetoric  proves  by  means  of  enthy- 
meme  or  imperfect  demonstration,  and  by  example  or 
imperfect  induction.  There  is  no  kind  of  proof,  it 
would  appear,  that  cannot  be  reduced  to  these  two 
arguments.  The  enthymeme  is  a  syllogism  in  which 
reasoning  is  carried  on  by  probabilities  or  signs.  Ex- 
ample, like  induction,  consists  in  judging  of  a  thing  by ' 
other  particular  things  similar  to  the  one  in  question, 
but  example  does  not  proceed  from  the  part  to  the 
whole,  it  proceeds  only  from  the  part  to  the  part. 
Rhetoric  determines  the  points  of  view  that  give  rise 
to  enthymemes  and  examples  :  this  determination  is  the 
object  of  oratorical  topic. 

Aristotle  distinguishes  three  kinds  of  speech  :  the 
deliberative,  the  legal  and  the  epideictic  ;  he  also  lays 
down  the  rules  governing  each. 

Such  are  the  oratorical  means  relating  to  the  object. 
The  speaker's  rtk  is  to  have  himself  regarded  as 
intelligent,  upright,  and  benevolent. 

Finally,  the  means  relating  to  the  listener  consist  in 
being  able  to  rouse  passion  and  to  lull  it  to  sleep. 
Aristotle  dwells  at  length  on  this  part  of  his  subject, 
giving  proof  of  a  very  shrewd  psychological  sense. 
He  makes  an  interesting  study  of  the  influence  of 
age  and  environment  on  character  and  disposition. 

Following  on  these  theories,  which  constitute  the 
basis  of  rhetoric,  come  studies  on  elocution  and  dis- 


148  STUDIES  IN  PHILOSOPHY 

position,  denoting  a  considerable  degree  of  truth  and 
sagacity  in  judgment,  along  with  a  profound  knowledge 
of  the  matter  in  hand. 


XXIV. — ESTHETICS 

Aristotle  divided  philosophy  into  three  parts  :  the 
theoretical,  the  practical  and  the  poetical,  or  the  one 
relating  to  art.  Though  he  made  no  attempt  to  develop 
this  latter,  the  proofs  and  examples  he  gives  show  him 
to  be  the  founder  of  esthetics. 

Aristotelian  esthetics  does  not  proceed  so  much  from 
the  concept  of  the  beautiful  as  from  that  of  art  ;  all  the 
same,  a  theory  of  the  beautiful  is  therein  outlined. 
According  to  Aristotle,  coordination,  symmetry  and 
precision  form  the  essential  characteristics  of  the 
beautiful.  Sensible  manifestation  is  not  an  essential 
element  of  the  beautiful,  which  shows  forth  as  being 
realised  more  especially  in  the  mathematical  sciences. 
The  beautiful  dwells  in  the  general.  Poetry,  which 
bears  upon  the  general,  is  more  beautiful,  more  serious 
and  philosophical  than  history,  which  is  contained  in 
the  particular. 

Aristotle,  like  Plato,  regards  imitation  as  the  essence 
of  art.  Art  results  from  man's  tendency  to  imitate 
and  the  pleasure  he  thereby  obtains.  What  man 
imitates  is  nature,  that  is  to  say,  according  to  the 
Aristotelian  philosophy,  not  only  the  outer  appearance, 
but  the  inner,  the  ideal  essence  of  natural  things.  Art 
is  capable  of  representing  things  as  they  are  or  as  they 
should  be.  The  representation  is  all  the  more  beautiful 
in  proportion  as  the  artist  proves  himself  able  to  com- 
plete, in  the  way  in  which  nature  herself  was  going,  the 
work  she  necessarily  leaves  unfinished.  All  art  tends 


ARISTOTLE  149 

to  represent  the  general  and  the  necessary.  This  is 
true  even  of  comic  poetry,  the  real  aim  of  which  is  the 
representation  of  characters. 

The  arts  include  more  than  one  kind  of  utility,  or 
service.  They  produce  distraction,  moral  culture,  intel- 
lectual enjoyment,  and  that  particular  effect  which 
Aristotle  calls  cleansing,  or  purification  (/edBapo-is). 
Purification  is  the  proprium  of  the  highest  arts,  more 
especially  of  serious  poetry. 

What  is  this  famous  purification  ?  It  is  not  exactly 
moral  improvement,  but  rather  the  suppression,  by 
homeopathic  treatment,  of  some  passion  that  troubled 
and  domineered  over  the  soul.  Moreover,  it  is  im- 
portant to  note  that  not  all  excitation  to  passion  is 
capable  of  producing  this  curative  effect.  Excitation 
of  a  salutary  nature  is  that  which  comes  from  art,  it  is 
subject  to  law  and  propriety,  and,  by  magnifying  the 
object  of  the  passions,  detaches  them  from  the  circum- 
stances of  individual  life  in  order  to  apply  them  to  the 
destiny  common  to  all  men. 

Aristotle  gives  no  systematic  classification  of  the 
arts,  the  highest  of  which,  according  to  him,  are  music 
and  poetry. 

XXV.— POETICS 

Almost  all  that  is  left  of  Aristotle's  Poetics  deals 
with  the  study  of  tragedy,  though  he  is  known  to  have 
dealt  fully  with  poetics. 

Poetry  arises  from  the  tendency  to  imitation.  A 
tragedy  is  the  imitation  of  a  serious  and  complete 
action,  of  a  certain  extent,  in  noble  language  and 
a  dramatic  form  devoid  of  narrative  :  an  imitation 
that  excites  terror  and  pity,  thereby  cleansing  the  soul 


150  STUDIES  IN  PHILOSOPHY 

of  these  passions.  In  the  persons  and  destinies  of  his 
heroes,  the  tragic  poet  offers  us  general  types  of  nature 
and  human  life.  He  shows  us  immutable  laws  which 
dominate  and  control  apparently  accidental  events. 
Hence  the  efficacy  of  tragedy  in  cleansing  the  soul  of 
all  its  inordinate  affections. 

The  most  important  part  of  tragedy  is  action. 
Action  ought  to  be  natural.  Not  that  the  author 
should  simply  set  forth  what  has  happened,  he  ought 
also  to  show  what  might  have  happened,  what  is  pos- 
sible either  according  to  the  laws  of  probability  or 
according  to  those  of  necessity.  Action  ought  to  be 
one  and  complete.  It  should  be  impossible  to  disturb 
or  curtail  any  part  of  the  work  without  disuniting  and 
spoiling  the  ensemble.  For,  in  any  whole,  that  which 
can  be  added  or  taken  away,  without  the  change  being 
noticed,  forms  no  part  of  that  whole. 

The  only  unity  on  which  Aristotle  insists  is  that  of 
action.  He  does  not  mention  unity  of  place,  and,  as 
regards  unity  of  time,  merely  states  that,  speaking 
generally,  in  tragedy  an  effort  is  made  to  confine  the 
action  within  a  single  day  or  to  go  beyond  that  limit 
but  slightly. 

He  determines  the  rules  that  refer  to  the  parts  of 
the  action,  to  the  characters,  which  ought  to  be  more 
finished  and  beautiful  than  they  are  in  real  life,  and  also 
to  composition  and  elocution. 

He  regards  tragedy  as  superior  to  epic  poetry 
because  its  unity  is  more  strict  and  confined,  whereas 
an  epic  poem  includes  parts,  each  one  of  which  would 
suffice  to  form  material  for  a  tragedy. 


ARISTOTLE 


XX  V  I. GR  A  M  M  A  R 


151 


In  ancient  times  Aristotle  was  looked  upon  as  the 
founder  of  grammar  and  criticism,  for  he  had  written 
works — now  lost — on  the  subject  of  poetical  explana- 
tion and  the  criticism  of  poets.  Such  indications  with 
reference  to  grammar  as  we  possess  are  not  given  for 
themselves,  but  only  as  they  affect  something  else. 
None  the  less  are  they  important  in  the  formation  of 
the  science  of  grammar.  Aristotle  applied  his  usual 
powers  of  observation  to  the  subject  of  grammar  ;  but 
the  theory  of  language  was  then  in  its  infancy  :  hence 
the  vagueness  and  obscurity  frequently  met  with  in  his 
assertions. 

He  recognises  three  parts  of  speech  :  noun,  verb 
and  conjunction.  The  two  former  are  subject  to 
inflection.  Nouns  are  divided  into  masculine,  feminine, 
and  neuter. 

Words  are  based  rather  on  mutual  agreement 
amongst  men  than  on  nature.  Subsequently,  in  their 
formation,  it  is  less  the  principle  of  analogy  than  the 
arbitrary  that  dominates. 

XXVII. — SPEECHES  AND  POEMS 

Several  speeches  of  Aristotle  are  mentioned,  includ- 
ing a  ^0705  SiKaviicos  or  Apology,  in  which  he  defends 
himself  against  the  accusation  of  impiety,  a  Eulogy  of 
Plato,  a  Eulogy  of  Alexander ;  but  the  authenticity  of 
these  works — now  lost — has  been  much  disputed. 

He  also  composed  poems,  a  few  authentic  lines  of 
which  remain,  though  many  fragments  are  of  very 
doubtful  authenticity.  The  most  important  of  these 
is  a  portion  of  a  scolion  in  honour  of  Hermias  of 


152  STUDIES  IN  PHILOSOPHY 

Atarnea,  his  friend.  Aristotle  here  sings  of  virtue, 
to  which,  like  the  ancient  heroes  of  Greece,  Hermias 
has  sacrificed  his  life.  Mention  may  also  be  made  of 
a  few  distachs  of  an  elegy  to  Eudemus,  composed  in 
honour  of  Plato,  "  a  man  whom  the  wicked  may  not 
even  praise." 

The  following  is  the  fragment  of  the  Scolion  to 
Hermias  : — 

Virtue,  object  of  effort  on  the  part  of  the  race  of  mankind, 
supreme  reward  of  life  !  For  thee,  O  virgin,  for  thy  beauty, 
the  Greeks  are  ready  to  brave  death,  to  endure  terrible,  never- 
ending  toil.  So  beautiful  is  the  fruit  thou  dost  engender  in 
the  heart,  immortal  fruit  more  precious  than  gold,  nobility  or 
soft-eyed  slumber  !  For  thee,  Hercules,  the  son  of  Zeus,  and 
the  sons  of  Leda  bore  many  a  trial,  for  they  were  noble  hunters 
in  pursuit  of  the  power  thou  bestowest.  Through  love  of  thee, 
Achilles  and  Ajax  entered  the  abode  of  Hades.  Thou,  too, 
wert  ever  the  object  of  the  love  of  Atarnes'  son  ;  for  the  sake 
of  thy  beauty  he  deprived  his  eyes  of  the  glorious  light  of  the 
sun.  That  is  why  he  is  praised  in  song  for  his  noble  deeds  ; 
the  Muses  shall  magnify  his  name  and  make  it  immortal,  the 
Muses,  Mnemosyne's  daughters,  who  honour  the  majesty  of 
Jupiter  the  protector  of  hospitality,  and  who  likewise  honour  the 
glory  of  faithful  friendship. 

XXVIII.— LETTERS 

Aristotle's  letters  have  been  celebrated  by  Demetrius 
and  other  authors  as  being  models  of  epistolary  style. 
Simplicius  states  that  the  style  of  these  letters  com- 
bined clearness  with  charm  of  diction  to  a  degree 
attained  by  no  other  known  writer.  Diogenes  men- 
tions letters  to  Philip,  the  letters  of  the  Selymbrians, 
four  letters  to  Alexander,  nine  to  Antipater,  and  others 
to  Mentor,  Ariston,  Philoxenes,  Democritus,  &c.  As 
the  fragments  that  have  come  down  to  us  are  for  the 


ARISTOTLE  153 

most  part  unauthenticated,  we  are  unable  to  judge 
for  ourselves  of  either  the  contents  or  the  form  of 
Aristotle's  letters. 


XXIX. — ARISTOTLE  AS  A  WRITER 

Aristotle  wrote  in  the  Attic  language  of  his  age. 
The  multitude  of  new  ideas  he  undertook  to  express, 
however,  had  a  considerable  influence  upon  the  instru- 
ment he  used.  The  consideration  of  things  in  their 
individuality,  the  clear  delimitation  of  scientific  domain, 
the  effort  to  form  concepts  exempt  from  every  sensible 
element,  are  all  reflected  in  his  language  and  style. 
As  Aristotle's  logical  analysis  only  ceases  when  it  has 
grasped  the  final,  specific  differences,  so  also,  in  Aris- 
totelian language,  apparent  synonyms  are  distinguished 
from  each  other  and  defined  with  great  preciseness. 

Aristotle  had  two  ways  of  defining  terms  :  the 
scientific  determination  of  the  meanings  of  traditional 
words,  and  the  creation  of  new  terms.  He  used  both 
methods,  especially  the  former.  He  mainly  starts  with 
an  ordinary  term ;  and  then,  sometimes  restricting, 
sometimes  extending  its  meaning,  he  makes  it  the  exact 
expression  of  a  logical  concept.  Traditional  language, 
however,  was  full  of  gaps.  To  fill  them  up,  Aristotle 
coined  words,  always,  as  far  as  possible,  seeking  a  basis 
to  work  upon  in  tradition  itself.  Owing  to  the  perfec- 
tion of  the  terminology  thus  constituted,  he  proved 
himself  the  true  founder  of  the  language  of  science 
throughout  the  world. 

The  following  are  instances  of  expressions  coined  by 
Aristotle  :   aSialperos  (individual)  ;  aireia-Qat  TO  ev 
(petitio  principii,    begging    of    the    question)  ; 
(immediate)  ;  avd\va-t,<s  (analysis)  ;  avo/j,oiopepr)s  (hetero- 


154  STUDIES  IN  PHILOSOPHY 


geneous)  ;  dvrtyacris  (contradiction)  ;  dTroSeircriKos  (de- 
monstrative) ;  a7ro^)aa-49  (affirmation)  ;  yevirco?  (generic)  ; 
SiXOTopia  (dichotomy)  ;  e'yu/Tret/H/co?  (empiric)  ;  evavnor^ 
(opposition)  ;  evepjeta  (energy)  ;  every?  (unity)  ;  eWe- 
Xe^eta  (entelechy)  ;  egwrepiKos  (exoteric)  ;  e7ra/cTt/co9 
(inductive)  ;  erepor^  (alterity  or  otherness)  ;  rj0uc6s 
(morals)  ;  0eo\ojtK^  (theology)  ;  KariyyopiKos  (categori- 
cal) ;  \oytKos  (logical)  ;  opyaviicos  (organic),  &c. 

The  following  instances  may  be  quoted  in  which 
Aristotle  confined  himself  to  a  scientific  determination 
of  the  meaning  of  the  term  :  avrlOea-is  (antithesis)  ; 
aglmjua  (axiom)  ;  evavrios  (contrary)  ;  ewTrdp^eiv  (to  be 
immanent)  ;  eTraywyij  (induction)  ;  ea-^arov  (last)  ; 
iSiov  (characteristic  property  of  a  species)  ;  o-vfifiefirjKos 
(accident)  ;  crv\\oY%ecr0ai  (to  reason)  ;  a-vve^?  (con- 
tinuous) ;  crvve^eta  (continuity)  ;  crvv6\ov  (whole)  ; 
v\t)  (matter)  ;  v-rrofceifjievov  (substratum). 

Finally  we  will  take  a  few  instances  of  the  distinc- 
tions he  draws  between  concepts,  by  means  of  analysis 
and  opposition  :  761/05  (genus)  ;  etSo?  (species)  ;  /az^o-t? 
(movement)  ;  evepyeia  (act)  ;  dvTi<j>ao-i<;  (contradiction) 
and  evavriov  (opposition)  ;  iroietv  (to  make)  and 
(to  do)  ;  Svva/Mi?  (potency)  and  evepyeia  (act)  ;  eT 
(induction)  and  o-v\\oyia-fj,6<;  (deduction)  ;  ovo-ia  (essence) 
and  <rv/j,{3€{3r)tc6Ta  (accidents)  ;  StaXe/crt/co?  (dialectic)  and 
aTToBeiKTiKos  (demonstrative)  ;  irporepov  ry  fyvaei  (anterior 
per  se)  and  Trporepov  Trpb?  ^a?  (anterior  from  our 
standpoint). 

Aristotle's  style  is  no  less  personal  than  his  language. 
The  ancients  extolled  his  fluency  and  charm  ;  the 
words  flowed  from  his  lips,  said  Cicero,  in  a  golden 
stream.  Such  praise  evidently  applies  to  his  dialogues, 
his  published  works.  In  his  didactic  works  (Trpay- 
which  alone  have  come  down  to  us,  we  note 


ARISTOTLE  155 

the  exactness  of  his  definitions,  inimitable  clearness, 
precision  and  brevity,  a  strictness  and  exactness  in  the 
meaning  of  words,  suggestive  of  the  language  of 
mathematics.  In  a  word,  Aristotle's  style  is  dis- 
tinguished by  an  exact  appropriation  of  form  to  content. 
Frequently,  however,  especially  in  such  of  his  works  as 
are  incomplete,  Aristotle  writes  with  a  certain  degree  of 
aridity  and  carelessness.  Not  only  are  the  sentences 
not  arranged  in  periods,  but  there  are  numerous  anaco- 
lutha  and  parentheses,  which,  in  no  small  measure, 
militate  against  clearness.  At  times,  too,  in  these 
abstract  dissertations,  we  come  across  passages  that  are 
not  lacking  in  fire  and  eloquence.  Of  such  a  character 
is  the  end  of  chapter  7,  book  10,  of  the  Nicomachean 
Ethics  : 

The  life  of  the  statesman  and  of  the  soldier,  then,  though 
they  surpass  all  other  virtuous  exercises  in  nobility  and  grandeur, 
are  not  leisurely  occupations,  but  aim  at  some  ulterior  end,  and 
are  not  desired  merely  for  themselves. 

But  the  exercise  of  the  reason  seems  to  be  superior  in 
seriousness  (since  it  contemplates  truth),  and  to  aim  at  no  end 
beside  itself,  and  to  have  its  proper  pleasure  (which  also  helps 
to  increase  the  exercise) ;  and  its  exercise  seems  further  to  be 
self-sufficient,  and  leisurely,  and  inexhaustible  (as  far  as  anything 
human  can  be),  and  to  have  all  the  other  characteristics  that 
are  ascribed  to  happiness. 

This,  then,  will  be  the  complete  happiness  of  man,  i.e. 
when  a  complete  term  of  days  is  added  ;  for  nothing  incomplete 
can  be  admitted  into  our  idea  of  happiness. 

But  a  life  which  realised  this  idea  would  be  something  more 
than  human  ;  for  it  would  not  be  the  expression  of  man's 
nature,  but  of  some  divine  element  in  that  nature — the  exercise 
of  which  is  so  far  superior  to  the  exercise  of  the  other  kind  of 
virtue  (i.e.  practical  or  moral  virtue),  as  this  divine  element  is 
superior  to  our  compound  human  nature.1 

1  I.e.  our  nature  as  moral  agents,  as  compounds  of  reason  and  desire. 


156  STUDIES  IN  PHILOSOPHY 

If,  then,  reason  be  divine  as  compared  with  man,  the  life 
which  consists  in  the  exercise  of  reason  will  also  be  divine  in 
comparison  with  human  life.  Nevertheless,  instead  of  listening 
to  those  who  advise  us  as  men  and  mortals  not  to  lift  our 
thoughts  above  what  is  human  and  mortal,  we  ought  rather, 
as  far  as  possible,  to  put  off  our  mortality  and  make  every 
effort  to  live  in  the  exercise  of  the  highest  of  our  faculties  j 
for  though  it  be  but  a  small  part  of  us,  yet  in  power  and  value 
it  far  surpasses  all  the  rest.1 


XXX. — ARISTOTLE'S  INFLUENCE 

The  first  effect  of  Aristotle's  teaching  was  to  bring 
into  being  the  Peripatetic  school,  which  flourished  for 
a  period  of  from  two  to  three  centuries,  and  whose 
principal  representatives  are  :  Theophrastus  of  Lesbos 
(372.^287  ?  B.C.),  Eudemus  of  Rhodes  (fourth  century), 
Aristoxenus  of  Tarentum  (born  about  350  B.C.),  sur- 
named  the  Musician,  Decearchus  of  Messena  (flourished 
320  B.C.)  and  Strato  of  Lampsacus  (flourished  287  B.C.). 
Critolaus,  a  member  of  the  embassy  sent  to  Rome 
in  156  B.C.,  by  which  philosophy  was  introduced  into 
the  Roman  world,  was  a  Peripatetic  philosopher.  The 
school  was  distinguished  for  its  minute  investigations  in 
logic,  morals  and  natural  science,  but  the  naturalistic 
tendency  gradually  prevailed  over  the  metaphysical. 
Strato  even  went  so  far  as  to  identify  divinity  with  the 
<f>va-is  which  acts  unconsciously  throughout  the  world, 
and  to  substitute  for  the  Aristotelian  teleology  an 
altogether  mechanical  explanation  of  things,  based  on 
the  properties  of  heat  and  cold. 

With  the  publication  of  Aristotle's  works  by 
Andronicus  of  Rhodes,  about  70  B.C.,  began  the  long 
list  of  interpreters  and  commentators  of  the  Stageirite, 

1  F.  H.  Peter's  translation. 


ARISTOTLE  157 

including  Boethus  of  Sidon,  Nicolas  of  Damascus, 
Alexander  of  Aphrodisias  in  Cilicia,  surnamed  the 
Exegete  far  excellence  (icar  egoxtfv),  Porphyry  of  Bat- 
anaea,  the  Neoplatonist,  Themistius  of  Paphlagonia, 
Philopon  of  Alexandria  and  Simplicius  of  Cilicia. 

Though  the  Peripatetic  school  consists  mainly  of 
disciples  not  very  advanced  in  metaphysics  or  of  purely 
erudite  commentators,  still,  the  master's  doctrines  are 
very  vigorous  and  instinct  with  life  in  philosophies 
which  did  not  originate  with  him  but  were  largely 
inspired  by  his  influence.  The  principle  of  the  Stoics, 
intermediary  between  potency  and  act,  and  limited  by 
tension,  immanent  in  all  things,  the  intelligent  and 
supreme  final  cause,  would  indeed  appear  to  be  nothing 
else  than  the  <f>v<n,s  of  Aristotle,  into  which  the  vovs  would 
seem  to  be  absorbed.  Through  the  precise  distinction 
he  made  between  mechanism  and  finality,  between  the 
physical  and  the  metaphysical  order  of  things,  between 
chance  and  intelligent  action,  Aristotle  rendered  possible 
Epicureanism,  which  seems  largely  to  be  made  up  of 
the  doctrines  which  Aristotle  defined  or  created  for  the 
purpose  of  refuting  them.  Neoplatonism  itself,  in  the 
matter  of  its  doctrine  regarding  the  vovs,  is  greatly 
indebted  to  Aristotle.  The  Neoplatonists  endeavoured 
to  reconcile  Plato  and  Aristotle  ;  and  Plotinus  main- 
tained that  his  doctrine  of  the  transcendent  one  from 
which  the  vov?  emanates,  was  the  inevitable  consequence 
of  Aristotelian  teaching. 

After  defending  ancient  philosophy  to  the  very  end, 
Aristotelianism,  becoming  embodied  in  the  beliefs  of 
the  Middle  Ages,  transformed  them  into  philosophical 
doctrines.  It  was  mainly  owing  to  the  influence  of 
Aristotle  that  there  developed,  in  that  period  of  religious 
mysticism,  the  spirit  of  logic  and  of  rational  speculation. 


158  STUDIES  IN  PHILOSOPHY 

Tardily  and  indirectly  did  Aristotle's  writings  pene- 
trate into  the  western  world.  Even  in  the  middle  of 
the  twelfth  century,  only  small  portions  of  the  Organon 
were  known,  to  wit,  the  Categories  and  the  Hermeneia, 
in  the  Latin  translation  of  Boetius.  These,  along  with 
the  EtVo7&>y?7  of  Porphyry  and  the  Timaeus  of  Plato, 
formed  almost  the  entire  possessions  of  philosophical 
antiquity.  From  A.D.  1150  to  1210,  the  other  works 
of  Aristotle  appeared  in  the  form  of  a  Latin  version  of 
Arabic  translations,  which  in  their  turn  had  been  trans- 
lated by  Christian  Syrians,  from  Syriac  translations,  in 
the  ninth  century.  Shortly  afterwards  (thirteenth  cen- 
tury), the  Greek  text  was  communicated  to  the  scholars 
of  the  West,  mainly  by  Greeks  from  Constantinople  ; 
and  a  translation  direct  from  the  Latin  was  substituted 
for  the  indirect  translations.  Robert  Greathead,  Albert 
le  Grand  and  Saint  Thomas  were  the  principal  persons 
engaged  in  this  refining  process  of  translating  into  Latin. 

As  showing  how  dependent  on  his  will  is  man's 
intelligence,  people  of  the  most  diverse  opinions, 
strangely  enough,  found  in  Aristotle  a  rational  basis 
for  their  beliefs  and  aspirations.  There  could  be 
nothing  apparently  more  one  than  the  Middle  Ages, 
for  Aristotle  was  invoked  by  everybody,  though,  as  a 
matter  of  fact,  there  were  as  many  Aristotles  as 
philosophers.  There  were  even  Aristotles  who  had 
only  the  name  in  common  with  the  Stageirite. 

It  was  Aristotle's  Organon  that  gave  rise  to  the 
famous  quarrel  between  the  universities,  which  lasted 
from  the  ninth  to  the  end  of  the  eleventh  century. 
About  this  time,  complete  systems  of  Aristotelian 
philosophy  grew  up  amongst  the  Arabs  and  Jews,  who 
had  possession  of  all  the  master's  writings.  The  Arabs, 
who  were  naturalists  and  monotheists,  were  captivated 


ARISTOTLE  159 

by  Aristotle's  teachings  about  God  and  by  his  investi- 
gations into  natural  history.  AverroSs,  of  Cordova 
(A.D.  1126-1198),  regards  himself  as  a  true  Aristotelian 
when  maintaining  that  active  understanding  is  an 
emanation  from  God,  that  it  is  one  for  all  men  and 
alone  is  immortal.  Moses  Maimonides,  a  Jew  of 
Cordova  (A.D.  1135-1204),  finds  no  difficulty  in  re- 
conciling miracles  and  the  creation  of  matter  with 
Aristotelianism. 

The  most  brilliant  period  of  Christian  scholasticism 
is  also  that  during  which  Aristotle's  authority  is  at  its 
highest.  Though  his  doctrines  on  physics,  which  are 
regarded  as  advocating  the  eternity  of  the  world  and  of 
time,  are  for  a  certain  period  regarded  with  suspicion, 
from  the  year  A.D.  1230,  the  whole  of  his  works  begin 
to  be  used  as  text-books  for  lessons  in  philosophy.  Just 
as  the  truths  of  faith  are  the  expression  of  supernatural 
illumination,  so  the  Aristotelian  doctrine  is  the  ex- 
pression of  natural  illumination.  Reason  does  not 
coincide  with  faith,  but  it  is  moving  towards  it. 
Aristotle,  as  representing  reason,  is  the  forerunner  of 
Christ  in  the  things  of  nature,  as  Saint  John  the  Baptist  is 
his  forerunner  in  those  of  grace.  Thus  defined,  circum- 
scribed and  subordinated,  Aristotelianism  becomes  the 
origin  of  what  has  since  been  called  deism  and  natural 
religion.  At  that  time  there  was  found  in  it  all  that 
theology  required.  Naturally  it  cannot  demonstrate 
the  truth  of  the  dogmas,  for  that  would  be  contradictory  ; 
still,  it  refutes  objections  brought  against  them  and 
establishes  their  probability.  In  particular,  it  sets  up  a 
theory  of  substantial  form  and  of  real  and  separable 
accidents,  which  makes  transubstantiation  conceivable 
in  the  persistence  of  the  same  sensible  elements  in  the 
Eucharist. 


i6o  STUDIES  IN  PHILOSOPHY 

And,  indeed,  Aristotelianism  is  as  favourable  to  dis- 
sent as  it  is  to  orthodoxy.  Amaury  of  Chartres  and 
David  of  Dinant  (thirteenth  and  fourteenth  centuries) 
claim  that  it  upholds  pantheism,  for  the  one  identifies 
the  God  of  the  Stageirite  with  form,  the  other  with 
universal  matter.  The  German  mystics,  too,  Theodoric 
of  Freiburg  and  "  Meister  Eckhart  "  (thirteenth  and 
fourteenth  centuries),  present  their  doctrine  of  the 
substantial  union  of  the  soul  with  divinity,  as  the 
development  of  the  Aristotelian  theory  of  the 


And  lastly,  Aristotle  is  not  only  the  master  of 
philosophers  in  the  Middle  Ages  ;  he  is  even  regarded 
as  the  patron  of  those  who,  in  opposition  to  the  Church 
and  the  philosophy  of  the  times,  claim  to  harness  and 
control  the  mysterious  forces  of  nature.  These  re- 
probates look  upon  Aristotle  as  a  magician.  He  is 
credited  with  having  written  alchemical  treatises  on  the 
occult  philosophy  of  the  Egyptians,  and  is  placed,  with 
Plato,  at  the  head  of  the  list  of  oecumenical  alchemists. 
Alchemists  called  themselves  the  new  commentators  of 
Plato  and  Aristotle. 

Thus  we  find  Aristotle,  in  the  Middle  Ages,  every- 
where stirring  up  the  minds  of  men  and  regarded  as  an 
authority  :  his  main  work,  however,  was  undoubtedly 
the  organisation  of  that  Christian  philosophy  which  was 
so  complete  and  detailed,  so  logical  and  firmly  based 
throughout,  that  it  seemed  destined  to  last  for  ever.  This 
philosophy  held  sway  in  the]  colleges  of  the  University 
of  France  up  to  the  eighteenth  century.  In  the  Sorbonne, 
in  1624,  it  was  forbidden,  under  penalty  of  death,  to  pro- 
pound doctrines  opposed  to  those  held  by  the  ancients. 
Even  in  1671,  the  professors  were  called  upon  to 
respect  Peripateticism  under  penalty  of  exclusion.  Only 


ARISTOTLE  161 

at  the  beginning  of  the  eighteenth  century  did  scholastic 
Aristotelianism  make  way  for  new  ideas. 

It  was  from  faith,  not  from  reason,  that  the  first 
really  savage  attack  came.  Not  only  did  Luther 
note  how  important  were  the  differences  that  divided 
Aristotelian  philosophy  from  Christianity,  he  even  re- 
garded it  as  impious  to  seek  for  a  reconciliation 
between  God-given  faith  and  sin-stained  reason.  Aris- 
totelian philosophy,  the  work  of  man,  with  its  claim  to 
deal  with  things  divine,  could  be  nothing  else  than 
error  and  sacrilege  ;  religion,  once  reconciled  thereto, 
could  only  become  distorted  and  misrepresented.  Aris- 
totle was  an  arch-heretic  :  religion  would  only  be  safe 
on  condition  his  doctrines  were  utterly  abolished. 

Opposed  in  the  name  of  the  Christian  religion, 
Aristotelianism,  in  spite  of  its  glorious  revival  by  the 
scholars  of  the  Renaissance,  Pomponatius,  Scaliger, 
Vanini,  Gennadius,  and  George  of  Trebizond,  speedily 
became  an  object  of  attack  by  science  and  philosophy. 
Bacon  saw  in  the  Aristotelian  method  nothing  but 
deduction  applied  to  the  data  of  opinion  and  language  ; 
in  his  eyes,  Aristotelian  metaphysic  was  only  the  claim 
to  explain  things,  exclusive  of  mechanical  causes,  by 
supernatural  and  divine  actions.  He  therefore  con- 
demned the  philosophy  of  Aristotle,  as  being  contrary 
to  the  conditions  of  science,  which  latter  seeks 
mechanical  explanations  and  proceeds  by  induction. 
Descartes  looked  upon  Aristotelianism  as  the  doctrine 
that  realised  sensible  qualities,  and  explained  phenomena 
by  these  chimerical  entities.  Barren  and  obscure  ideas, 
these  abstractions  could  not  possibly  be  the  principles 
of  things.  In  direct  opposition  to  Aristotle,  Descartes 
restores  quality  to  quantity,  not  quantity  to  quality. 

It  appeared  as  though  the  Aristotelian  doctrine  would 

M 


1 62  STUDIES  IN  PHILOSOPHY 

have  definitely  lived,  when  Leibnitz  triumphantly  re- 
stored it  to  philosophy,  declaring  that  in  the  theory  of 
substantial  forms  and  entelechy,  when  rightly  under- 
stood, there  was  more  truth  than  in  the  entire  philo- 
sophy of  the  moderns.  Following  in  the  steps  of 
Aristotle,  Leibnitz  placed  substance  in  a  principle  of 
action,  relegated  extent  and  matter  from  the  class  of 
substance  to  that  of  phenomenon,  and  reconciled  final 
with  efficient  causes  by  making  mechanism  dependent  on 
finality.  Aristotelianism,  since  the  time  of  Leibnitz, 
has  maintained  a  place  of  its  own  in  philosophy,  more 
particularly  playing  an  important  part  in  the  formation 
of  the  Hegelian  system. 

However  great  his  place  in  history,  can  it  be  said 
that  Aristotle,  even  at  the  present  time,  is  one  of  the 
masters  of  human  thought  ? 

As  regards  philosophy  strictly  so  called,  there  can 
be  no  question  as  to  the  answer  that  must  be  given. 
It  appears  as  though  Aristotelianism  responds  particu- 
larly to  the  preoccupations  of  modern  times.  The 
two  doctrines  that  until  recently,  have  occupied  the 
largest  place  in  the  world  of  philosophy  were  Kantian 
idealism  and  evolutionism.  Now,  Aristotle's  system 
may  without  disadvantage  be  set  up  against  these  two 
systems. 

It  is  opposed  to  Kantism.  As  a  matter  of  fact, 
Kant  rejects  the  dependence  of  the  mind  in  respect  of 
being,  the  ontological  value  attributed  to  the  laws  of 
the  mind,  the  theoretical  unconditioned  and  the  sub- 
ordination of  practice  to  theory  ;  all  of  which  belong  to 
the  very  essence  of  Aristotelianism.  The  philosophy  of 
Kant  has  been  set  up  in  direct  opposition  to  dogmatic 
philosophy,  of  which  Aristotle  is  the  representative 


ARISTOTLE  163 

par  excellence.  But  if  Kant  discovered  a  new  conception 
of  things,  a  conception  which  must  henceforth  be 
examined  by  all  interested  in  philosophy,  it  cannot  be 
affirmed  that  he  fully  succeeded  in  getting  his  hypo- 
thesis accepted  universally.  If  this  hypothesis  has 
on  its  side  the  testimony  of  conscience,  which,  by  the 
way,  it  undertakes  to  satisfy,  it  cannot  obtain  the 
frank,  complete  approval  of  the  intellect.  This  latter 
persists  in  saying,  with  Aristotle  :  "  Everything  has  a 
reason  of  its  own,  and  the  first  principle  must  be  the 
final  reason  of  things.  Now,  explanation  implies  deter- 
mination, and  the  final  reason  cannot  be  anything  else 
than  fully  determined  being.  When  we  consider  the 
infinite  and  the  finite,  it  is  the  finite,  in  so  far  as  it  is 
intelligible,  that  is  the  principle  ;  the  infinite,  in  so 
far  as  it  is  unintelligible,  can  only  be  phenomenon."  As 
regards  Aristotle  and  Kant,  what  we  have  to  do  is  to 
find  out  whether  the  supremacy  must  be  attributed  to 
the  will  or  to  the  intellect ;  now,  even  at  the  present 
time,  this  question  does  not  appear  to  have  been 
answered  once  for  all. 

The  position  of  Aristotelianism  as  compared  with 
evolutionism  is  quite  different.  Not  only  does  it  not 
oppose  the  latter,  it  even  recognises  and  includes 
it,  at  the  same  time  affording  the  means  of  going 
beyond  it.  Historically,  it  is  one  of  the  most  direct 
antecedents  of  evolutionism.  Whether  in  nature  or  in 
man,  Aristotle  shows  that  everywhere  we  have  con- 
tinuity— a  process  of  development  from  the  lower  to 
the  higher.  Plants  imply  minerals,  animals  imply 
plants,  man  implies  animals,  and  man  is  nothing  but 
the  completion  of  the  being  roughly  outlined  in  the 
lower  productions  of  nature.  Even  in  man,  imagination 
springs  from  sensation,  memory  from  imagination,  and 


1 64  STUDIES  IN  PHILOSOPHY 

the  intellect  cannot  think  without  images.  We  can  find 
no  scientific  thesis  of  evolutionism  that  would  be  in- 
compatible with  the  natural  philosophy  of  Aristotle. 
But  is  this  mechanical  order  of  things  the  absolute 
order  ?  Do  these  explanations  fully  satisfy  the  in- 
telligence ?  This  is  the  question  Aristotle  asks,  a 
question  he  finds  it  impossible  to  answer  along  the 
lines  of  spiritualistic  metaphysics. 

To  our  philosopher,  the  order  which  proceeds  from 
the  indeterminate  to  the  determinate,  from  genus  to 
species,  cannot  be  regarded  by  the  intellect  as  the 
absolute  order  of  the  generation  of  things,  for  the  in- 
determinate always  admits  of  other  determinations  than 
those  it  receives  in  the  real  world.  Though  man  is  the 
completion  of  the  animal,  still,  the  animal  admitted  of 
other  determinations  than  those  that  made  it  into  a  man. 
Why  do  genera  find  their  realisation  in  certain  species 
rather  than  in  others  ?  The  reason  of  this  choice  from 
amongst  all  possible  developments  can  be  found  only 
in  the  very  being  which  is  the  term  of  the  development. 
The  perfection  of  this  being  must  be  a  force  controlling 
the  evolution  of  the  matter  from  which  it  is  to  be  born. 
In  this  way,  the  order  which  proceeds  from  the  inde- 
terminate to  the  determinate  does  not  exclude  ;  it  calls 
for  a  symmetrically  contrary  order,  the  hidden  principle 
of  its  direction  and  realisation.  And  so  Aristotle 
reconciles  the  evolutionistic  mechanism  with  finality  by 
making  a  distinction  between  the  order  of  things  in 
time  and  that  of  things  in  the  absolute.  Evolutionism 
is  truth  from  the  standpoint  of  the  senses  ;  from  that 
of  the  intellect,  however,  the  imperfect  exists  and  is 
determined  only  with  a  view  to  the  more  perfect. 
The  finalistic  explanation  is  the  justifiable  and  indis- 
pensable complement  of  the  mechanistic  one. 


ARISTOTLE  165 

Thus  Aristotelianism  still  has  a  place  of  its  own  in 
philosophy.  But  has  it  not  become,  for  the  future, 
banned  and  barred  from  science  ? 

Here  a  distinction  must  be  made  between  the 
moral  sciences,  on  the  one  hand,  and  the  mathematical 
and  physical  sciences  on  the  other.  Aristotle's  ethics, 
and  even,  in  many  important  respects,  his  politics,  far 
from  being  forgotten,  are  in  greater  vogue  than  ever 
nowadays.  The  recommendation  to  live  as  a  man 
when  one  is  born  a  man,  and  to  attribute  real 
sovereignty  in  politics  to  reason  and  law,  are  by  no 
means  on  the  point  of  sinking  into  oblivion.  But  the 
sciences  dealing  with  nature,  all  henceforth  positive, 
seem  to  have  little  in  common  with  the  natural  philo- 
sophy of  the  great  metaphysician. 

In  order  to  express  a  fair  judgment  on  this  subject, 
it  should  at  once  be  stated  that  a  man  may  have  exercised 
great  influence  on  the  development  of  the  sciences 
without  any  of  his  ideas  being  recognised  in  present- 
day  teachings.  The  sciences  are  built  up  stage  by 
stage  ;  and  though  some  particular  ancient  theory  may 
not  be  recognised  in  modern  theories,  it  may  well  have 
played  its  part  in  paving  the  way  for  their  reception. 
Now,  merit  of  this  kind  may  certainly  be  attributed  to 
Aristotle.  He  advanced  theories  and  concepts  which 
may  be  vastly  different  from  modern  methods  and 
principles,  and  yet  have  none  the  less  controlled  the 
formation  of  these  very  principles.  For  instance,  we 
have  the  Aristotelian  theory  of  induction  which  doubtless 
determines  rather  the  end  to  be  attained  than  the  means 
to  be  employed,  and  prefers  to  regard  this  end  as  being 
the  discovery  of  types  and  not  that  of  laws,  but  which 
is  none  the  less  very  important  because  of  the  precision 
with  which,  in  induction,  it  shows  how  we  have  to  set 


1 66  STUDIES  IN  PHILOSOPHY 

free  the  necessary  from  the  contingent,  the  universal 
from  the  particular.  Such  also  are  the  ideas  of  genus 
and  species,  potency  and  act,  mechanical  blending  and 
qualitative  combination,  chance,  in  reference  to  the  con- 
junction of  causes  independent  of  one  another,  con- 
tinuity in  the  scale  of  beings,  classification  of  the 
sciences,  etc. 

But  the  simple  acknowledgment  that  Aristotle  has 
supplied  science  with  many  starting  -  points  is  not 
sufficient.  Many  of  his  principles  may  still  quite  well 
be  recognised  in  the  spirit  of  contemporary  science 
itself.  His  great  principle  that  there  are  laws  in 
nature  and  that  they  can  be  discovered  only  by  deducing 
them  from  experience  by  the  aid  of  reflection,  his 
constant  wish  to  investigate  things  in  their  details,  to 
understand  them  not  by  means  of  vague  formulae,  but 
in  themselves  with  their  own  characteristics,  his  definition 
of  cause  as  existing  in  that  element  which  makes  pro- 
duction known  as  necessary,  his  doctrine  of  biological 
continuity  and  of  the  solidarity  of  the  higher  with 
respect  to  the  lower ;  all  these  essential  features  of 
Aristotelian  philosophy  may  be  met  with  in  modern 
science.  Though  an  authority  belonging  to  the  past, 
Aristotle  has  not  ceased  to  be  a  master,  even  in  these 
days. 

The  objection,  however,  will  be  urged  that  Aristotle 
is  finalistic  and  that  science  does  not  now  trouble  itself 
with  the  consideration  of  ends. 

Perhaps  there  is  some  misunderstanding  here. 
Aristotelian  finality  is  not  the  building  up  of  the  world, 
as  though  it  were  a  watch,  by  an  artisan  who  sets 
before  him  an  idea  and  calculates  how  to  realise  it.  It 
consists,  we  may  say,  of  the  three  following  principles  : 
ist,  throughout  the  world,  order  is  the  rule,  disorder 


ARISTOTLE  167 

the  exception  ;  this  is  equivalent  to  saying  that  the 
combinations  of  phenomena  which  result  immediately 
from  the  laws  of  nature,  harmoniously  united  in  types, 
and  consequently  normal  in  their  development,  are  far 
more  numerous  than  the  combinations  due  to  the 
fortuitous  conjunction  of  laws  independent  of  one 
another  ;  2nd,  in  every  individual  there  is  an  organising 
force  or  Averts  by  virtue  of  which  it  tends  to  be  and  to 
realise  a  certain  form  ;  3rd,  the  specific  types  are  strictly 
determined,  separated  from  one  another,  and  immut- 
able. Is  it  quite  certain  that  finality,  thus  interpreted, 
is  altogether  absent  from  modern  science  ? 

The  first  of  these  three  principles  signifies  that  it  is 
possible  to  obtain  knowledge  of  fundamental  laws  by 
means  of  observation  and  induction.  In  contrast  with 
this  theory  we  have  the  mathematical  theory  of  Descartes, 
according  to  which  there  are  really  no  qualitative  and 
multiple  laws  of  nature,  but  only  various  determina- 
tions of  homogeneous  and  mathematical  quantity. 
But  though  we  have  the  Cartesian  conception  as  an 
ideal  representing  complete  science,  the  Aristotelian 
method  of  advance  is  still  the  one  best  suited  to  our 
means  of  knowledge.  The  only  thing  in  which  Aristotle 
erred  was  in  imagining  that  by  the  process  of  induction 
we  could  arrive  at  simple  and  absolute  laws  which 
presuppose  nothing  anterior  to  themselves. 

The  second  principle  bears  a  striking  resemblance 
to  that  of  the  struggle  for  life.  Here,  too,  we  pre- 
suppose in  every  individual  a  tendency  to  exist  and 
develop  along  fixed  lines.  It  is  true  that  modern 
science  would  like  to  reduce  life  itself  to  a  mechanism  ; 
all  the  same,  it  acknowledges  that  life,  as  we  find  it, 
plays  the  part  and  possesses  the  characteristics  that 
Aristotle  attributed  to  it.  The  entire  difference  con- 


1 68  STUDIES  IN  PHILOSOPHY 

sists  in  regarding  as  derived  what  Aristotle  looked  upon 
as  primitive  ;  but  until  this  reduction  is  effected,  we  do 
not  think  we  are  wrong  in  saying:  everything  takes 
place  as  though  there  were  in  each  living  being  a 
tendency  to  exist,  and  that  in  some  determinate  manner. 
Finally,  the  third  principle,  which  still  counts 
adherents  amongst  scientists  themselves,  is  not,  as 
Aristotle  understood  it,  in  absolute  contradiction  to  the 
teachings  of  the  evolutionists,  from  the  physical  point 
of  view.  What  is  it  that  Aristotle  means  ?  He  does 
not  wish  to  affirm  that  the  history  of  the  beings  of 
nature  began  in  time,  with  the  creation  of  separate 
species :  he  means  that  the  realisation  of  a  certain 
number  of  types,  both  distinct  from  and  in  harmony 
with  each  other,  is  the  end  and  rule  of  the  productions 
of  nature.  He  admits  that  nature,  for  the  most  part, 
succeeds  in  realising  this  end  ;  but,  apart  from  the 
perfectly  regular  productions  of  nature,  he  acknow- 
ledges productions  partly  regular,  partly  irregular. 
Now,  if  we  leave  the  past  out  of  account,  and  also  any 
beginning  in  time,  about  which  Aristotle  did  not  trouble 
himself,  we  shall  find  no  very  great  divergence  between 
this  point  of  view  and  that  of  evolutionism.  In  contra-, 
distinction  to  materialism  and  the  doctrine  of  chance, 
evolutionism  recognises  that  species  exist,  at  the  present 
time,  at  all  events.  It  also  recognises  the  tendency  in 
nature  towards  an  increasingly  complete  specification. 
The  principle  of  Aristotle,  then,  subsists,  even  in  these 
days,  in  the  hypothetical  form  at  any  rate,  the  only 
form  a  principle  can  admit  of  in  science  ;  everything 
takes  place  as  though  there  were  a  hierarchy  of  ideal 
forms,  distinct  from  one  another,  and  which  the  beings 
of  nature  tend  to  realise. 


JACOB   BOEHME,   THE   GERMAN 
PHILOSOPHER 

"  Gott  ist  von  der  Natur  frei,  und  die  Natur  ist  doch  seines 
Wesens." — J.  BOEHME  (Vom  dreifachen  Leben  des  Menschens). 

I 

IT  is  not  the  custom,  even  in  Germany,  to  assign  a 
place  of  importance  in  the  history  of  philosophy  to 
Jacob  Boehme,  the  shoemaker  theosophist  of  the  Renais- 
sance. Along  with  Hegel,  he  is  recognised  as  a  man  of 
powerful  mind  ;  but  whilst  it  is  admitted  that  from  the 
whole  of  his  obscure,  involved  writings  a  certain  number 
of  doctrines  capable  of  being  understood  to  some  extent 
by  the  intellect  can  be  evolved,  these  doctrines  are 
regarded  as  coming  under  the  category  of  theology  and 
Christian  edification  rather  than  as  monuments  of  profane 
and  rational  science.  Such  appreciation  is  natural  in 
France  where  philosophy,  in  the  spirit  of  Descartes, 
mostly  depends  on  the  understanding,  and  is  suspicious 
of  anything  resembling  mysticism.  In  Germany,  how- 
ever, philosophy  has  not  adopted  the  rationalistic  form 
in  so  constant  a  fashion.  Alongside  of  Leibnitz,  Kant, 
Fichte,  and  Hegel,  the  Schoolmen,  so  to  speak,  of 
modern  Germany,  we  find  philosophers  of  belief,  religion, 
or  feeling,  such  as  Hamann,  Herder,  Jacobi,  Schelling 
the  theosophist,  and  the  famous  Christian  philosopher 
Franz  von  Baader.  These  latter,  as  against  the  former, 

169 


170  STUDIES  IN  PHILOSOPHY 

are  mystical  dissidents,  just  as,  in  former  times,  Eckhart 
and  Tauler  were  opposed  to  Thomist  rationalism.  Even 
the  German  philosophers  of  concept  and  reflection,  the 
followers  of  Kant  and  Hegel,  if  we  consider  the  basis 
and  spirit  of  their  teaching,  and  not  the  form  in  which 
they  set  it  forth,  are  not  so  free  from  mysticism  and 
theosophy  as  would  seem  to  be  the  case,  or  even  as  they 
state.  For  they  too  look  upon  the  veritable  absolute  as 
being  not  in  space  or  thought,  but  rather  in  spirit,  which 
is  regarded  as  superior  to  the  categories  of  the  under- 
standing ;  they  too  endeavour  to  base  nature  on 
this  absolute.  Now,  taking  into  consideration  this 
element  of  mysticism  and  theosophy,  set  forth  in 
Germany  not  merely  by  a  whole  series  of  important 
philosophical  systems,  but  even  by  the  preeminently 
classical  systems,  if  we  inquire  into  the  origins  of 
German  philosophy,  we  can  hardly  fail  to  bestow  con- 
siderable attention  upon  the  shoemaker  theosophist.  We 
will  seriously  ask  ourselves  whether  he  did  not  deserve 
the  name  of  German  philosopher  given  him  during  his 
lifetime  by  his  admirer  and  friend,  Dr.  Walther. 

True,  at  first  sight,  the  name  scarcely  seems  to  suit 
him.  Boehme  is  neither  a  scientist,  a  dialectician,  nor 
even  a  disinterested  investigator.  The  son  of  peasants, 
his  first  occupation  was  that  of  a  cowherd.  Then  he 
became  a  shoemaker  at  GOrlitz,  the  town  adjoining  his 
birthplace,  and  here  he  conscientiously  practised  his  trade 
in  the  fear  of  the  Lord.  He  married  the  daughter  of 
a  worthy  butcher  living  in  the  town,  Catharina  Kuntz- 
schmann,  by  whom  he  had  four  sons,  and,  it  is  said, 
two  daughters.  He  brought  up  his  sons  in  his  own 
station  of  life  and  made  workmen  of  them.  He  lived 
in  piety,  simplicity,  and  Christian  meekness,  and  was 
ever  engaged  in  meditation  on  religious  things.  But 


JACOB  BOEHME  171 

it  was  his  continual  desire,  he  tells  us,  to  seek  in  the 
heart  of  God  for  a  refuge  from  divine  wrath  and  the 
wickedness  of  the  devil.  He  wrote  a  considerable 
number  of  books.  But  what  was  the  source  of  his 
inspiration  ?  He  had  read  neither  the  classic  authors 
nor  the  Schoolmen,  and  was  acquainted  only  with 
mystical  and  theosophical  writings.  And  even  for 
what  he  knows  he  is  indebted  to  personal  and  super- 
natural revelations.  Four  times  the  heavenly  light  was 
revealed  to  him,  when  he  saw  either  Christ  or  the 
eternal  Virgin  ;  during  the  few  moments  these  visions 
lasted  he  learned  more  than  he  would  have  done  had  he 
attended  classes  for  years.  At  the  beginning  of  each  of 
his  books  we  find  the  words  geschrieben  nach  g'ottlicher 
Erleuchtung,  written  by  divine  enlightenment. 

The  work  corresponds  with  the  conditions  under 
which  it  was  composed.  It  is  a  mixture  of  abstruse 
theology,  alchemy,  speculations  on  the  undiscernible, 
and  the  incomprehensible,  fantastic  poetry  and  mystic 
effusions  ;  in  fact,  a  dazzling  chaos.  His  first  book  is 
entitled,  The  dawn  at  its  rise,  or  the  root  and  mother  of 
-philosophy r,  astrology,  and  theology  considered  in  their  true 
principle :  a  description  of  nature,  in  which  is  seen  how  all 
things  were  in  the  beginning,  etc.  Boehme  herein  sets 
forth  the  genesis  of  the  holy  Trinity,  the  creation  and 
fall  of  the  Angels,  the  creation  and  fall  of  man,  the 
redemption  and  the  end  of  the  world.  He  sees,  and 
would  have  others  see,  far  more  than  he  demonstrates  ; 
his  science  is  a  metaphysical  hallucination.  Accord- 
ingly he  is  constantly  doing  violence  to  language,  requir- 
ing it  to  express  the  inexpressible.  He  uses  the  terms 
of  ancient  mysticism,  of  alchemy  and  philosophy  ;  he 
imposes  on  them  meanings  of  extraordinary  subtilty, 
and  insists  on  there  being  the  infinite  and  the  mysteri- 


172  STUDIES  IN  PHILOSOPHY 

ous  at  the  base  of  all  thought.  Is  it  possible  that  from 
such  a  work  anything  can  be  gleaned  by  the  historian 
of  philosophy,  unless  by  an  arbitrary  interpretation  he 
transforms  into  concepts  what,  on  the  part  of  the  author, 
is  pure  intuition  and  imagination  ? 

In  forming  an  opinion  of  this  man,  whose  sole  aim 
was  to  set  the  spirit  free  from  the  letter,  it  would  be 
unbecoming  to  judge  by  appearances.  In  reality,  Boehme 
is  not  the  simple,  ignorant  man  he  tells  us  he  is.  He 
was  open-minded  and  possessed  of  a  keen  intellect,  as 
his  first  teachers  immediately  recognised.  He  lived  in 
a  country  and  at  an  epoch  in  which  the  greatest  of  all 
problems  were  being  discussed.  The  mysticism  of  old 
was  still  flourishing  in  Germany  during  the  times  of 
Schwenckfeld  and  Sebastien  Franck.  At  the  same  time, 
ever  since  Nicolas  de  Cusa,  there  had  been  developing, 
beneath  the  influence  of  Italian  naturalism,  a  profound 
and  brilliant  theosophy  represented  by  Agrippa  von 
Nettesheim  and  Paracelsus,  the  rehabilitation  and  deifica- 
tion of  that  nature  which  the  mystics  of  the  Middle 
Ages  were  destroying.  In  another  direction,  over  against 
the  moral  optimism  of  Eckhart  and  his  disciples,  Luther 
had  recently  set  up  the  doctrine  of  a  positive,  radical 
evil,  rising  up  to  oppose  God  and  incapable  of  being 
brought  within  the  compass  of  mere  diminution  or 
deprivation.  The  new  principles  had  early  entered 
either  into  connection  or  into  conflict  with  the  principle 
of  ancient  mysticism.  Protestantism  was  already  attempt- 
ing that  reconciliation  of  its  mystical  with  its  Pauline 
origins,  its  spiritualistic  monism  with  its  moral  dualism, 
and  its  principle  of  liberty  with  that  of  discipline,  which 
she  is  still  following.  Theosophy  was  united  with 
mysticism  in  Valentin  Weigel,  who  submitted  as  matter 
for  the  subjective  reflection  of  Eckhart,  the  man  of 


JACOB  BOEHME  173 

Paracelsus,  a  resume  and  perfection  of  the  three  natures, 
the  terrestrial,  sidereal,  and  the  divine,  of  which  the 
created  universe  consists. 

From  his  youth  onwards,  Boehme  eagerly  took  an 
active  part  in  this  movement  of  ideas.  In  his  wander- 
ings to  and  fro  as  a  journeyman  before  becoming  a 
master-shoemaker,  he  conversed  of  things  religious  and 
theosophical ;  he  observed,  read,  and  reflected.  Though 
he  read  but  little,  what  he  did  read  was  important  and 
full  of  profound  thought.  The  Bible  was  for  him  the 
book  of  books,  that  thrilling,  deep  word  which,  especi- 
ally since  the  days  of  Luther,  has  ever  been  the  most 
powerful  incentive  to  reflection.  But  Boehme  read  the 
writings  of  many  other  masters  besides.  He  read 
Schwenckfeld,  noting  his  objections  to  that  doctrine 
of  vicarious  atonement  which  tends  to  replace  by  ex- 
ternal and  accidental  action  the  internal  working  of 
grace,  the  only  possible  source  of  essential  conversion. 
He  read  Paracelsus,  and  was  delighted  to  find  in  him 
an  enthusiastic  apostle  of  life,  a  revealer  of  the  magic 
power  of  imagination,  a  seer  who  finds,  in  the  world 
and  in  natural  man,  that  image  of  God  which  mystics 
had  ceased  to  find  therein.  He  studied  alchemy,  trying 
to  discover  its  true,  its  spiritual  meaning.  To  him, 
transmutation  was  the  symbol  of  the  new  birth  to  which 
man  is  called  ;  the  philosopher's  stone  found  its  realisa- 
tion for  him  in  the  power  of  faith  and  of  surrender  to 
God.  He  read  Valentin  Weigel,  and  became  imbued 
with  the  spiritual  mysticism  this  pious  pastor  inherited 
from  Tauler,  from  German  theology,  from  Schwenckfeld, 
and  from  Sebastien  Franck  ;  thanks  to  him,  also,  he  con- 
ceived the  idea  of  combining  mysticism  with  theosophy. 

Boehme  read  not  only  books  of  written  characters, 
he  also  read  the  book  of  nature.     Every  manifestation 


174  STUDIES  IN  PHILOSOPHY 

of  nature  is  instruction  for  him  ;  matter  is  not  a  being 
apart,  foreign  to  spirit  ;  it  is  spirit  itself,  revealed  and 
visible.  The  stars,  the  sun,  the  elements  of  the  earth,  life 
everywhere,  in  its  origin  and  in  every  one  of  its  phases, 
the  growing  tree,  the  animal  with  its  desires  and  dis- 
interested instincts,  man  with  his  inner  life,  his  struggle 
with  evil,  his  defeats  and  triumphs — all  these  things 
Boehme  contemplates  and  meditates  upon,  and  in  this 
immediate  and  religious  communion  with  nature,  waits 
for  her  to  infuse  into  him  her  own  spirit  and  reveal  the 
mysteries  of  being. 

It  is  eternal,  interior,  and  living  being  that  he  seeks 
everywhere  and  in  all  things.  Thus,  the  phenomena  of 
nature,  like  the  teachings  set  forth  in  books,  are 
signs  for  him  to  decipher,  not  the  object  about  which 
knowledge  is  sought.  The  reason  why  he  reads  and 
observes  is  to  have  matter  on  which  his  spirit  may 
dwell  for  reflection.  It  is  Boehme's  endeavour  to  set 
the  spirit  free  from  the  letter,  to  find  out  the  force 
which  works  at  the  heart  of  inert  phenomena,  and  to 
penetrate  to  the  very  source  of  all  reality.  Therefore 
inner  experience  and  reflection  are,  once  for  all,  his  true 
means  of  investigation.  True,  he  was  an  illuminate  ; 
his  meditation  was  a  prayer  ;  his  discoveries,  divine 
revelations.  Still,  what  matters  the  explanation  the 
individual  himself  gives  of  the  channel  along  which 
his  ideas  entered  his  consciousness  ?  Is  Descartes' 
analytical  geometry  any  the  less  true  because  he  claimed 
that  he  owed  its  invention  to  the  assistance  of  the  holy 
Virgin  ?  It  may  be  because  of  the  way  in  which  the 
human  mind  is  constituted  that  he  at  first  attributed  to 
supernatural  revelation  the  new  ideas  that  arose  within 
him,  impressing  him  by  their  beauty  and  illuminat- 
ing power,  and  that  he  regarded  them  as  entering  his 


JACOB  BOEHME  175 

mind  from  without.  Plato's  essences,  the  z/oOs  of  Aris- 
totle, the  Christian  ideal,  the  supreme  principles  of 
knowledge  and  action,  were  looked  upon  as  beings  and 
things  in  themselves,  before  they  came  to  be  explained 
by  the  laws  of  the  human  mind.  The  natural  has  first 
been  supernatural ;  for  the  genius  does  not  know  how 
he  arises  ;  to  himself  he  appears  as  a  god  visiting  his 
creature.  Boehme,  indeed,  is  not  content  to  receive 
into  his  own  intelligence  the  revelations  of  divine 
intelligence  ;  he  is  a  seer  of  visions.  Increate  wisdom, 
the  eternal  Virgin,  appeared  to  him  several  times. 
But  enthusiasm,  even  when  of  a  somewhat  sickly 
nature,  is  just  as  likely  to  strengthen  as  it  is  to 
weaken  the  powers  of  the  human  mind,  and  a  shock  to 
the  organism  is  nothing  but  the  result  of  the  excessive 
tension  to  which  the  mind  has  had  to  subject  the  body 
for  the  realisation  of  its  creations.  The  thinking  reed1 
bends  beneath  the  effort  of  thought,  even  more  than 
beneath  the  weight  of  matter.  After  all,  there  is  only 
one  interpretation,  only  one  standard  of  either  a  thinker's 
or  an  artist's  work,  and  that  is  the  work  itself.  The 
author  is  the  mould  which  is  broken  that  the  statue 
may  be  made  visible. 

II 

What  is  it,  then,  that  we  find  in  the  work  of  Boehme 
when  considered  in  itself,  both  in  its  spirit  and  inner  mean- 
ing, as  the  author  would  have  it  studied,  and  in  its  real 
and  objective  content,  as  history  would  have  it  studied  ? 

First  of  all,  what  is  the  motive  of  the  theosophist 
shoemaker's  reflections  ? 

1  Pascal  in  his  Pens&s  (Edition  Ha<vet,  i.  6)  says  :  "  L'homme  n'est  qu'un 
roseau  le  plus  faible  de  la  nature,  mais  c'est  un  roseau pensant."  (Translator's 
note.) 


1 76  STUDIES  IN  PHILOSOPHY 

"  From  my  youth  up,"  he  tells  us,  "  I  have  sought 
only  one  thing  :  the  salvation  of  my  soul,  the  means  of 
gaining  possession  of  the  kingdom  of  God."  Here, 
apparently,  is  nothing  more  than  an  altogether  practical 
and  religious  object ;  but  in  Boehme's  mind,  this  object 
is  destined  to  raise  the  most  profound,  metaphysical 
speculations. 

He  learnt  from  the  mystics  what  it  means  to  possess 
God.  One  must  take  care,  so  these  masters  teach,  not 
to  liken  the  possession  of  God  to  the  possession  of  any- 
thing material.  God  is  spirit,  i.e.  for  the  man  who 
understands  the  meaning  of  the  term,  a  generating 
power  previous  to  all  essence,  even  the  divine.  God 
is  spirit,  i.e.  pure  will,  both  infinite  and  free,  with 
the  realisation  of  its  own  personality  as  its  object. 
Henceforward,  God  cannot  be  accepted  by  any  passive 
operation.  We  possess  him  only  if  he  is  created 
within  us.  To  possess  God  is  to  live  the  life  of 
God. 

On  the  other  hand,  Boehme  learnt  from  Luther 
that  the  natural  man  is  not  simply  a  son  separated 
from  his  father,  that  between  God  and  his  creature 
there  is  something  more  than  inert  space,  unresisting 
non-being.  The  natural  man  has  rebelled  against  his 
creator  :  between  him  and  God,  sin  raises  its  head, 
like  a  real,  positive  power,  endeavouring  to  defeat  the 
divine  action.  Evil  is  not  non-being,  it  is  a  real  being 
that  combats  the  principle  of  good.  Everywhere  in 
nature  Boehme  finds  that  effective  warfare  being  waged, 
which  Luther  enabled  him  to  see  in  the  human  con- 
science. Whether  he  beholds  sun  and  stars,  clouds  or 
rain  or  snow,  creatures  with  reason  or  creatures  without 
reason,  such  as  wood,  stones,  earth,  or  elements ; 
no  matter  in  which  direction  he  turns,  he  sees  every- 


JACOB  BOEHME  177 

where  evil  over  against  good,  anger  opposing  love, 
affirmation  opposed  to  negation.  Even  justice,  here 
below,  is  at  grips  with  its  contrary.  For  the  godless 
are  as  prosperous  as  the  god-fearing,  barbarous  nations 
possess  the  richest  lands  and  enjoy  the  good  things 
of  earth  more  than  do  the  servants  of  God.  Observing 
these  things,  Boehme  tells  us,  I  fell  into  a  state  of  deep 
melancholy  and  my  spirit  was  troubled.  Not  a  single 
book,  of  all  those  with  which  I  was  acquainted,  brought 
me  any  consolation.  And  the  devil  was  there,  watching 
for  me,  and  filling  my  mind  with  heathenish  thoughts 
such  as  I  should  be  ashamed  to  express  here.  Is  it  true 
that  God  is  love,  as  Christianity  teaches,  that  God  is 
omnipotent,  that  there  is  nothing  which  has  reality 
in  his  presence  ?  Such,  doubtless,  are  the  questions 
Boehme  felt  starting  to  life,  deep  in  his  consciousness. 
Gladly  would  the  devil  have  seen  him  give  up  all 
hope  of  fathoming  the  mystery  and  sink  to  sleep  in 
indifference.  Boehme,  however,  guessed  his  designs 
and  determined  to  foil  them. 

How  was  he  to  reconcile  the  end  of  human  activity, 
of  which  mystics  had  so  noble  a  conception,  with  the 
reality  of  things,  so  concisely  stated  by  the  founder 
of  Protestantism  ?  If  mankind  and  the  whole  of  nature 
have  radically  rebelled  against  God,  how  can  one  main- 
tain the  possibility  of  the  birth  of  God  within  the 
human  soul  ?  If  man,  like  a  decayed  tree,  can  will 
and  do  nothing  but  evil,1  there  is  no  middle  course 
to  adopt,  it  would  appear,  between  leaving  the  tree  to 
rot,  and,  after  uprooting  it,  flinging  it  into  the  fire. 
If  nature  is  absolutely  opposed  to  God,  either  God 
has  no  power  over  her,  or  he  ought  to  destroy  her. 

To  maintain  the  spiritual  and  optimistic  ideal  of  the 

1  According  to  Luther's  expression. 

N 


178  STUDIES  IN  PHILOSOPHY 

mystics,  whilst  at  the  same  time  regarding  nature  from 
the  pessimistic  standpoint  of  Luther,  and,  in  a  more 
general  way,  from  a  realistic  standpoint  :  such  is  the 
task  Boehme  sets  himself.  This  task  determines  itself 
in  his  mind  as  follows.  Whereas  the  mystics  wished  to 
know  how  God  can  be  born  in  that  which  is  not  himself, 
Boehme  asks  himself  how  God  can  be  reborn  in  that  which 
has  violently  separated  from  him.  Now,  he  imagines 
he  can  solve  this  problem  if  he  is  able  to  discover  both 
the  source  of  divine  existence  and  the  origin  of  the 
world  and  of  sin.  This  science  will  be  regeneration 
itself.  For  knowledge,  when  it  penetrates  to  its  source 
and  origin,  blends  and  unites  with  action  and  reality. 
To  see  things  from  the  standpoint  of  God  is  to  be 
reborn  to  divine  life. 

The  following,  therefore,  is  to  be  the  fundamental 
division  of  Boehme's  system :  ist,  How  does  God 
engender  himself?  2nd,  Why  and  how  did  God 
create  the  world  and  how  did  evil  enter  therein  ? 
3rd,  How  can  God  be  reborn  in  the  heart  of  the 
corrupt  creature,  and  what  is  the  final  end  of  all 
beings  ? 

As  we  see,  this  is  the  question  of  the  beginning  and 
the  end,  stated  in  all  its  generality  and  dominating  all 
others.  Whereas  the  ancients  tried  to  discover  a 
posteriori  what  stable,  determinate  principles  lie  hidden 
beneath  the  movement  and  indetermination  of  pheno- 
mena, and  knew  no  mean  between  an  altogether 
illusory,  indeterminate  absolute  such  as  chance,  and  a 
full  and  perfected  absolute  such  as  intelligence,  our 
philosopher,  for  whom  the  whole  of  nature  is  the  result 
of  an  action,  tries  to  find  out  how  the  absolute  itself 
came  into  being,  in  so  far  as  it  is  this  and  not  that ; 
even  as  regards  God,  he  descends  from  infinite  power 


JACOB  BOEHME  179 

to  the  production  of  determinate  being.  The  philosophy 
of  the  ancients  was  a  classification,  more  than  anything 
else,  that  of  Boehme  is  to  be  a  construction.  The 
problem  of  the  genesis  of  things  has  been  substituted  for 
that  of  their  essence.  And  as  the  being  whose  genesis  is 
here  sought  and  whose  internal  movement  should  explain 
nature  is  distinctly  the  conscious,  free  and  acting  person, 
the  system  we  are  about  to  study  appears  before  us  as 
the  dawn  of  a  new  philosophy,  which  may  be  called  the 
philosophy  of  personality,  considered  in  itself  and  in  its 
connection  with  nature. 

What  method  does  Boehme  recommend  in  this 
enquiry  ? 

The  problem  now  before  us,  we  must  remember,  is 
to  see  being  proceed  from  its  primary  source,  that  is, 
to  apprehend  the  transition  from  nothing  to  something. 
Now,  the  means  at  the  disposal  of  ordinary  philosophy 
are  powerless  for  such  a  task.  What  will  erudition 
give  us,  except  opinions,  abstract  ideas  ?  The  Bible 
itself,  if  we  seek  enlightenment  therein  without  going 
farther  back  in  time,  is  nothing  but  a  dead  letter,  a 
symbol  that  cannot  be  explained.  It  is  the  same  with 
the  senses  and  the  reason  as  it  is  with  erudition.  The 
senses  enable  us  to  know  only  the  cut-and-dry  appear- 
ances of  things  and  their  products,  not  their  real  nature 
and  inner  life.  Exterior  reason,  or  the  natural  elaboration 
of  the  data  of  experience,  is  as  dead  as  the  materials  it 
brings  together.  It  analyses  and  separates  ;  and  the 
objects  it  considers,  thus  snatched  from  the  living 
whole  of  which  they  formed  part,  are  no  more  than 
fictitious  beings,  incapable  of  telling  us  anything  of 
their  origin  and  true  nature.  It  is  this  exterior  reason, 
which,  seeing  the  wicked  in  this  world  of  ours  prosper 
equally  with  the  good,  insinuates  to  man  that  evil  is  the 


180  STUDIES  IN  PHILOSOPHY 

equal  of  good,  and  consequently  that  the  existence  of 
the  God  of  religion  is  problematical. 

All  these  methods  have  the  same  flaw  :  they  are 
passive  and  dead.  They  presuppose  a  given,  realised 
object,  and  set  the  mind,  like  a  motionless  mirror, 
opposite  that  object.  A  living  method,  alone,  enables 
us  to  penetrate  into  the  mysteries  of  life.  Being,  alone, 
knows  being  ;  we  must  generate  with  God  in  order  to 
understand  generation.  Therefore  the  true  method 
consists  in  witnessing,  or  rather  taking  part  in  the 
divine  operation  whose  end  is  the  blossoming  and 
dominion  the  rule  of  the  personality  ;  it  is  knowledge  as 
consciousness  of  action:  a  method,  indeed,  which  proceeds 
from  cause  to  effect,  whereas  any  purely  logical  method, 
limited  to  the  working  out  of  the  data  of  experience, 
is  and  can  be  nothing  more  than  a  vain  effort  to  rise 
from  effect  to  cause. 

But  then,  how  can  man  thus  place  himself  at  the 
standpoint  of  God  ?  It  is  impossible  for  him  to  ascend 
to  God  :  there  is  no  transmutation  of  creature  into 
creator.  Still,  though  man  cannot  ascend  into  God, 
God  can  descend  into  man.  Not  that  God  can  be 
evoked  and  materially  constrained,  as  it  were,  by  the 
practices  of  false  magic  or  outward  devotion,  but  rather 
that  God  descends  into  man,  when  man  dies  to  his 
corrupt,  inborn  nature,  to  give  himself  up  to  divine 
action.  Christ  said  that  you  "  must  be  born  again,"  if 
you  would  see  uthe  kingdom  of  God."  The  conversion 
of  the  heart  opens  the  eye  of  the  intelligence.  Just 
as  the  exterior  man  sees  the  exterior  world,  so  the  new 
man  sees  the  divine  world  in  which  he  is  living.  And 
this  return  to  God  is  possible  for  man,  since  man  was 
created  in  the  image  of  God.  He  has  only  to  go  down 
to  the  deepest  recesses  of  himself  and  set  free  the  interior 


JACOB  BOEHME  181 

man  from  the  exterior  man  in  order  to  participate  in 
divine  life.  "  Reflect  on  thyself,  search  thyself,  find 
thyself:  this  is  the  key  of  wisdom.  Thou  art  the 
image  and  child  of  God.  Such  is  the  development  of 
thy  being  ;  eternal  birth  in  God.  For  God  is  spirit, 
and  likewise  in  thee  that  which  commands  is  spirit  and 
is  the  creation  of  divine  sovereignty." 

Once  man  thus  adopts  the  eternal  standpoint  of 
universal  genesis,  everything  which  at  the  outset  was 
only  veil  and  mist  interposed  between  himself  and  the 
light,  becomes  a  transparent  symbol,  a  faithful  expres- 
sion. Erudition,  the  Bible,  tradition,  concepts,  the 
phenomena  of  nature,  all  these  things,  though  dead  in 
themselves,  become  animated  and  living  when  regarded 
with  the  eye  of  the  spirit.  The  eternal  word,  speaking 
within  ourselves,  tells  us  the  true  meaning  of  the  written, 
the  sensible  word.  Nor  is  this  all,  for  between  the 
within  and  the  without  there  is  reciprocity  of  action. 
Of  a  surety,  the  sight  of  exterior  things,  in  itself  alone, 
would  never  have  revealed  to  us  the  principle  which 
these  things  manifest,  this  principle  wills  to  be  under- 
stood in  itself.  Primary  being,  however,  is  to  us 
nothing  but  empty  form  ;  it  is  by  the  correct  inter- 
pretation of  phenomena  that  it  assumes  body  and  is 
determined.  All  the  same,  it  could  never  find  adequate 
expression  in  phenomena.  Being  infinite,  spirit  could 
not  be  wholly  manifested,  for  all  manifestation  takes 
place  by  means  of  the  finite.  Spirit  is  eternal  mystery 
in  its  essence.  Therefore  not  only  should  we  make 
use  of  phenomena  in  order  to  catch  a  faint  glimpse  of 
the  details  of  divine  perfection,  but  we  should  also 
remember  that  phenomena  are  never  anything  else  than 
an  imperfect  manifestation  of  this  perfection.  And 
when  we  speak  of  the  origin  of  God  and  of  things,  we 


1 82  STUDIES  IN  PHILOSOPHY 

ought  to  appeal  to  all  the  images  with  which  our  senses 
and  reason  supply  us,  and  always  look  upon  these  images 
as  but  clumsy  metaphors  which  should  be  understood  in 
spirit  and  in  truth.  The  wisdom  of  God  is  beyond  all 
description. 

Ill 

This  maxim  meets  with  its  application  at  the  very 
first  step  theosophy  attempts  to  take.  To  begin  with, 
we  have  to  set  forth  the  birth  of  God,  the  way  in  which 
God  generates  himself.  To  speak  of  the  birth  of  God, 
however,  taking  these  words  literally,  is  to  speak  the 
devil's  language  ;  it  is  saying  that  eternal  light  flashed 
out  of  darkness,  that  God  had  a  beginning.  Still,  I  am 
compelled  to  employ  this  term  :  the  birth  of  God  ; 
otherwise,  thou  couldst  not  understand  me.  Restricted 
as  we  are,  we  speak  only  by  parcelling  things  out,  by 
breaking  the  unity  of  the  whole.  In  God  there  is 
neither  Alpha  nor  Omega,  neither  birth  nor  develop- 
ment. I,  however,  am  compelled  to  place  things  one 
after  the  other.  The  reader  must  by  no  means  read 
me  with  the  eyes  of  flesh. 

Eternal  nature  generates  itself  without  any  begin- 
ning. How  does  this  generation  come  about  ? 

Boehme  here  sets  himself  the  famous  problem  of 
self-originated  existence — of  aseity.  Whereas,  however, 
by  this  term  the  Schoolmen  understand  a  mere  property 
of  perfect  being,  a  property,  too,  that  is,  above  all, 
negative  ;  Boehme  insists  that  the  strange  expression, 
"  God  self-caused,"  shall  have  a  precise,  concrete  and 
positive  meaning.  To  fathom  the  mystery  it  contains 
is,  to  him,  the  first  and  main  question,  the  solution  of 
which  will  throw  light  on  all  other  questions.  Nor  does 
he  think  he  ought  to  abandon  the  search  until  he  has 


JACOB  BOEHME  183 

reconstructed  in  thought  the  logical  sequence  of  the 
operations  by  which  God  rises  from  a  state  of  nothing- 
ness to  one  of  fulness  of  existence. 

What,  then,  was  there  in  the  beginning  ?  From 
what  germ  did  God  generate  himself? 

In  the  beginning  was  being  which  presupposes 
nothing  anterior  to  itself;  in  which,  consequently, 
nothing  is  essence,  or  nature,  or  finite,  determinate 
form  :  for  everything  that  exists  as  a  determinate  thing 
demands  a  cause  and  a  reason.  We,  for  our  part,  can 
conceive  of  this  being  only  as  the  eternal  no-thing,  the 
infinite,  the  abyss,  the  mystery.  Boehme  uses  the  word 
Ungrund  to  designate  this  first  source  of  things,  mean- 
ing thereby  that,  beneath  God,  there  is  nothing  to  serve 
him  as  a  foundation,  and  also  that  in  the  first  being 
the  ground  or  reason  of  things  is  not  yet  manifested. 
Thus,  the  primordial  infinite  in  itself  is  nothing  but 
silence,  rest  without  beginning  or  end,  absolute  peace 
and  eternity,  unity  and  identity.  In  it  is  neither  goal, 
nor  place,  nor  even  the  impulse  to  seek  and  find.  It 
is  free  from  suffering,  that  companion  of  desire  and 
quality.  It  is  neither  light  nor  darkness.  It  is  an  un- 
fathomable mystery  unto  itself. 

Such  is  the  initial  condition  of  divinity.  Is  it  also 
its  fulfilment?  If  the  answer  is  in  the  affirmative, 
God  is  reduced  to  being  nothing  more  than  an  abstract 
property,  wanting  in  force,  intelligence  and  science  ;  he 
is  rendered  incapable  of  creating  the  world  in  which 
the  very  perfections  he  lacks  are  to  be  found.  But  it 
is  impossible  that  God  should  be  an  inert  being,  dwell- 
ing somewhere  beyond  the  skies.  The  Father  is  omni- 
potent and  omniscient  ;  he  is  the  essence  of  gentleness 
and  love,  pity  and  blessing.  The  world,  too,  derives 
from  him  all  the  perfections  to  be  found  therein.  Then 


1 84  STUDIES  IN  PHILOSOPHY 

how  is  the  transition  to  be  brought  about  from  God, 
who  is  nothingness,  to  God  the  person  and  creator  ? 

Here  we  come  to  the  main  point  in  Boehme's 
system.  The  solution  of  the  problem  of  eternal 
generation,  given  by  our  theosophist,  is  the  distinctive 
task  he  set  himself;  it  opens  up  a  new  path  along 
which  many  philosophers  were  subsequently  to  proceed. 

Of  course,  the  mystics  of  old  had  already  taken  up 
this  line  of  research.  Eckhart  asked  himself  how 
merely  potential,  motionless  and  inactive  divinity,  which 
is  the  first  being,  becomes  the  living  and  personal  God, 
who  alone  is  true  God.  He  explained  the  transition 
from  the  one  to  the  other,  by  considering  the  part 
played  by  the  image  or  idea  of  God,  which  emanates 
spontaneously  from  primordial  power,  just  as  from 
each  of  our  tendencies  there  goes  forth  an  idea  that 
makes  it  objective  and  manifest.  Beholding  itself  in 
its  own  image,  absolute  substance  became  conscious  of 
itself  and  was  constituted  a  person. 

Boehme  is  inspired  by  this  doctrine,  but  he  does 
more  than  return  to  and  continue  it ;  with  that  sense 
of  concrete  existence,  of  life  and  nature,  which  char- 
acterises him,  he  can  find  no  satisfaction  in  the  abstract 
God  of  the  mystics  of  old.  Eckhart  had  almost  ex- 
plained how  God  becomes  conscious  of  himself ;  con- 
sciousness of  self,  however,  is  no  more  than  the  shadow 
of  existence.  In  order  then  that  God  may  really  be 
a  person  and  that  nature  may  find  in  him  the  elements 
of  a  positive  existence,  divine  generation  must  be  some- 
thing different  from  what  Eckhart  teaches. 

Boehme  starts  with  the  principle  that  God,  who  is 
mystery,  wills  to  reveal  himself  in  all  the  fulness  of  his 
being,  i.e.  to  manifest  himself  as  a  living  person,  capable 
of  creating.  In  so  far  as  he  pursues  the  revelation  of 


JACOB  BOEHME  185 

himself,  God  wills  and  posits  all  the  conditions  of  this 
revelation.  Now,  according  to  Boehme,  there  is  one 
supreme  law  which  governs  all  things,  both  divine  and 
human  :  that  all  revelation  calls  for  opposition.  As 
light  is  visible  only  when  reflected  by  a  dark  body, 
so  anything  whatsoever  is  posited  or  constituted  only 
by  being  set  over  against  its  opposite.  That  which 
meets  with  no  obstacle  always  goes  forward  and  never 
returns  within  itself,  never  manifestly  exists,  either  for 
itself  or  for  another.  Two  moments  may  be  distin- 
guished in  the  relation  of  the  given  principle  to  its 
contrary.  The  mere  presence  of  the  negative  principle 
over  against  the  positive  principle  manifests  the  latter 
only  as  a  potency  or  a  possibility.  If  it  is  desired  that 
this  potency  become  reality,  it  must  act  upon  the 
negative  principle,  discipline  it  and  make  thereof  its 
instrument  and  expression.  This  law  of  opposition 
and  reconciliation  governs  divine  genesis.  If  the  divine 
spirit  is  to  be  revealed,  it  will  not  remain  within  itself, 
it  will  create  its  contrary.  Nor  is  this  all ;  for,  acting 
on  this  contrary,  it  will  assimilate  it  to  itself  and 
spiritualise  it.  And  so  we  find  that  Boehme  is  to 
involve  God  in  a  series  of  oppositions.  In  proportion 
as  contradictions  and  reconciliations  come  about,  in  like 
proportion  will  divine  personality  be  realised.  The 
contrary  essence  or  nature  on  which  God  will  rely  in 
order  to  personify  himself,  will  constitute,  within  God 
himself,  the  eternal  basis  of  our  created  nature. 

Such  are  the  ideas  that  govern  Boehme's  system 
and  give  it  its  distinctive  character.  They  have  their 
centre  in  a  principle  which  may  be  formulated  as 
follows  :  being  is  constituted  as  potency  by  opposing 
itself  and  as  reality  by  reconciling  to  itself  that  which 
is  opposed  to  it.  These  general  ideas,  however,  are  not 


1 86  STUDIES  IN  PHILOSOPHY 

so  much  formulated  in  one  special  place,  as  employed 
in  the  development  of  the  system. 

In  the  beginning  was  no-thing.  This  no-thing  is  not 
absolute  nothingness.  On  the  contrary,  it  is  being 
itself,  eternal  Good,  eternal  gentleness  and  eternal  love, 
but  still,  being  in  itself,  i.e.  non-manifested.  And  so 
in  this  no-thing  there  dwells  an  internal  opposition.  It 
is  nothing,  and  it  is  all  ;  it  is  indifference  and  it  is 
excellence.  That  is  the  reason  this  no-thing  must 
appear  to  us  as  unstable  and  living.  It  will  move 
itself,  in  order  to  become  reconciled  with  itself. 

The  first  result  of  the  opposition  just  noticed  is  the 
scission  of  the  primordial  infinite  into  two  contraries  : 
desire  (Sucht)  and  will  (Wille).  No-thing  is  desire, 
because  it  is  mystery,  and  mystery  tends  to  manifest 
itself;  no-thing  is  the  desire  to  become  some-thing. 
But  the  object  to  which  it  tends  is  not  an  indeterminate 
one  :  it  is  the  manifestation  and  possession  of  oneself. 
And  so  the  infinite  is  desire  on  the  one  hand,  and  what 
is  called  will  on  the  other.  Unconscious  and  un- 
assuaged  desire  generates  will ;  but  will,  to  which 
belong  knowledge  and  understanding,  regulates  and 
determines  desire.  The  one  possesses  motion  and  life  ; 
the  other,  independence  and  power  of  command.  Will 
is  greater  than  the  power  which  gave  it  birth. 
This  duality  is  the  origin  of  all  the  oppositions  which 
the  march  of  divine  revelation  will  arouse.  Will  is 
the  germ  of  divine  personality  and  the  basis  of  all 
personality  ;  desire,  the  essence  and  body  of  will,  is 
the  germ  of  eternal  nature  and  the  basis  of  sensible 
nature. 

And  so  will  is  manifested  because  of  the  presence  of 
desire,  with  which  it  is  contrasted.     Yes  and  no,  how- 


\ 

\ 


JACOB  BOEHME  187 

ever,  are  not  two  things  outside  of  each  other  ;  they 
are  one  and  the  same  thing,  divided  only  to  allow  the 
yes  to  reveal  itself.  That  is  why  separation,  in  its  turn, 
is  an  unstable  condition.  The  yes  which,  in  this 
separation,  is  per  se  devoid  of  essence  and  looked 
upon  as  no-thing,  endeavours  to  make  itself  concrete 
by  absorbing  the  no  and  reconstituting  unity  to  its  own 
advantage.  On  to  the  two  opposite  terms,  desire  and 
will,  there  is  now  added  a  third,  the  idea  of  a  recon- 
ciliation of  the  first  with  the  second.  The  production 
of  this  third  term  is  the  work  of  imagination.  Speak- 
ing generally,  this  faculty  of  imagination  is  desire, 
applied  to  an  image  and  tending  to  absorb  it — as 
hunger  absorbs  food — and  then  to  produce  it  in  the 
outer  world,  transformed  into  a  living  reality  by  the 
action  of  the  subject  itself.  Now,  the  will  which  is  mind, 
and  whose  object  is  the  revelation  of  itself,  unites  with 
desire,  in  order  to  imagine  this  revelation  ;  and,  in 
doing  so,  become  capable  of  realising  it.  Imagination 
makes  the  will  into  a  magician.  What  the  will  wills  is 
determined  in  the  very  effort  it  makes  to  represent  it 
to  itself.  It  wills  to  find  and  lay  hold  upon  itself ; 
consequently,  to  form  an  interior  mirror  of  itself ;  and 
as  desire  is  the  matter  on  which  it  works,  it  wills  that 
infinite  desire,  fixing  itself  on  the  Good,  shall  become 
this  mirror. 

The  task,  then,  before  God  or  the  will  is  the  follow- 
ing :  the  regulating  of  desire  according  to  the  law  of 
the  Good,  and  hence,  the  forming  of  an  object  which  is 
a  mirror  of  the  will,  and  wherein  the  latter  can  con- 
template and  recognise  itself.  In  accomplishing  this 
task,  divine  will  is  to  issue  from  a  state  of  nothingness 
and  attain  to  reality. 

God  wills  to  manifest  himself,  to  form  a  mirror  of 


1 88  STUDIES  IN  PHILOSOPHY 

himself.  He  can  do  this  only  in  a  threefold  manner. 
First,  he  must  posit  himself  as  indeterminate  will, 
capable  of  willing  good  or  evil.  Such  a  will,  however, 
is  neither  good  nor  bad  :  God  must  come  out  of  this 
indifference.  He  does  this  by  generating  within  him- 
self the  one,  eternal  Good,  or  determinate  will.  This 
good,  which  is  God,  is  not  an  object  or  a  thing  ;  it 
remains  will,  though  strong,  infallible  will.  With  the 
generation  of  this  will,  a  beginning  has  been  constituted 
in  the  infinite,  a  foundation  has  been  formed  in  the 
abyss,  and  a  reason  for  things  has  been  superposed  upon 
the  eternal  mystery.  Nevertheless,  the  first  will  has 
not  exhausted  itself  in  the  generation  of  determinate 
will  :  it  retains  its  infinite  fecundity.  Thus,  from  the 
conjuncture  of  infinite  will  and  determinate  will  springs 
a  third  will,  to  wit,  will  that  goes  forth  of  itself  to 
produce  an  object.  The  object  resulting  from  this 
threefold  action  is  none  other  than  the  mirror  of  will 
itself,  eternal  wisdom.  This  image  is  not  God,  it  is 
only  the  image  of  God.  Still,  by  it,  God  is  henceforth 
self-revealed,  he  sees  himself  as  a  will  that  is  threefold 
and  one  at  the  same  time.  These  three  moments  of 
divine  activity  may  be  characterised  by  the  names  of 
willy  strictly  so  called,  reason  and  force.  They  may 
also  be  named  Father,  Son  and  Spirit.  These  are  not 
three  gods,  for  each  of  the  three  is  a  spiritual  being, 
and  separation  of  substances  exists  only  in  the  material 
world.  Nor  are  they  even  three  persons  ;  for  will,  as 
against  its  image  or  idea,  is  only  knowledge  and 
consciousness  of  itself,  it  does  not  yet  exercise  that 
empire  over  a  thing-being,  which  is  the  condition  of 
personality.  In  truth,  God  is  person  only  in  Christ. 
In  the  generation  that  has  just  been  considered,  there  is 
nothing  else  than  a  threefold  action  of  the  one  will. 


JACOB  BOEHME  189 

Eternal  wisdom,  whose  production  is  the  result  of 
this  action,  in  which,  too,  the  Trinity  sees  and  finds  itself 
acting,  is  not  a  fourth  will,  but  is  set  over  against  the 
Trinity  as  its  representation  or  object.  It  is  this 
reconciliation  of  desire  with  will  that  the  latter  had 
undertaken  to  effect.  Like  every  mirror,  it  is  passive  and 
does  not  generate  at  all.  It  is  the  eternal  virgin.  In  it 
are  all  the  divine  perfections,  though  rather  as  ideas  and 
paradigms  than  forces  and  living  beings.  For  these 
perfections  are  objects  of  will,  not  wills  themselves  :  and 
life  could  not  exist  without  will,  on  which  it  is  founded. 
Life  and  fruitfulness  belong  not  to  ideas  or  generalities, 
but  to  persons  only,  in  so  far  as  they  act  in  accordance 
with  ideas. 

Such  is  the  divine  genesis  following  the  appearance 
of  desire  and  will  in  the  heart  of  the  primordial  infinite. 
Here,  indeed,  we  have  God  far  removed  from  a  state 
of  nothingness.  He  knows  himself  as  will,  and  even 
good  will.  But  is  he  God  the  Father,  omnipotent  and 
omniscient,  love  and  pity,  light  and  joy,  of  whom  we 
try  to  catch  a  glimpse  and  whom  we  seek  ? 

This  God,  if  we  note  well,  by  no  means  realises 
personality  yet.  He  is  intelligence  ;  he  knows  himself. 
But  intelligence,  such  as  we  see  it  within  ourselves,  is 
not  something  concrete,  something  we  can  grasp.  It  is 
not  an  essence,  but  rather  the  potency  or  germ  of  an 
essence.  The  God,  whose  action,  altogether  interior, 
has  no  other  object  than  himself,  is  still  a  hidden,  an 
incompletely  revealed  God.  He  is  God  as  far  as  possi- 
bility will  allow  :  the  divine  ideal.  In  order  that  this 
ideal  may  be  realised  and  God  be  the  living  person,  will 
must  continue  the  work  of  eternal  generation,  which,  so 
far,  has  only  been  begun.  God  must  have  a  second  birth. 

Here,  more  especially,  the  law  of  contraries  will  find 


1 9o  STUDIES  IN  PHILOSOPHY 

its  application.  If  we  consider'  all  the  things  that  really 
exist  in  this  world,  we  find  they  are  made  up  of  yes  and 
no  :  In  Ja  und  Nein  bestehen  alle  T)inge.  Day  could 
not  be  without  night,  nor  night  without  day  ;  cold  is 
the  condition  of  heat,  and  heat  of  cold.  Do  away  with 
opposition  and  struggle,  and  everything  will  return  to 
silence  and  immobility,  everything  will  revert  to  a  state 
of  nothingness.  The  one,  in  so  far  as  it  is  one,  has 
nothing  that  it  can  will.  For  it  to  will  and  live,  it  must 
divide  into  two.  In  the  same  way,  unity  cannot  sense 
itself,  but  in  duality  sensation  is  possible.  For  a  being, 
then,  to  be  posited  as  real,  it  must  be  opposed  against  its 
contrary,  and  the  degree  of  opposition  is  the  measure  of 
the  degree  of  realisation. 

Now,  in  the  development  of  the  divine  activity  just 
considered,  God  was  not  opposed  against  anything  which 
might  rightly  be  called  his  contrary.  The  power  of 
objectivation  in  whose  presence  he  has  found  himself  and 
which  he  determined  and  limited  so  as  to  form  his  true 
image  of  it,  differed  from  him  only  as  idea  differs  from 
intelligence.  In  this  passive  principle,  there  is  nothing 
to  oppose  divine  action  :  a  mirror  reflects — without 
resisting — the  rays  that  fall  upon  it.  In  this  altogether 
ideal  opposition,  God  could  acquire  only  an  ideal 
existence.  In  order  that  he  may  assume  bodily  form, 
as  a  person,  he  must  be  engaged  in  strife  with  a  real 
contrary,  i.e.  with  a  positive  power  whose  action  is 
opposed  to  his  own.  Therefore  God  must  raise  up 
such  a  contrary,  become  connected  with  it,  oppose  it 
and  finally  discipline  and  permeate  it ;  only  thus  will 
the  work  of  divine  generation  be  accomplished.  How 
is  this  new  development  to  be  effected  ? 

The  will  that  has  realised  itself  in  the  evolution 
through  which  we  have  just  passed,  and  which  may  be 


JACOB  BOEHME  191 

called  reason,  is  still  pure  spirit,  infinitude,  a  mystery. 
But  mystery,  whilst  it  continues  such,  calls  for  revela- 
tion which  alone  determines  it  as  mystery.     Like   all 
contraries,   mystery  and   revelation   imply   each   other. 
Therefore,  the  will  could  not  remain  the  obscure  dark 
potency  it  still  is  (Finsterniss).     Within  its  murky  gloom 
is  kindled  a  new  desire,   the  desire  to  exist  in  a  real, 
concrete,  that  is  to  say,  corporeal  fashion.     But  it  is  not 
of  itself  that  darkness  glows  and  becomes  fire,  or  that 
motionless  reason  is  changed  into  the  desire  to  live.    The 
term  or  goal  to  which  divine  will  tends  is  the  realisation 
of  the  personality,  the  excellent  form  of  life.    At  the  basis 
of  reason,  then,  there  was  light  as  well  as  darkness,  the 
dawn   of  perfect   life   as   well   as   the   dim  desire   of 
life  in  general ;  and  it  was  by  contact  with  the  new- 
born light  that  the  dark  was  kindled  and  became  fire. 
The  desire  to  live  is,  at  bottom,  the  will  to  live  well. 
And  so  the  possible  God  divides  himself  into  desire 
of  life   in   general   and   will    to    realise    perfect    life. 
These  are  no  longer  two  abstract,  ideal  entities,  but 
rather  two  forces,  alike  positive  and  living.     And  these 
forces  first  appear  as  two  rival  energies,  ready  to  enter 
upon  a  struggle  with  each  other.     For  the  love  of  life, 
when  left  to  itself,  impels  the  being  to  exist  in  every 
possible  manner  :   it  makes  no  distinction  between  good 
and  evil,  the  beautiful  and  the  ugly,  the  divine  and  the 
diabolical.     On  the  contrary,  the  will  to  live  well  and 
be  a  person  requires  a  choice  from  amongst  all  possible 
forms  of  life,  and  excludes  those  that  do  not  conform 
with  the  ideal.     The  dividing  of  the  eternal  no-thing 
into  passivity  and  activity,  desire  and  will,  had  produced 
only  the  logical  opposition  of  subject  and  object.     The 
dividing    of  will   into    negative    and    affirmative    will, 
fire  and  light,  force  and  love,  results  in  a  real  opposition 


192  STUDIES  IN  PHILOSOPHY 

and  the  beginning  of  internal  warfare  within  the  heart 
of  divinity.  The  first  of  the  two  rival  powers,  force  or 
life  in  general,  is  the  principle  and  the  mother  ;  the 
second,  love  or  enlightenment,  is  the  law  and  the  end. 
The  one  is  the  substratum  of  real  nature,  the  other  that 
of  divine  personality. 

In  this  second  opposition,  God  awakes  to  personal 
life  ;  set  over  against  nature,  however,  as  against  some 
inimical  power,  he  is  at  first  nothing  more  than  latent 
energy,  mere  capacity  for  love  and  light.  In  order 
that  this  energy  may  be  displayed  and  realised,  love 
must  enter  into  relation  with  force,  imposing  its  law 
on  this  latter.  And  so  the  progress  of  divine  revelation 
demands  a  reconciliation  of  the  two  contraries  that  have 
sprung  up  in  the  heart  of  will.  Now,  that  this  recon- 
ciliation may  come  about,  it  must  in  the  first  place  be 
posited  as  both  idea  and  goal  :  afterwards,  the  divine  will 
must  work  to  realise  this  idea.  But  the  reconciliation 
of  force  with  love,  or  of  fire  with  light,  is  nothing  else 
than  the  realisation  of  that  eternal  wisdom,  which  divinity 
has  formed  as  a  mirror  wherein  to  contemplate  and  know 
itself.  Thus,  what  has  to  be  done  is  to  bring  down  the 
idea  from  the  empty  heights  of  a  transcendent  heaven,  in 
order  to  blend  it  with  living  forces  and  manifest  it  in 
a  corporeal  nature.  Ideal  wisdom  as  an  object  to  be 
realised  :  that  is  the  third  term  superposed  on  the  two 
contraries  into  which  divine  will  has  divided  itself. 

How  will  the  new  task  resulting  from  the  position 
of  these  three  terms  be  accomplished  ?  Here  we  are  on 
the  plane  of  life  :  matter,  agent  and  end  are,  each  of 
them,  beings  endowed  with  force  and  activity.  It  is  by 
the  cooperation  of  these  three  principles  that  reconcilia- 
tion will  be  brought  about.  If  love  is  an  action  that 
tends  to  temper  force,  force  is  an  unconscious  impulse 


JACOB  BOEHME  193 

towards  love  ;  and  the  idea  itself,  ideal  wisdom,  seized 
with  the  desire  to  live,  tends  to  its  own  realisation  : 
the  virgin,  God's  companion,  aspires  after  the  mani- 
festation of  the  divine  wonders  slumbering  within  her. 
From  these  elements,  eternal  magic  forms  God  in 
person.  Will  is  linked  in  imagination  to  the  idea  it  pur- 
poses to  realise  ;  contemplating,  it  becomes  enamoured 
of  it ;  and,  eagerly  desiring  union,  seizes  upon  and 
absorbs  it.  It  absorbs  it  in  order  to  generate  it  within 
itself  and  produce  it  in  the  form  of  a  reality.  On  its 
side,  too,  the  idea  is  active  and  desires  existence  :  it  is 
a  soul  seeking  for  itself  a  body.  It  goes  to  meet  the 
will  that  is  calling  to  it.  The  idea  is  accordingly 
realised,  beneath  the  generating  action  of  imagination 
and  desire  :  spirit,  by  a  wholly  interior  operation, 
devoid  of  any  pree"xistent  corporeal  reality,  takes  to 
itself  a  nature,  an  essence  and  a  body. 

This  realisation  of  divine  wisdom  is  a  wonderful  and 
complex  work  which  it  is  important  to  consider  in 
detail. 

God  effects  it  by  means  of  seven  organising  spirits 
which  he  generates  with  a  view  to  this  task.  These 
spirits  are  the  forces  born  in  the  heart  of  the  dark 
element  beneath  the  influence  of  the  light  element,  forces 
whose  mission  it  is  to  transform  the  will  which  says 
"  no "  into  the  will  which  says  "  yes "  ;  to  discipline 
and  deify  nature.  Boehme  here  resumes  and  adapts  to 
his  system  the  ancient  kabbalistic  doctrine  of  the  seven 
natural  essences,  the  last  of  which  is  the  divine  kingdom. 
According  to  Boehme,  the  seven  spirits  are  born  in 
succession  from  one  another,  and  their  succession  marks 
the  progress  of  nature  in  the  direction  of  God.  The 
first  three  bring  nature  or  the  dark  element  to  the  point 
at  which  contact  will  be  possible  between  itself  and  the 

o 


1 94  STUDIES  IN  PHILOSOPHY 

light  element.  The  fourth  realises  this  contact,  and  the 
last  three  cause  light  and  love  to  reign  over  nature,  now 
prevailed  upon  and  induced  to  follow  spontaneously. 

First,  there  springs  up  in  the  will,  desire  properly 
so  called,  or  the  egoistic  tendency.  The  will  wills  to 
be  something.  Now,  there  is  nothing  over  against  it,  the 
possession  of  which  is  capable  of  determining  it.  There- 
fore it  takes  itself  as  object  and  wills  everything  for 
itself.  It  then  imagines  itself  to  be  something,  though 
it  is  still  nothing  but  hunger  and  emptiness.  This  first 
essence  is  the  dark,  the  solid,  the  force  of  contraction, 
the  salt  of  the  alchemists. 

Then  there  comes  about  motion,  as  the  second  essence 
or  second  natural  spirit.  For,  as  it  is  infinite  and  void, 
the  will  cannot  find  any  satisfaction  in  taking  itself  as 
object.  Therefore  it  turns  without  and  becomes  the 
acute,  the  bitter  ;  pain  :  that  spur  of  sensibility,  the 
force  of  expansion,  the  mercury  of  the  philosophers. 

Meanwhile  the  two  forces  thus  produced  are  in  con- 
flict with  each  other.  The  first  directs  being  towards 
itself,  the  second  directs  it  towards  something  else.  The 
result  of  this  opposition  is  the  third  essence,  restlessness, 
the  incessant  motion  of  a  soul  that  cannot  find  its  good 
within  itself  and  knows  not  where  to  look  for  it.  The 
two  forces  in  the  soul,  the  forces  of  contraction  and 
expansion,  are  contradictory,  and  yet  they  cannot  be 
separated  from  each  other.  The  soul,  void  in  itself, 
cannot  remain  fixed  in  egoism  :  moved  by  ego'fsm  even 
when  it  goes  forth  from  itself  and  seeks  its  good  without, 
it  cannot  attain  to  abnegation  and  love.  It  is  con- 
tinually fleeing  from  and  seeking  itself.  This  restless 
motion  is  that  of  a  wheel,  a  motion  which  reaches  no 
goal  and  yet  is  in  perpetual  pursuit  of  itself.  Thus, 
the  third  essence  has  for  its  expression  :  rotation,  or 


JACOB  BOEHME  195 

the  combination  of  the  centripetal  and  the  centrifugal 
forces.  It  forms  the  base  of  the  sulphur  of  the 
alchemists. 

Nature  rises  to  this  point  of  herself,  but  here  her 
power  stops.  She  has  shaken  off  the  dull  slumber  and 
ignoble  ease  of  ego'fsm,  and  sought  without  for  the 
thing  she  could  not  find  within.  To  the  eye  of  the 
body,  however,  the  exterior  infinite  is  just  as  void  as 
the  interior  infinite  ;  and  the  soul  has  done  no  more 
than  abandon  itself  to  two  contradictory  impulses  and 
bring  itself  into  a  state  of  embarrassment,  placing  itself 
before  a  spinning-wheel,  as  it  were.  This  interior  con- 
tradiction in  a  being  which  seeks  for  rest  by  means  of 
agitation,  is  an  intolerable  torture ;  but  nature,  of 
herself,  cannot  put  an  end  to  it.  She  has  exhausted 
all  her  resources  :  nothing  within  will  extricate  her  from 
the  state  in  which  she  is.  Salvation  can  come  only 
from  what  is  above  nature,  i.e.  from  God  or  eternal 
freedom.  But  how  will  these  two  contrary  powers 
succeed  in  reuniting  with  each  other  ? 

The  restlessness  with  which  nature  is  tortured  has 
this  advantage,  that  it  manifests  her  weakness  and  cries 
out  to  her  that  she  cannot  suffice  unto  herself  and  form 
a  whole.  The  man  who  knows  his  wretchedness  is  not 
so  wretched  as  he  who  is  ignorant  of  it.  Under  the 
influence  of  the  spirit  hovering  above  nature,  the  latter 
soon  feels  an  anxious  desire  for  freedom.  Something 
tells  the  soul  that  it  must  give  itself  to  that  which  is 
superior  to  it,  that  in  self-sacrifice  it  will  find  itself,  that 
in  dying  unto  itself  it  will  be  born  in  very  truth.  On 
the  other  hand,  spirit  and  freedom  need  nature  in  order 
to  manifest  and  realise  themselves.  If  nature  has  a  dim 
consciousness  of  her  own  law  and  harmony  in  spirit, 
the  latter  seeks  in  nature  his  own  reality  and  body. 


196  STUDIES  IN  PHILOSOPHY 

Spirit  wills  to  exist,  as  nature  tends  to  free  herself  from 
suffering.  Thus  impelled  in  the  direction  of  each  other, 
spirit  and  nature  approach  each  other.  Nature,  how- 
ever, has  her  own  distinctive  motion,  and  her  force 
of  inertia.  The  new  desire  she  feels  only  just  shows 
itself,  it  does  not  modify  her  wonted  course.  And 
so  nature  comes  into  collision  with  spirit  whom  she  is 
seeking  and  who  now  comes  down  to  her  ;  from  the 
impact  a  new  phenomenon  is  born  :  the  lightning  flash. 
This  is  the  fourth  moment  in  the  march  of  existence, 
the  fourth  essence.  This  moment  is  the  manifestation 
of  the  contact  of  nature  and  spirit.  In  the  flash  of  the 
lightning,  the  dark,  the  coarse  and  the  violent,  all  that 
makes  up  the  ego'lstic  tendency  of  nature,  is  swallowed 
up  and  reduced  to  nothingness.  The  darkness  is  illu- 
mined and  becomes  living,  manifest  fire,  the  centre  of 
light.  Henceforth,  nature  is  subject  to  and  capable  of 
realising  spirit.  There  has  come  to  pass  a  divine  law 
which  will  henceforth  apply  to  all  beings.  All  life,  accord- 
ing to  this  law,  implies  a  dual  birth.  Suffering  is  the 
condition  of  joy,  only  through  fire  or  by  way  of  the 
cross  do  we  attain  to  light.  Per  crucem  ad  lucem.  Both 
in  the  intellectual  and  the  physical  order  of  things, 
parturition  is  preceded  by  a  state  of  unrest  and  anxiety. 
Nature  labours  and  suffers,  feeling  she  has  not  the 
strength  necessary  to  bring  forth  the  fruit  she  has  con- 
ceived. Suddenly,  however,  a  supernatural  effort,  as  it 
were,  takes  place  ;  suffering  and  joy  clash  together  in 
one  indivisible  instant,  the  lightning  flashes  forth  and 
the  new  being  passes  out  of  darkness  into  light.  Hence- 
forth the  child  of  flesh  is  in  possession  of  his  own  form 
and  will  develop  by  himself  in  accordance  with  his  con- 
trolling idea  ;  the  fruit  of  intelligence  is  no  longer  a 
chaos  of  vague,  incoherent  ideas,  it  is  a  conscious  thought, 


JACOB  BOEHME  197 

sure  of  itself,  entering  unhesitatingly  into  the  expression 
which  manifests  it. 

With  the  appearance  of  the  lightning  flash,  the  first 
existence  of  divine  nature,  the  development  of  the 
negative  triad,  comes  to  an  end.  At  the  same  time 
there  begins  the  development  of  a  positive  triad,  repre- 
senting the  second  and  definitive  existence  of  nature. 
Contraction,  expansion  and  rotation,  will  be  found  in 
the  march  of  this  regenerate  nature,  though  in  a  new, 
a  supernatural  sense. 

The  new  concentration  is  the  work  of  love  :  the  uni- 
fying power  of  the  spirit.  Beneath  its  influence,  forces 
abandon  all  their  violence  and  take  delight  in  each  other. 
Ego'fstic  passions  die  away,  and  in  place  of  the  unity 
of  individuals,  each  one  of  whom  claims  to  exist  alone, 
there  is  substituted  a  unity  of  penetration  where  each 
seeks,  in  its  accord  with  the  whole,  a  participation  in 
true  unity.  Thus,  love  is  the  fifth  spirit,  the  fifth 
essence.  Its  symbol  is  water,  which  extinguishes  the 
fire  of  desire  and  confers  second  birth,  birth  according 
to  the  spirit. 

Still,  beings  ought  to  do  more  than  simply  melt 
into  each  other.  Their  unification  cannot  be  absorption 
and  annihilation.  The  march  of  revelation  ought  to 
make  multiplicity  perceptible,  even  to  that  profound 
spiritual  unity  conferred  by  love.  And  so  there  appears 
a  sixth  spirit,  which  releases  the  elements  of  the  divine 
symphony  and  causes  them  to  be  heard  in  their  indi- 
viduality as  well  as  in  their  relation  to  the  general  effect. 
This  sixth  spirit  is  the  intelligent  word,  or  sound,  by  means 
of  which  voices  cease  to  be  indistinct  noises,  for  they 
acquire  that  determination  which  makes  them  discern- 
ible and  comprehensible  in  themselves.  As  love  was 
the  unification  of  the  multiple,  so  the  sixth  essence  is 


198  STUDIES  IN  PHILOSOPHY 

the  perception  of  the  multiple  in  the  heart  of  unity 
itself. 

All  that  now  remains  to  be  done,  in  the  completion 
of  the  task  of  realising  God,  is  to  collect  and  coordinate 
all  the  forces  that  have  successively  created  each  other. 
If  the  higher  is  to  govern  the  lower,  it  is  not  to 
substitute  itself  therefor  and  annihilate  it,  for  the  lower 
is  its  reality  and  its  very  existence  ;  deprived  of  this 
support,  the  higher  element  would  be  dissipated  in  the 
void  of  transcendent  space.  Light  can  only  exist  with 
darkness  as  its  background.  Therefore  there  appears 
a  seventh  spirit  which,  winning  over  the  lower  to  the 
higher  by  persuasion,  and  bringing  down  the  higher 
into  the  lower  by  grace,  summons  the  whole  of  nature, 
great  and  small,  first  and  last,  to  the  manifestation  of 
the  divine  will.  This  essence  is  body  or  the  spirit  of 
harmony.  By  its  action  the  revelation  of  the  Eternal 
is  finally  accomplished.  Wisdom  is  no  longer  an  idea. 
It  is  a  kingdom  of  living  beings,  the  kingdom  of  God 
or  of  Glory. 

Thus  Boehme  regards  as  a  reality  and  an  essential 
condition  of  divine  life,  this  uncreated  heaven,  the 
kingdom  of  the  Father,  the  glory  of  God,  of  which 
the  Scriptures  speak  in  so  many  places  though  the 
language  used  is  often  interpreted  metaphorically.  The 
lily  is  clad  in  beauty,  a  beauty  surpassing  the  splendour 
of  Solomon.  Man  has  his  vesture  of  glory  :  his  wealth 
and  his  home,  power  and  honours,  all  that  manifests  his 
invisible  personality.  God,  too,  reveals  himself  in  a 
phenomenon  that  has  no  other  content  thaa  himself,  and 
yet  is  distinct  from  him.  The  Glory  of  God  is  his 
vesture,  his  outer  form,  his  body  and  reality  :  it  is  God 
seen  from  without. 

To  describe  the  harmony  and  beauty  of  this  kingdom 


JACOB  BOEHME  199 

of  Glory  is  impossible.  It  sums  up  all  we  see  on  earth, 
though  in  a  state  of  perfection  and  spirituality  to  which 
the  creature  cannot  attain.  Its  colours  are  more  brilliant, 
its  fruits  more  savoury,  its  sounds  more  melodious,  and 
its  whole  life  more  happy.  Along  with  purity  of  spirit, 
divine  beings  possess  the  full  reality  of  body.  Their 
life  is  not  an  imperfectly  satisfied  desire  :  it  is  being 
in  all  its  fulness  and  completion.  Above  all  else  it  is 
harmony,  reconciled  with  the  free  and  perfect  growth  of 
all  individuals.  Consider  the  birds  in  our  forests :  each 
praises  God  in  its  own  fashion,  in  all  keys  and  modes. 
Do  we  find  God  offended  by  this  diversity,  does  he 
impose  silence  on  the  discordant  voices  ?  All  forms  of 
being  are  precious  in  the  sight  of  Infinite  Being.  But  if 
divine  gentleness  is  manifested  in  our  world,  a  fortiori, 
beings  in  the  kingdom  of  Glory  are  free  from  all  restraint, 
since  all  in  that  kingdom,  each  according  to  its  nature, 
not  only  seek  God  but  also  possess  and  manifest  him. 

Such,  in  its  completion,  is  eternal  nature,  the  revela- 
tion of  the  divine  mystery.  She  carries  within  herself 
three  principles,  the  three  reasons,  as  it  were,  or  bases 
of  determination,  born  of  primordial  no-thing.  The 
first  principle  is  the  substratum  of  the  first  three  qualities, 
or  of  nature  left  to  herself.  It  is  darkness,  or  latent 
fire,  waiting  for  the  spark  in  order  to  become  manifest. 
Boehme  generally  calls  it  fire.  The  second  principle  is 
the  substratum  of  the  last  three  qualities,  i.e.  of  the 
form  or  expression  of  ideal  wisdom.  This  is  the  prin- 
ciple of  light.  Each  of  these  principles  is  eternal  ;  and, 
in  a  sense,  they  exclude  each  other.  Fire  admits  of  no 
limit,  it  devours  everything  with  which  it  is  brought 
into  contact.  Light  is  the  absolute  of  sweetness  and 
joy,  the  negation  of  darkness,  the  goal  of  all  aspiration. 
The  former  is  the  life  of  the  all  or  of  the  inde- 


200  STUDIES  IN  PHILOSOPHY 

terminate  infinite  :  the  latter  is  the  life  of  God  or  of 
the  excellent,  the  determinate  one.  Still,  neither  of  these 
two  principles  can  suffice  unto  itself.  In  vain  does  fire 
will  to  be  the  whole,  it  is  only  a  part.  In  vain  does 
light  scorn  darkness  :  it  is  realised  only  when  reflected 
from  the  dark.  That  is  why  a  third  principle  is  neces- 
sary, which  will  unite  the  first  to  the  second,  in  such  a 
manner  as  to  produce  real  existence.  This  third  principle 
is  body.  By  it,  spirit  incarnates  in  matter  and  becomes 
real  and  living.  This  union  of  the  first  principle  with 
the  second  is,  after  all,  not  a  complete  absorption,  and 
the  three  principles  remain  irreducible.  Indeed,  the 
operation  which  places  fire  under  the  laws  of  light  does 
not  annihilate  even  the  basis  of  fire.  Infinitude  of  life 
subsists  beneath  the  form  of  perfection  that  determines 
it.  The  divine  command  is  not  addressed  to  slaves  ; 
it  wills  to  have  free  beings  and  finds  them.  Fire,  light, 
body,  i.e.  life,  good  and  their  union  in  one  real  being  ; 
such  are  the  three  principles  of  divine  nature. 

We  must  now  take  care  not  to  identify  this  nature 
with  the  true  God.  However  excellent  she  be, 
divine  nature  exists  neither  in  herself  nor  for  her 
own  benefit.  She  is  the  realisation  of  the  perfections 
comprised  in  the  idea  of  wisdom.  She  is  the  eternal 
virgin,  who,  at  the  voice  of  God,  has  come  down  from 
the  limbo  of  the  possible  into  the  paradise  of  actual 
existence.  Nature  will  now  return  thanks  to  her 
creator,  handing  over  to  him  her  life  and  her  bodily 
existence.  The  eternal  virgin,  fecundated  by  spirit, 
henceforth  brings  to  birth,  and  the  fruit  of  her  womb 
is  God  the  person,  i.e.  the  God  who  not  only  knows 
and  possesses  himself,  but  also  projects  himself  with- 
out, in  love  and  action.  Whereas  the  latter,  God  the 
person,  set  before  himself,  as  a  mirror  of  his  infinite 


JACOB  BOEHME  201 

will,  eternal  wisdom  or  the  idea  of  divinity,  God  con- 
stituted himself  only  as  ideal  trinity,  a  possible  person- 
ality. By  giving  himself  in  nature  a  living  contrary, 
and  bending  this  contrary  to  the  laws  of  his  good  will, 
God  enters  upon  a  differentiation  which  is  real  and  no 
longer  ideal,  and  hence  attains  to  effective  personality, 
that  of  the  Christian  trinity.  Self-knowledge  confers 
only  existence  for  self;  action  alone  generates  absolute 
existence  and  completes  personality. 

Now,  this  action  is  threefold  :  it  posits  three  per- 
sons corresponding  to  the  three  principles  of  nature. 
In  the  first  place,  God  is  the  will  that  presides  over 
life  in  general,  or  over  eternal  fire.  In  this  sense  he  is 
the  Father,  power,  justice,  divine  wrath  :  he  is,  as  it 
were,  the  consciousness  of  infinite  vital  activity.  God, 
however,  does  not  desire  life  for  life  itself.  His  will 
is  to  have  life  as  a  realisation  of  idea,  to  generate  the 
living  word.  This  is  why  the  Father  gives  birth  to 
the  Son,  who  is  the  consciousness  of  the  second  prin- 
ciple or  of  light,  and  wills  the  subordination  of  life  to 
good,  its  raison  d'etre.  By  the  Son,  the  God  of  love 
and  compassion,  the  fire  of  wrath  is  for  ever  appeased. 
Accordingly,  the  Son  is  greater  than  the  Father.  Still, 
the  existence  of  the  good  will  as  against  the  universal 
will  to  live  is  not  sufficient  to  realise  the  good  ;  these 
two  wills  must  come  together  and  become  reconciled, 
and  this  is  what  takes  place  in  a  third  consciousness 
and  a  third  person,  whence  proceeds  the  third  principle 
called  the  Holy  Ghost. 

Thus,  whilst  forming  eternal  nature,  and  by  reason 
of  the  very  activity  expended  in  forming  it,  God  truly 
constitutes  himself  Father,  Son,  and  Spirit,  without  on 
that  account  abdicating  his  unity.  Because  the  three 
realisations  of  God  are  indeed  persons  and  not  things, 


202  STUDIES  IN  PHILOSOPHY 

they  do  not  come  under  that  law  of  time  and  space 
which  insists  that  unity  is  incompatible  with  multi- 
plicity. Personality  admits  of  mutual  penetration  : 
further  than  that,  it  implies  it.  Only  in  its  union 
with  other  persons  can  a  personal  being  be  constituted 
as  such.  In  so  far  as  a  being  is  conceived  as  external 
to  other  beings  it  is  constituted  in  space  and  attributes 
to  itself  individuality,  that  enemy  of  true  personality. 
Ego'fsm  is  the  basis  of  individuality  :  it  is  the  gift  of 
oneself  that  makes  the  person. 

The  generation  of  God  is  now  accomplished.  God 
is  perfect  personality  realised  in  three  persons,  each  of 
which  is  the  part  and  the  whole,  at  one  and  the  same 
time.  These  three  persons  are  the  Father,  or  the  con- 
sciousness of  force  ;  the  Son,  or  the  consciousness  of 
good  ;  and  the  Spirit,  or  the  consciousness  of  the 
harmony  set  up  in  God  between  force  and  good.  And 
over  against  God,  as  being  his  work  and  his  glory, 
stands  arrayed  eternal  nature,  in  whom  all  possibilities 
are  realised,  in  proportion  as  they  express  divine  per- 
fection. 

Such  is  the  teaching  of  Boehme  regarding  the  birth 
of  God.  Through  the  theological  and  alchemical 
symbols  in  which  this  teaching  is  clothed  for  the  pur- 
pose of  self-manifestation  is  it  not  clear  that  it  pos- 
sesses a  philosophical  meaning  and  import  ?  The  main 
idea  of  the  teaching  is  that  the  person  is  the  perfect 
being  and  must  exist  ;  consequently,  that  all  the  condi- 
tions of  the  person's  existence  must  themselves  be 
realised.  From  this  principle  all  else  follows.  Per- 
sonality, says  Boehme,  implies  thought  and  action  ; 
now,  in  order  to  think  and  act,  one  must  be  in  rapport 
with  something  opposed  to  oneself.  Thought  must 
have  some  object  to  consider  and  resolve  into  itself ; 


JACOB  BOEHME  203 

action  must  have  matter  which  it  may  subdue  and 
spiritualise.  This  law  is  universal ;  absolute  person- 
ality itself  could  not  escape  from  it  without  contradic- 
tion. On  the  other  hand,  absolute  being  must  be  self- 
caused,  must  depend  on  nothing  foreign  to  itself.  Thus 
if  absolute  being  wills  to  become  person,  it  must  draw 
from  itself  an  object  opposed  to  itself,  to  which  its  in- 
telligence applies,  and  which  its  activity  modifies.  It  is 
necessary  that  the  one  infinite  divinity  be  transformed 
of  itself  into  a  duality,  one  of  whose  terms  will  be  the 
true  God,  and  the  other  will  be  nature,  of  whom  this 
God  has  need.  Thus  conceived  of  as  being  subject 
and  agent  as  against  object  and  matter  springing  from 
his  own  inmost  being,  God  has  a  task  to  perform  :  the 
solution  of  the  antinomy  he  has  created  within  him- 
self; and  by  the  accomplishment  of  this  task  he  realises 
himself  qua  person.  His  action  and  thought,  life  and 
existence,  are  henceforward  something  else  than  the 
shadow  of  human  life  and  activity  :  they  are  perfect 
types  of  which  the  existence  of  creatures  affords  us 
nothing  but  feeble  images. 

Now,  what  is  this  system  wherein  God  generates 
himself  by  positing  and  rising  above  his  contrary  ?  Is 
it  not  that  ancient  doctrine  of  Night  as  a  first  principle 
which  Aristotle  had  already  repudiated  in  his  prede- 
cessors ?  The  first  being,  said  Aristotle,  is  not  the 
imperfect,  but  the  perfect ;  in  the  order  of  phenomena, 
the  perfect  is  subsequent  to  the  imperfect ;  but  in  the 
order  of  being,  it  is  the  perfect  that  is  first  and 
absolute.  Boehme's  doctrine,  like  that  of  the  old 
theologians,  appears  to  be  only  an  anthropomorphism 
or  a  naturalism.  He  noticed,  we  may  say,  that  in  the 
case  of  man,  indetermination  precedes  determination  ; 
that  struggle  is  the  condition  of  life  and  progress  ; 


204  STUDIES  IN  PHILOSOPHY 

that  an  image  is  necessary  for  the  understanding,  and 
matter  for  the  will ;  that  the  action  of  our  faculties 
consists  in  assimilating  to  oneself  external  objects  ;  and 
he  transferred  to  God  this  condition  of  human  existence. 

Even  were  this  judgment  well  founded  we  could  not 
regard  it  as  a  condemnation  of  the  doctrine,  purely  and 
simply.  Though  Boehme's  system  were,  in  reality,  to 
apply  only  to  finite  beings,  it  would  not,  on  that 
account,  be  without  importance.  We  must  forgive 
our  theosophist  for  his  imperfect  teaching  as  to  the 
history  of  the  divine  trinity,  if,  when  thinking  he  is 
speaking  to  us  of  God,  he  is  really  speaking  of  our- 
selves, and  that  with  much  sagacity.  The  great  prin- 
ciple that  will  is  the  basis  of  life  and  existence,  and 
that,  in  its  turn,  life  finds  in  freedom  its  end  and  raison 
d'etre^  will  lose  none  of  its  interest  by  being  concerned 
only  with  the  created  world  instead  of  being  applied  to 
the  Creator  as  well.  This  strange  system,  whose  very 
opulence  is  utter  confusion,  and  whose  glory  is  dazzling 
lightning,  contains  many  a  delicate,  modest,  and  psycho- 
logical observation,  many  a  wise,  practical,  and  moral 
reflection.  As  Boehtne  tells  us,  it  is  in  the  depths  of 
his  consciousness  that  he  seeks  after  divinity  ;  it  is 
because  God  generates  himself  in  man  that  man  can 
be  made  acquainted  with  divine  generation.  What 
wonder  if  his  knowledge  of  God  is,  above  all,  knowledge 
of  ourselves  ? 

Moreover,  it  does  not  follow  that  Boehme,  from  the 
metaphysical  point  of  view,  is  a  mere  naturalist.  With- 
out delighting,  as  he  does,  in  speculations  that  we 
cannot  possibly  verify  regarding  the  birth  and  develop- 
ment of  God,  at  all  events  we  can  see  the  difference 
between  his  teaching  and  that  rejected  by  Aristotle. 
According  to  the  ancient  philosophy  of  chaos  and  the 


JACOB  BOEHME  205 

infinite,  the  generation  of  the  perfect  by  the  imperfect 
was  the  absolute  reality  of  things.  To  Boehme  there  is 
no  before  or  after  in  God,  the  absolute.  It  is  our  con- 
dition as  finite  and  belonging  to  nature  that  forces  us 
to  regard  God  from  the  standpoint  of  nature,  and  to 
picture  to  ourselves  his  life  as  being  progressive. 

This  is  not  all,  however.  The  chaos  of  the  ancients 
was  a  given  nature,  a  thing,  and  that  the  most  confused 
and  indeterminate  conceivable  ;  and  from  this  thing,  by 
a  necessary  process  of  development,  determinate  and 
perfect  being  was  brought  forth.  The  standpoint  of 
the  ancients  was  an  objective  one.  Aristotle,  under  the 
name  of  pure  action,  contrasts  the  thing  that  is  wholly 
determinate  with  the  thing  that  is  wholly  indeter- 
minate ;  whereas  Neo-Platonism,  returning  to  the  idea 
of  progress,  posits,  as  first  being,  a  unity  which,  superior 
or  inferior  to  intelligence  and  life,  unnamable  and 
unintelligible,  still  seems  to  be  only  the  thing,  stripped 
of  the  last  of  its  qualities  by  the  final  effort  of  abstrac- 
tion. The  principle  of  our  theosophical  mystic  is 
something  quite  different.  A  Christian  and  a  spirit- 
ualist, he  assigns  the  first  place  to  personality  in  its 
most  perfect  form.  From  the  point  of  view  at  which 
he  is  placed,  indetermination,  infinitude,  no-thing  have 
quite  different  meanings  from  those  contained  in  ancient 
philosophy.  No  longer  is  no-thing  the  lack  of  quality 
and  perfection  in  a  thing  that  can  exist  only  if  it  is 
determinate  ;  it  is  the  infinite  fecundity  of  a  spirit 
which  is  by  its  very  potency,  and  is  exhausted  by  none 
of  its  productions.  Negative,  from  the  outer  stand- 
point of  objectivity,  Boehme's  principle  is  altogether 
positive  from  the  inner  standpoint  of  life  and  genera- 
tion. In  itself  this  principle  is  not  the  imperfect,  it  is 
the  perfect ;  and  the  progress  admitted  by  Boehme, 


206  STUDIES  IN  PHILOSOPHY 

though  in  a  way  relative  to  the  human  mind,  is  pro- 
gress in  manifestation,  not  in  the  intrinsic  perfection  of 
God.  The  system  of  the  metaphysical  world  has  been 
inverted  ;  no  longer  is  it  intelligence  that  depends  on 
the  intelligible,  it  is  the  intelligible  that  depends  on  in- 
telligence. It  is  no  longer  the  subject  that  derives  its 
existence  from  the  object,  it  is  the  object  that  exists  by 
the  subject.  The  reason  this  substitution  has  come 
about  is  that  man  has  discovered  in  that  which  consti- 
tutes the  foundation  of  the  subject,  in  mind  and  will, 
something  irreducible  that  baffles  description,  and  which 
he  regards  as  more  real  in  its  indetermination  and 
nothingness  than  all  the  tangible  realities  of  given  sub- 
stance. 

Thus,  Boehme's  course  is  by  no  means  that  of 
the  Pythagoreans  or  even  of  the  Neo-Platonists.  The 
progress  which  proceeds  from  will  to  its  workings 
cannot  be  assimilated  to  the  progress  which  proceeds 
from  the  indeterminate  thing  to  the  determinate. 
The  theology  of  Boehme  is  not  an  evolutionistic 
monism. 

Nor,  on  the  other  hand,  is  it  a  system  of  dualism  : 
Does  it  not,  indeed,  appear  as  though  Boehme  escapes  one 
danger  only  to  fall  into  an  opposite  one  ?  How  does 
Boehme  maintain  the  perfection  of  the  divine  principle 
unless  it  be  by  positing,  outside  of  God,  as  a  subject  of 
evil,  a  hostile  and  coeternal  principle  ?  And,  accord- 
ing to  him,  God  himself  is  one  with  and  responsible  for 
this  latter  principle.  Per  crucem  ad  lucem  :  this  is  both 
the  divine  and  the  human  law.  No  light  without  dark- 
ness, no  action  without  matter,  no  subject  without  object, 
no  God  without  nature.  Is  it  not  just  this  universal 
and  necessary  coexistence  of  two  principles,  the  one 
positive,  the  other  negative,  that  is  called  dualism  ? 


JACOB  BOEHME  207 

It  cannot,  indeed,  be  denied  that  Boehme  sees  in 
matter  the  condition  of  the  manifestation  of  spirit  ;  this 
is  even  an  essential  part  of  his  system.  But  Boehme 
does  not  regard  himself  as  a  dualist  on  that  account. 
In  his  eyes  it  is  monstrous  to  make  evil  the  equal  of 
good,  and  nature  the  equal  of  God.  The  negative 
principle  does  not  exist  in  itself,  but  only  by  the  action 
of  the  positive  principle,  which  creates  it  in  order  to 
manifest  therein.  God  alone  is  sovereign  ruler,  and 
it  is  the  internal  motion  of  divine  will  that  posits 
matter,  outside  of  God,  as  the  condition  of  this  very 
motion.  Matter  is  the  exterior  aspect,  the  phenomenon 
of  the  invisible  action  of  spirit.  It  fixes  in  dead  forms 
the  continuous  flashing  forth  of  living  light.  Dependent 
on  spirit  as  regards  her  origin,  nature  is  subject  to 
spirit  as  regards  her  final  purpose.  Her  end  is  to  supply 
spirit,  by  manifesting  it,  with  the  object  it  needs  in 
order  to  lay  hold  of  and  personify  itself.  She  resists 
spirit  only  in  order  to  afford  it  an  opportunity  to  display 
its  might,  her  instinct  is  an  intelligence  that  is  ignorant 
of  itself ;  her  passion,  an  unconscious  desire  for  freedom. 
Far  from  nature  being  the  equal  of  God,  it  is  at  God's 
summons  that  she  begins  to  exist,  and  the  limit  of  her 
development  is  her  exact  adaptation  to  the  will  of 
spirit. 

Thus  Boehme's  theology  borders  on  dualism  as  it 
did  on  evolutionism  without  running  counter  to  it  or 
foundering  therein.  At  bottom,  Boehme  purposes  to 
find  a  middle  term  between  these  two  doctrines.  In 
his  opinion,  the  mystics  of  old  were  in  the  wrong  when 
they  rejected  dualism  altogether.  This  was  the  reason 
they  could  not  realise  the  philosophy  of  personality  that 
they  had  conceived.  Their  God  lacks  the  conditions 
of  real  existence,  he  does  not  outstep  the  limits  of  ideal 


208  STUDIES  IN  PHILOSOPHY 

existence.  It  is  only  by  borrowing  from  dualism 
the  idea  of  the  eternal  existence  of  matter  as  contrary 
to  spirit  and  giving  this  matter  as  a  body  to  divine 
spirit,  that  divine  personality  can  be  conceived  of  as 
really  existing.  But,  on  the  other  hand,  God  the 
person  must  remain  infinite  being  outside  of  which 
nothing  exists  in  itself.  Dualism  is  repugnant  to 
religious  thought  which  would  have  God  not  only  a 
form  and  an  ideal,  but  also  omnipotent  and  independent 
being.  Thus,  matter  must  not  be  a  first  being  for  the 
same  reason  that  God  is,  its  very  existence  must  result 
from  the  working  of  divine  power.  How  can  matter 
issue  from  God  and  yet,  at  the  same  time,  be  the  con- 
trary of  God  ?  Boehme  solves  the  difficulty  by  saying 
that  God,  in  order  to  reveal  himself,  makes  himself 
objective  and  real,  and  that  this  object  and  this  exterior 
reality,  though  posited  by  God,  are  not  confounded 
with  him,  because  will,  which  is  the  basis  of  his  being, 
is  infinite  ;  its  efforts  cannot  possibly  be  wasted.  Thus 
God  has  a  nature  or  body  that  is  not  himself  and  that 
forms  his  real  existence  ;  but  this  body  is  posited  by 
God  and  is  none  other  than  his  will  itself,  seen  from 
without.  In  this  phenomenon  of  God,  the  eternal 
mystery  is  revealed,  without  the  revelation  ever  dispelling 
the  mystery.  Nature  is  of  the  essence  of  God  but  God 
is  independent  of  nature.  This  system  is  a  kind  of 
concrete  or  naturalistic  spiritualism. 

IV 

The  knowledge  of  divine  genesis  is  the  first  we  need, 
in  order  to  attain  to  the  possession  of  God.  But  this 
knowledge  is  not  enough.  It  was  a  mistake  on  the 
part  of  the  mystics  to  believe  that  all  science  was  com- 


JACOB  BOEHME  209 

prised  in  the  science  of  God.  Nature  and  man  cannot 
be  explained  by  a  mere  diminution  of  perfect  essence. 
In  creatures  there  is  something  peculiar  to  themselves 
that  distinguishes  them  from  God  and  even  allows  them 
to  rebel  against  him.  Evil,  the  work  of  creatures,  is 
not  a  non-being,  it  is  a  being  that  says  no  ;  hatred  that 
would  destroy  love  ;  violence  that  would  break  the  law. 
Accordingly,  there  is  a  science  of  nature,  apart  from 
the  science  of  God.  The  difficulty  consists  in  account- 
ing for  this  distinction  whilst  maintaining  that  relation 
of  dependence  which  should  link  all  science  with  that  of 
absolute  being. 

The  first  problem  raised  by  the  existence  of  nature  is 
that  of  creation.  On  this  point  Boehme  cannot  adopt 
the  doctrine  usually  called  theism.  According  to  this 
doctrine,  it  would  appear  that  God  made  the  world 
from  absolute  no-thing,  i.e.  created  it  by  his  infinite 
will  alone,  without  using  any  matter  at  all,  either 
sensible  or  suprasensible.  But  such  a  world  would  have 
no  true  reality,  for  its  reality  would  not  be  founded  in 
God.  It  would  be  simply  a  possible  and  ideal  world, 
like  the  very  principle  to  which  it  would  owe  its  birth  : 
intelligence  without  matter  creates  only  ideas.  There- 
fore there  is  no  true  personality  in  creatures.  The 
reason  some  are  good  and  others  bad,  some  predestined 
to  happiness  and  the  rest  given  up  to  damnation,  is  not 
because  there  are  living  and  opposite  energies  in  the 
souls  of  creatures,  it  is  because  it  has  so  been  willed 
by  the  God  who  transcends  all  arbitrary  wills.  Idealism 
and  fatalism  are  the  consequences  of  the  doctrine  of 
theism. 

Still,  if  Boehme  rejects  theism,  will  he  not,  as  a  con- 
sequence, sink  into  pantheism  ?  We  know  that  he 
recognises  in  God  the  existence  of  a  nature.  Is  it  not 

p 


210  STUDIES  IN  PHILOSOPHY 

this  nature  that  is  to  constitute  the  substratum  of 
visible  nature  ?  Can  the  latter  be  anything  else  than  a 
development  of  the  former  ;  and  must  we  not  say,  with 
the  pantheists,  that  the  world  is,  if  not  God  himself,  at 
all  events  the  body  and  manifestation  of  God  ? 

Certainly  such  an  interpretation  would  be  contrary 
to  Boehme's  plan,  which  is  even  more  energetically 
opposed  to  pantheism  than  to  theism.  Surely,  he  says, 
in  one  sense  God  is  everything,  heaven  and  earth,  spirit 
and  world ;  for  everything  has  its  origin  in  him.  But 
then,  what  becomes  of  his  glorious  immensity  if  the 
world  is  the  standard  of  his  perfection  ?  Doubtless  he 
created  the  world  by  his  wisdom  and  might  :  but  he  did 
not  form  it  so  that  he  himself  might  become  more 
perfect.  His  perfection  is  complete  independently  of 
all  creation.  God  formed  the  world  so  as  to  be  mani- 
fested in  a  manner  that  would  be  sensible.  Let  not 
sophists  tell  me  that,  in  my  doctrine  of  the  divine 
nature,  I  am  confounding  God  with  the  world.  I  am 
not  confounding  exterior  with  interior  nature.  The 
latter  is  truly  living  and  is  perfect.  The  other  has 
nothing  but  a  derived  life,  and  remains  imperfect.  No, 
the  exterior  world  is  not  God,  nor  could  it  without 
blasphemy  be  called  God.  To  say  that  God  is  all,  that 
God  is  himself  and  heaven  and  earth  and  the  outer 
world,  is  to  speak  as  the  heathen,  to  make  profession  of 
the  devil's  religion. 

Boehme's  problem,  therefore,  is  to  derive  matter 
from  spirit  and  yet  not  sink  into  theism,  and  to  base 
sensible  nature  on  divine  nature  without  falling  into 
pantheism.  How  does  he  solve  the  problem  ? 

Whereas  the  birth  of  God  was  a  mere  generation, 
i.e.  a  magical  production  accomplished  by  spirit  through 
its  two  powers  at  once  homogeneous  and  contrary, 


JACOB  BOEHME  211 

and  without  any  pre-existing  matter,  the  birth  of  the 
world  is  a  creation,  or  production  brought  about  by  a 
spiritual  agent  through  matter.  The  spiritual  agent  is 
the  one  God  in  three  persons.  Matter  is  eternal 
nature.  Neither  of  these  two  principles  is  the  world,  or 
contains  it.  God  the  person,  as  such,  is  pure  spirit. 
Eternal  nature  is  perfect  harmony,  in  which  beings, 
although  distinct,  interpenetrate  :  it  is  a  multiplicity 
each  part  of  which,  in  its  own  way,  expresses  the  unity 
of  the  whole.  These  perfections  radically  distinguish 
God  and  the  divine  nature  from  the  sensible  and 
created  world,  which,  on  the  one  hand,  is  material,  and 
on  the  other  consists  of  parts  and  fragments  exterior  to 
one  another.  But  though  God  the  person  and  eternal 
nature  are  not  the  world,  they  contain  its  elements  ;  the 
world  has  its  own  mobility  and  reality  so  far  as  there  is 
in  it  something  of  the  divine  perfection.  And  first  ,God, 
seeing,  from  all  eternity,  in  wisdom,  the  ideas  of  things, 
formed  the  design  of  creating  the  world,  i.e.  of  causing 
to  exist  in  corporeal  fashion  what  existed  in  him  in 
essential  fashion,  or  rather  of  causing  to  appear  separate 
what,  in  him,  was  together.  He  formed  this  design 
from  love  alone,  without  being  constrained  or  forced 
thereto  in  any  way.  There  is  not  the  slightest  reason 
for  creation.  Its  wherefore  is  a  mystery  and  admits  of 
no  revelation  whatsoever.  If  creation  had  its  first 
origin  in  the  manifested  God  and  not  in  the  primordial 
abyss,  it  would  be  explained,  it  would  be  necessary,  and 
would  force  itself  upon  God.  But  God  wills  to  have 
children,  not  masters.  Though  the  world  depends  on 
God,  God  has  no  need  of  the  world. 

The  world  was  not  made  from  some  thing,  i.e. 
brute  matter,  the  absolute  contrary  of  a  person.  It  was 
made  of  the  divine  nature,  in  the  sense  that  the  seven 


212  STUDIES  IN  PHILOSOPHY 

spirits  constituting  this  nature  realised  in  the  form  of 
bodies  the  ideas  contained  in  wisdom.  The  productions 
of  these  spirits  in  the  world  of  Glory  were  figures  with 
floating  contours,  instinct  with  life  and  spirituality :  the 
infinite  visible  in  the  finite.  The  same  spirits  now 
fix  the  idea  in  hard  compact  matter  which  conceals 
the  infinite  that  it  realises.  In  the  world  of  Glory  the 
real  and  the  ideal  balance  each  other  :  in  the  created 
world,  it  is  the  real  that  predominates. 

Such  is  the  portion  of  God  the  person,  such  the 
portion  of  the  divine  nature  in  creation.  A  third 
worker,  however,  intervenes  in  order  to  realise  the 
world,  this  worker  is  the  creature  itself.  Just  as  when 
the  artist  is  working,  the  work  itself,  that  wills  to  be, 
furthers  by  its  distinctive  life  the  efforts  of  will  and 
intelligence  ;  so  the  creature,  when  brought  to  the 
threshold  of  existence  by  the  union  of  spirit  and  increate 
nature,  endeavours  to  cross  this  threshold  and  display 
itself  in  fulness  of  light.  All  spirit  is  a  soul  which  desires 
a  body.  Now,  the  creative  word  had  the  effect  of 
breaking  the  bond  that  held  together  the  spiritual  forces 
in  union  and  harmony.  Each  of  them,  thenceforward, 
wills  to  exist  for  itself,  to  become  manifest  in  accordance 
with  its  distinctive  tendency. 

What,  then,  is  creation  ?  It  is  the  introduction  of 
space  and  time  into  the  world  of  particular  wills.  Deep 
in  the  heart  of  eternity,  wills,  individual  in  themselves, 
were  universal  in  their  object.  Realised  in  bodies 
separated  from  one  another  by  time  and  space,  wills  are 
thereby  detached  from  the  all  and  thrown  back  upon 
themselves.  Thus,  space  and  time  are  the  special 
foundation  of  the  reality  of  the  sensible  world.  Here, 
there  is  nothing  that  does  not  come  from  God,  but 
nothing  that  was  in  God  could  produce  this  form  of 


JACOB  BOEHME  213 

existence  by  mere  development  :  it  is  by  a  free,  original 
act,  a  veritable  creation,  that  God  causes  the  world  of 
discontinuity  and  exteriority  to  appear. 

God,  then,  is  by  no  means  swallowed  up  in  his 
creation,  any  more  than  the  intelligence  of  man  is 
exhausted  by  being  manifested.  The  divine  will  is  as 
tenuous  as  a  no-thing.  No  given  solid  being  is  capable 
of  enclosing  it  within  itself  and  making  it  immovable. 
Besides,  the  world  does  not  issue  from  God  himself,  but 
from  his  glory,  i.e.  his  exterior  form.  And  this  very 
glory,  the  periphery  of  divinity,  remains  after  creation 
what  it  was  before.  For  if  the  less  is  included  in  the 
more,  the  more  is  not  included  in  the  less  ;  a  fortiori^  the 
different  cannot  be  included  in  the  different.  Neither 
as  subject  nor  as  object  is  divinity  absorbed  in  its 
sensible  manifestation.  Creation  is  not  at  all  a  trans- 
formation of  force. 

Thus  God  creates,  at  the  same  time,  from  nothing 
and  from  matter.  God  the  person  creates  with  the 
divine  nature  as  matter,  but  personality  and  the  divine 
nature  alike  have  their  root  in  the  primordial  no-thing, 
in  the  mystery  of  infinite  will. 

Now  what  is  it  that  God  creates,  what  are  the 
essential  parts  of  the  world  system  ?  The  model  and 
instruments  of  creation  are  found,  under  the  form  of 
eternity,  in  the  divine  wisdom  and  nature.  Creation  is 
to  be  the  realisation  of  this  wisdom  and  nature  under  the 
form  of  time  and  separation.  And  so  there  is  a  relation 
between  created  things  and  eternal  things,  and  it  is 
to  a  certain  extent  possible  to  deduce  from  the  latter 
the  knowledge  of  the  former,  by  placing  oneself  at  the 
standpoint  of  God.  This  deduction  is  what  is  called 
the  philosophy  of  nature,  a  speculation  destined  later 
on  to  assume  a  considerable  degree  of  development 


2i4  STUDIES  IN  PHILOSOPHY 

in    Germany,    and    rudiments    of   which    we    find    in 
Boehme's  theosophy. 

The  construction  of  the  exterior  world  is  brought 
about  in  a  manner  similar  to  that  of  the  interior,  divine 
world.  In  sensible  bodies  as  in  eternal  nature,  it  is 
personality  that  seeks  manifestation  for  itself  :  the  only 
difference  is  that  this  manifestation,  which  is  fully 
effected  in  eternal  nature,  remains  of  necessity  incom- 
plete in  sensible  nature.  In  the  world  there  will  then  be 
three  principles  corresponding  to  the  three  divine  prin- 
ciples :  fire,  light,  and  the  union  of  these  two  principles 
in  corporeity.  Of  the  first  and  second,  without  appeal- 
ing to  the  third,  God  forms  the  angels,  who  are  still 
as  near  to  divine  perfection  as  the  created  condition 
permits  of.  The  angels  are  spirits  only.  They  do  not 
exist  of  themselves,  however,  and  their  body,  though 
spiritual,  is  harder  and  more  compact  than  the  glorious 
body  of  divinity.  The  angels  are  not  yet  placed  in 
time  ;  they  enjoy  a  derivative  eternity  intermediary 
between  absolute  eternity  and  the  succession  of  parts 
independent  of  one  another.  At  the  same  time  that 
God  formed  the  angels  from  the  first  two  principles,  he 
formed  from  the  third  a  terrestrial  nature,  more  concrete 
and  material  than  the  divine  though  still  subject  to  spirit 
and  relatively  harmonious.  This  nature  is  governed  by 
the  angels.  All  these  beings  were  created  in  order  that 
divine  light,  reflected  from  harder  surfaces,  might 
appear  more  shining,  that  sound  might  have  a  clearer 
ring,  and  the  kingdom  of  joy  extend  beyond  the  circle 
of  divine  glory.  Not  that  the  manifestation  of  God 
might  thereby  become  more  perfect,  for  it  is  at  the  cost 
of  a  diminution  of  harmony  that  any  particular  quality 
thus  becomes  more  vivid,  but  rather  that  it  was  ex- 
pedient for  infinite  power  and  love  to  realise  possibilities 


JACOB  BOEHME  215 

which,  though  they  had  no  place  in  the  divine  nature, 
still  showed  forth  the  signs  of  perfection. 

To  fulfil  their  destiny,  the  angels  must  proceed  from 
Father  to  Son,  from  wrath  to  love,  after  the  fashion  of 
God  himself.  Besides,  they  were  created  free,  and,  like 
God,  determine  themselves,  without  compulsion  from 
without.  They  are  masters  of  their  determinations. 
Now,  whereas  one  portion  of  the  angels  made  their  own 
freedom  of  will  conformable  with  the  divine  will,  another 
portion  rebelled  against  God.  Lucifer  was  the  chief  of 
these  rebel  angels  and  the  first  author  of  evil  :  he  sinned 
freely  in  accordance  with  his  own  will  and  without 
compulsion. 

Sin  came  about  in  the  following  manner.  A  com- 
pound of  nature  and  spirit,  Lucifer,  employing  his  own 
free  will,  fixed  his  imagination  on  nature.  Beneath  the 
gaze  of  this  magician,  nature  was  transformed;  from 
being  dark  she  became  shining  ;  full  of  defects,  she 
decked  herself  with  all  simple  perfections  ;  from  being 
a  part  she  became  so  puffed  up  as  to  appear  like  the  all. 
The  soul  of  the  angel  became  enamoured  of  this  idol, 
desiring  it  exclusively.  In  doing  so,  it  rejected  God 
and  separated  from  him. 

Then  hell  was  created.  Lucifer  obtained  what  he 
wished  for  :  separation.  This  result  he  obtained  not 
by  the  transcendent  intervention  of  God,  but  by  the 
immediate  effect  of  wrath  or  nature  to  whom  he  had 
devoted  himself.  Hell  is  the  principle  of  darkness, 
nature,  force,  life  pure  and  simple,  given  up  to  itself 
and  henceforth  contradictorily  opposed  to  love  and  light, 
and  so  deprived  of  all  direction,  control,  and  harmony. 
Hell  is  life  that  has  no  other  end  than  to  live.  Thanks 
to  Lucifer,  it  was  now  let  loose. 

Nor  was  this  all ;  Lucifer  was  created  eternal.    The 


2i 6  STUDIES  IN  PHILOSOPHY 

desire  for  life  and  the  desire  for  good,  which  God  had 
implanted  within  him,  had  not  as  their  common  support 
a  sensible  body  subject  to  succession  and  consequently 
capable  of  breaking  with  its  past  habits.  The  free  will 
of  a  mere  spirit  is  exhausted  in  a  single  act.  Lucifer's 
fault,  therefore,  is  irremediable.  No  conversion  is 
possible  for  him,  for  he  is  nothing  more  now  than  fire  and 
wrath,  and  light  has  no  longer  any  hold  upon  him.  The 
hell  he  has  created  is  as  eternal  as  his  own  will  itself. 

And  yet,  the  terrestrial  nature  ruled  by  the  angels 
suffers  from  the  effects  of  their  wrongdoing.  Con- 
fusion finds  its  way  into  this  nature.  Love,  being  exiled 
therefrom,  the  bond  uniting  the  forces  is  broken,  and  each 
of  these  latter  escapes  and  goes  wherever  it  pleases.  We 
no  longer  have  personal  unity,  in  which  the  parts  are 
the  organs  of  a  whole  ;  but  individual  multiplicity,  in 
which  each  part  regards  itself  as  the  whole  to  the 
exclusion  of  the  rest. 

Such  now  is  nature  :  the  earth  is  formless  and  void, 
darkness  covers  the  face  of  the  deep.  The  spirit  of 
God,  however,  hovers  above  his  shattered  work,  and 
the  Father  resolves  to  effect  a  new  creation  by  drawing 
nature  out  of  the  darkness  into  which  she  has  fallen. 
This  creation  is  the  one  related  by  Moses.  God  said, 
"  Let  there  be  light !  "  and  the  light  was  separated  from 
the  darkness.  In  seven  days,  in  accordance  with  the 
number  of  divine  spirits,  God  restored  nature  to  a 
state  of  harmony.  He  did  not,  purely  and  simply, 
destroy  Lucifer's  work  ;  he  gave  nature  a  weapon 
against  evil  and  an  instrument  of  regeneration,  to  wit, 
time.  Thanks  to  succession  in  time,  to  conceive  is  no 
longer  to  act ;  will  may  halt  at  the  very  brink  of  the 
precipice.  Even  when  accomplished,  the  act  no  longer 
exhausts  activity.  Henceforth,  the  good  are  neither 


JACOB  BOEHME  217 

fixed  in  good  nor  the  evil  in  evil.  To  time  is  attached 
space,  which  makes  individuals  relatively  independ- 
ent of  one  another.  Life  in  space  and  time  has 
for  its  object  sensible  matter,  i.e.  matter  properly  so 
called. 

The  term  and  perfection  of  creation  is  man,  the 
excellent  and  harmonious  concentration  of  the  three 
principles.  There  are  in  man  three  parts  :  soul  or  the 
infinite  power  of  good  and  evil ;  mind  or  intelligence 
and  sound  will ;  and  body,  or  concrete  reality.  The 
first  of  these  three  parts  corresponds  to  the  principle 
of  fire,  the  second  to  that  of  light,  and  the  third  to  that 
of  essence  or  reality.  The  three  principles  are  mani- 
fested in  man  with  all  the  perfection  that  existence  in 
time  and  space  implies. 

Man's  duty  is  to  subordinate  within  himself  two  of 
these  principles  to  the  third,  i.e.  will  and  action  to 
the  law  of  good,  and  his  end  is  to  generate  the  king 
of  nature,  whom  God  has  resolved  to  create  in  order 
to  dethrone  Lucifer.  As  God  the  Father  eternally  wills 
to  generate  his  heart  and  his  Son,  so  the  soul  ought  to 
fix  its  will  in  the  heart  of  God.  Adam  is  to  be  the 
seed  of  the  Christ.  The  task  that  has  fallen  to  man, 
however,  is  by  no  means  a  purely  spiritual  one.  The 
paradise  in  which  he  is  placed  and  which  he  must  cause 
to  blossom  forth  is  a  sensible  nature.  It  is  by  working 
to  draw  out  of  this  nature  all  the  treasures  she 
contains,  and  bring  them  to  light,  that  man  prepares 
for  the  coming  of  the  Son.  The  world,  developing  in 
time  and  space,  consists  of  individuals  separated  from 
one  another  :  these  individuals  have  to  be  united  in  one 
common  homage  paid  to  the  Eternal,  and,  without  their 
distinctive  characteristics  being  effaced,  these  latter  must 
be  raised  to  participation  in  absolute  personality. 


2i 8  STUDIES  IN  PHILOSOPHY 

This  is  the  destiny  prescribed  for  man,  though  not 
imposed  upon  him.  His  will  is  free.  In  him  there  is 
fire  and  light,  violence  and  gentleness,  egoism  and  self- 
denial.  In  addition,  as  the  result  of  his  terrestrial 
nature,  there  is  a  temporal  will,  set  between  these  two 
principles  and  capable  of  being  turned  in  the  direction 
of  the  one  or  the  other.  Man,  therefore,  possesses  all 
the  conditions  of  freedom,  and  is  able,  as  he  pleases, 
either  to  be  lost  himself  or  to  find  himself  effectively  by 
self-renunciation. 

How  has  he  used  this  power  ?  That  is  a  question 
of  fact,  it  finds  an  answer  in  tradition  and  experience. 
Now,  we  know  that  man,  following  the  example  of 
Lucifer,  disobeyed  God  and  fell  from  his  original  state 
of  nobility.  The  fall  of  man,  according  to  the  Mosaic 
account,  when  interpreted  in  the  light  of  the  spirit, 
was  brought  about  in  the  following  manner. 

Giving  reins  to  his  imagination,  man  began  to  con- 
template and  admire  nature,  in  preference  to  God.  By 
degrees,  he  attributed  to  his  idol  every  imaginable  per- 
fection, making  her  the  all,  including  even  divinity 
itself.  Then  he  grew  enamoured  of  her  and  ardently 
longed  to  engender  her  as  he  saw  her  in  his  imagination. 
Forgetful  of  the  rights  of  spirit,  he  wished  nature, 
untrammelled,  to  be  all  she  was  capable  of  being. 
Soon  afterwards,  in  accordance  with  the  law  of  being, 
the  idea  of  image  and  desire  became  a  body  ;  nature 
proclaimed  her  autonomy,  and  man  fell  beneath  the 
sway  of  the  violent,  egoistic  forces  he  had  let  loose. 
Such,  abridged,  is  the  story  of  the  fall.  The  sacred 
text,  however,  enables  us  to  distinguish  its  different 
phases  and  note  its  various  stages. 

The  starting-point  was  the  desire  to  know  things, 
no  longer  in  their  union  and  harmony,  as  God  had 


JACOB  BOEHME  219 

made  them,  but  by  separating  and  analysing  them, 
attributing  to  them  a  fictitious  individuality.  Man 
was  determined  to  know  what  hot  and  cold,  moist 
and  dry,  hard  and  soft,  and  all  the  other  qualities, 
taken  separately,  were  in  themselves.  In  death,  the 
congealer  and  disperser,  he  was  determined  to  dis- 
cover the  secret  of  life,  the  organiser.  No  longer  had 
that  divine  fruit,  concrete  knowledge,  any  savour  or 
attraction  for  him :  he  was  determined  to  taste  of 
abstract  knowledge,  parcelled  out,  the  fruit  of  terrestrial 
nature.  Nature,  thereupon,  responded  to  his  desire  by 
making  this  latter  objective  in  the  form  of  the  tree 
of  the  knowledge  of  good  and  evil.  This  tree  of 
temptation  is  none  other  than  the  sensible  realisation 
of  the  will  to  know  good  and  evil  separately,  in  so 
far  as  they  are  opposite  and  contradictory.  Through 
it,  man  sees  good  and  evil  as  two  things  exterior  to 
each  other,  according  to  the  condition  of  objects  set 
in  space  :  he  is  able  to  choose  the  latter  to  the  exclusion 
of  the  former.  The  fact  of  having  raised  up  the  tree 
of  analytical  science  is  the  first  sin,  that  of  under- 
standing. This  is  a  dangerous  declivity,  for  man  now 
conceives  evil,  and,  consequently,  is  capable  of  willing 
it;  still,  this  does  not  yet  constitute  the  fall,  since  he 
possesses  the  power  to  choose  between  good  and  evil. 

A  second  temptation  follows  the  first.  Hitherto, 
Adam  has  had  the  eternal  virgin  for  his  companion  ; 
hitherto,  the  ideal  or  the  image  of  God  has  been  the 
object  of  his  thought.  Having  begun  to  look  upon 
things  from  the  view-point  of  analysis,  in  their  ter- 
restrial form,  he  became  enamoured  of  the  world  of 
forces  and  instincts  which  henceforth  appeared  before  his 
gaze.  He  wished  to  live  an  animal  life,  to  reproduce 
himself  after  the  fashion  of  the  beasts.  The  image  of 


220  STUDIES  IN  PHILOSOPHY 

God  was  effaced,  the  virgin  fled  before  the  passion 
kindled  within  him.  Then  Adam  fell  asleep  :  for  the 
image  of  the  world  is  not  of  like  nature  to  the 
image  of  God.  The  latter,  which  slumbers  not,  con- 
stantly keeps  awake  the  spirit  contemplating  it.  But 
the  image  of  the  world,  being  subject  to  succession 
in  time,  tires  the  sight  and  engenders  sleep.  A  change 
of  condition  was  then  brought  about.  Man  had  fallen 
to  sleep  in  the  world  of  angels,  the  world  of  eternity  : 
he  awoke  in  time,  in  the  exterior  world.  Before  him 
he  saw  the  human  objectivation  of  his  earthly  desire  in 
the  form  of  a  woman  created  by  God  during  his  sleep. 
Aware  that  the  woman  came  from  himself,  man  sought 
to  unite  with  her,  to  unite  with  her  in  body.  This  is 
the  second  sin,  the  sin  of  sensibility.  Man  has  taken 
another  step  towards  perdition.  Still,  he  is  not  fallen, 
for  carnal  desires,  in  themselves,  do  not  deprive  man 
of  self-possession  ;  his  will  still  remains  his  own. 

The  fall,  that  neither  the  perversion  of  the  intelli- 
gence nor  that  of  sensibility  has  brought  about,  is  to 
be  effected  by  the  perversion  of  the  will.  The  devil 
breathed  into  man  the  desire  to  live  by  his  own  distinctive 
will,  to  suffice  unto  himself,  to  make  himself  God.  Man 
consented  to  the  temptation,  and,  by  disobedience,  set 
himself  over  against  God  as  his  equal.  From  that 
time  he  was  not  only  inclined  towards  evil,  he  plunged 
therein.  He  became  what  he  had  willed  to  be,  though 
in  a  way  contrary  to  what  he  had  imagined.  He 
became  god,  not  the  god  of  love,  light,  and  life,  the 
only  true  God,  but  the  god  of  wrath,  darkness,  and 
death,  who  is  nothing  more  than  the  sacrilegious  and 
diabolical  personification  of  the  mysterious  substratum 
of  divinity. 

Thereupon,  man  was  cursed  ;  or  rather,  he  declared 


JACOB  BOEHME  221 

himself  to  be  the  child  of  the  devil.  His  will,  evil  in 
itself,  separated  him  from  God,  and  dedicated  him  to 
wrath.  Following  on  this  curse,  the  world,  of  which 
man  was  both  the  resume  and  the  mover,  passed  from 
a  condition  of  harmony  to  one  of  individual  dispersion. 
Each  human  being  claimed  to  live  in  the  world  for 
himself  alone,  and  to  effect  his  own  development  with- 
out any  thought  for  his  neighbours.  The  struggle  for 
life  became  the  world's  only  law. 

Still,  man  was  not  condemned  by  God  for  all 
eternity  as  Lucifer  had  been  ;  the  conditions  of  the 
fall  were  different.  The  devil,  of  himself  alone,  was 
the  entire  cause  of  the  sin  he  had  committed.  Before 
him,  indeed,  evil  was  non-existent,  there  was  only  the 
possibility  of  evil.  Of  this  possibility,  Lucifer  had 
formed  evil  with  all  that  it  comprises,  its  matter  as 
well  as  its  form  :  he  was  the  author  of  the  motives 
that  had  tempted  him,  as  well  as  of  the  determination 
he  had  arrived  at  in  accordance  with  these  motives. 
The  position  of  man  was  quite  different.  Before  him, 
evil  was  already  in  existence  as  a  given  reality,  and, 
along  with  evil,  a  downward  tendency  to  new  falls. 
It  was  at  Satan's  solicitation  that  man  sinned.  Though 
the  decision  he  came  to  was  his  own,  the  motives  of 
this  decision  were  not  at  work.  They  were  within  him 
as  instincts,  a  pre-existing  nature.  Man  is  thus  respon- 
sible for  his  own  determination  alone,  not  for  the 
motives  to  which  he  has  yielded.  This  is  the  reason 
why  the  fall  of  Adam,  which  indeed  would  be  a  mortal 
one  were  man  left  to  himself,  is  not  irremediable.  It 
is  possible,  if  not  for  justice,  at  all  events  for  divine 
mercy,  to  set  the  tendency  towards  good,  deep  in  the 
human  soul,  in  opposition  to  evil  solicitations,  and  to 
give  man's  will,  which  is  temporal  in  its  nature,  the 


222  STUDIES  IN  PHILOSOPHY 

power  to  retract  its  resolution.  Will  God  now  come 
to  the  aid  of  man,  who  has  rebelled  against  him  ? 
Will  he  send  man  a  redeemer  and  saviour  ?  This  is 
what  no  necessity  either  commands  or  excludes,  it  is 
something  to  be  decided  in  the  mysterious  depths  of 
infinite  will. 


God,  having  already  restored  harmony  to  the  world, 
harmony  that  had  been  disturbed  by  Lucifer,  resolved 
to  summon  man  to  regeneration.  Good  and  evil  were 
now  in  the  presence  of  each  other,  not  only  in  eternity 
but  also  in  time  :  God  decided  to  bring  about,  as  far  as 
possible,  the  reconciliation  of  these  two  principles.  In 
accordance  with  the  divine  decrees  anterior  to  the  fall 
of  man,  the  Son  was  some  day  to  be  born  in  human 
form,  so  that  the  word  might  be  manifested  in  time. 
As  man  was  given  up  to  wrath  and  the  devil,  God 
decreed  that  the  coming  of  the  Christ  should  be  not 
only  the  coming  of  one  who  would  compass  human 
perfection,  but  also  that  of  a  redeemer  and  saviour. 
He  prepared  for  this  coming  by  the  series  of  events 
related  in  the  Old  Testament,  and  finally  gave  up  his 
Son  to  the  world  to  be  crowned  with  thorns  and 
crucified.  Per  crucem  ad  lucem !  The  Christ  is  a 
human  creature,  and  he  is  the  Son  of  the  eternal  Virgin. 
In  him  death  is  overcome.  He  who  suffers  with  him 
is  also  glorified  with  him. 

Still,  we  must  examine  more  closely  and  see  how 
man's  salvation  is  realised  by  Jesus  Christ. 

When  the  reason  hears  mention  of  God,  of  his 
nature  and  will,  it  imagines  that  God  is  something 
foreign  and  far  away,  living  outside  this  world  and 
above  the  stars,  ordaining  things  mechanically  after  the 


JACOB  BOEHME  223 

manner  of  a  force  situated  in  space.  Hence  reason, 
assimilating  God  to  his  creatures,  attributes  to  him  a 
mode  of  thought  and  action  analogous  to  that  of  man. 
It  believes  that  God,  before  creation,  deliberated  within 
himself  as  to  the  place  he  should  assign  to  each  creature. 
It  also  implies  that  God  decided  to  summon  a  portion 
of  mankind  to  heavenly  joy,  in  order  to  manifest  his 
grace,  and  condemned  the  rest  to  damnation,  in  order 
to  manifest  his  wrath.  Thus,  God  would  appear  to 
have  made  a  difference,  for  all  eternity,  between  men, 
for  the  purpose  of  manifesting  his  power  in  the  direction 
both  of  wrath  and  of  love. 

Most  certainly  there  is  an  election  of  grace,  though 
it  could  not  come  about  in  the  way  reason  imagines. 
Were  God  to  deliberate  and  come  to  a  decision  as  we 
do,  were  he  to  govern  things  from  without,  he  would 
be  divided  against  himself,  he  would  be  changeable, 
not  eternal.  Besides,  how  could  God  will  to  condemn 
a  portion  of  his  creatures  ?  God  is  love  ;  he  wills  the 
good  of  all  beings.  Election  and  damnation  are  not 
the  act  of  a  will  exterior  to  man.  Man  is  free, 
absolutely  free  ;  for  the  root  of  his  being  is  plunged 
in  the  eternal,  infinite  substratum  of  things.  There 
is  nothing  behind  the  human  will  capable  of  constraining 
it.  Itself  is  the  first  beginning  of  its  own  actions. 
Election  or  damnation  is  the  result  of  this  very  freedom. 
By  it,  man  can  turn,  as  he  pleases,  towards  light  or 
towards  darkness,  towards  love  or  towards  egoism  : 
man  can  make  either  an  angel  or  a  devil  of  himself. 
Within  himself  he  bears  his  own  paradise  and  hell  :  the 
exterior  paradise  and  hell  are  nothing  but  symbols  of 
good  and  evil  will.  Not  that  man  is  sufficient  unto 
himself  and  can  do  without  divine  grace.  His  good 
will  is  but  a  prayer,  unavailing  without  the  help  of 


224  STUDIES  IN  PHILOSOPHY 

God  ;  God  has  foreseen  from  all  eternity  that  he  either 
would  or  would  not  offer  up  this  prayer.  Free  actions, 
however,  remain  free  in  divine  foreknowledge,  which, 
sunk  in  the  primordial  deep,  cannot  be  distinguished 
from  the  common  substratum  of  all  wills. 

The  first  sign  and  the  first  effect  of  election  is  faith. 
Like  election,  faith  is  often  misunderstood.  Every  one 
boasts  of  having  faith.  Where  is  it  in  reality  ?  Present- 
day  faith  is  nothing  but  a  story  learnt  by  heart.  Where 
is  the  man  with  a  child-like  faith  in  the  birth  of  Jesus  ? 
Did  he  really  believe  it,  he  would  draw  nigh  to  the  Infant 
Jesus  whom  he  would  welcome  and  tenderly  nurture 
within  himself.  No  :  he  is  acquainted  only  with  the 
historic  child,  deceiving  his  conscience  with  vain  erudi- 
tion. Never  has  there  been  so  much  talk  of  faith,  and 
never  was  real  faith  more  lacking.  Would  you  have  a 
proof  of  this  ?  Never  before  has  there  been  so  much  dis- 
puting, so  much  judging  and  condemning  of  one  another. 
Does  God  judge  and  condemn  the  birds  of  the  forests 
because  each  of  them  praises  him  in  his  own  way,  and 
in  a  different  tone  from  the  rest  ?  Does  not  the 
infinite  might  of  God  admit  of  an  infinite  variety  of 
expressions  of  homage  ?  You,  who  persecute  your 
brothers,  are  more  useless  than  the  flowers  of  the  fields, 
more  foolish  than  beasts  lacking  in  intelligence.  You 
are  the  birds  of  prey  that  affright  the  other  birds, 
preventing  them  from  chanting  the  praises  of  God. 
To  believe  in  Jesus  Christ  from  an  historical  point  of 
view  is  no  more  helpful  than  believing  a  fable.  How 
many  Jews  and  Turks  are  more  Christian  than  those 
sham  Christians  who  know  what  Jesus  did  and  yet  do 
what  the  devil  does  !  But,  the  answer  will  come, 
we  believe  in  the  word.  Then  we  must  try  to  under- 
stand what  the  true  word  is.  The  Scriptures  are 


JACOB  BOEHME  225 

helpful,  but  they  are  not  the  word,  they  are  only  its 
mute,  obliterated  signs.  The  word  is  living,  for  it  is 
the  vehicle  of  the  spirit.  No  formula  can  define  it, 
for  it  is  infinite  as  God  himself.  That  is  why  true 
faith  is,  in  fine,  a  righteous  will,  freely  subject  to  the 
law  of  the  spirit.  It  consists  in  renewing  within 
oneself  the  birth  and  life  of  Christ,  his  baptism  and 
temptations,  sufferings  and  death.  The  imitation  of 
Christ  is  the  distinctive  mark  of  the  children  of  God. 
Consequently,  the  true  Christian  is  of  no  sect ;  he  may 
live  in  one,  but  he  does  not  belong  to  it.  His  religion 
is  interior,  it  cannot  be  confined  within  any  form. 

Faith,  when  thus  understood,  is  the  beginning  of 
regeneration.  What  is  to  be  thought  of  the  exterior 
means  and  methods  that  the  Churches  add  on  to  it? 
Speaking  generally,  works  are  nothing  in  themselves, 
and  the  Roman  Catholic  Church,  which  attributes  value 
thereto,  is  the  Babel  of  the  Christian  world.  Erroneous 
also  is  it  to  believe  that  faith  saves  us  because  through 
it  the  merits  of  Christ  would  be  attributed  to  us  from 
without,  just  as  a  new  form  may  be  given  to  passive 
matter.  Such  an  operation  would  not  change  the  root 
nature  of  the  soul,  it  would  not  be  a  second  birth. 
Faith  could  not  save  us  by  some  theurgic  operation 
that  compelled  divine  justice  to  benefit  us  :  it  saves  us 
only  by  the  sanctifying  grace  it  bears  within  itself,  and 
which,  from  without,  engenders  within  us  both  penitence 
and  the  redeeming  Christ.  Justification  is  sanctifica- 
tion.  It  is  not  the  object  of  faith  that  regenerates  us, 
it  is  faith  itself. 

For  this  reason  no  particular  means  of  regeneration 
is  efficacious  if  faith  be  not  the  soul  thereof.  True 
prayer  is  not  a  passive  request  for  divine  assistance,  it 
is  the  humble  action  of  the  will  that  recognises  its 

Q 


226  STUDIES  IN  PHILOSOPHY 

need  and  goes  to  God  as  for  food  ;  it  is  the  soul 
beseeching  and  receiving  sanctifying  grace.  True 
preaching  is  not  the  teaching  specially  given  by  the 
priest  or  even  by  the  Bible.  The  faithful  who  see  and 
hear  with  the  spirit  learn  from  all  creatures.  Sacraments 
are  not  aids  granted  to  man  without  himself  contributing 
thereto.  The  true  sacrament  consists  of  divine  grace 
descending  upon  the  soul,  which  can  appropriate  it  only 
by  faith.  And  regeneration,  the  object  of  prayer,  of 
sermons  and  sacraments  alike,  is  not  a  new  nature 
grafted  on  to  the  old :  it  is  the  spirit,  awakening  and 
expanding,  deep  within  the  nature  ;  it  is  the  person 
creating  himself  by  renunciation  of  the  individual  I, 
the  interior  man  who  is  substituted  for  the  exterior  man. 
Now,  of  what  nature  is  the  life  of  the  regenerate 
man  ?  Is  it  only  apathy  and  indifference,  mere  reflec- 
tion of  the  spirit  upon  itself,  annihilation  deep  in 
primordial  no-thing  ?  Spirit,  we  know,  is  not  this  inert 
no-thing  at  the  conception  of  which,  by  suppressing 
differences,  human  logic  arrives.  All  interior  being  tends 
to  become  exterior,  all  infinitude  is  the  desire  to  take 
form,  all  mystery  is  an  effort  to  reveal  itself,  all  spirit 
is  the  will  to  become  a  body.  So  also  is  it  with  the 
Christian  virtues.  They  do  not  remain  abstractions  ; 
they  develop  and  become  manifest.  They  become 
manifest  by  complete  renunciation  of  self,  by  total 
abandonment  to  the  will  of  God,  by  meekness,  by 
human  love,  by  communion  of  souls  in  spite  of  all 
outer  differences,  by  mastery  over  nature,  i.e.  over 
earthly  desires,  and  by  joy,  that  foretaste  of  eternity. 
The  new  man  does  not  destroy  the  old,  the  outer  man, 
though  he  takes  care  not  to  forget  himself  therein. 
Thou  art  in  the  world,  Christian  !  Thou  art  engaged 
in  an  honourable  trade.  Remain  in  it,  work,  act,  earn 


JACOB  BOEHME  227 

the  money  thou  need'st,  make  the  elements  produce 
all  they  are  capable  of  producing,  dig  in  the  ground  for 
silver  and  gold,  make  them  into  works  of  art,  build 
and  plant.  All  this  is  well  and  good.  Listen,  however, 
to  the  A  B  C  of  wisdom  :  Put  not  thy  soul  into  this 
exterior  life.  Chain  not  thy  free  spirit  down  in  this 
prison.  If  thou  retainest  thy  freedom,  all  that  thou 
do'st  in  the  world  will  prosper.  For  everything  sings 
forth  the  praises  of  God  to  him  who  has  ears  to  hear. 
Even  the  backslidings  of  thy  earthly  companion  shall 
not  harm  thy  soul,  but  they  will  be  beneficial  to  him. 
A  single  action  is  not  a  habit ;  a  powerful  tree  stands 
erect  before  the  raging  storm.  When  thou  see'st  the 
exterior  man  offend,  thou  wilt  the  better  understand 
the  frailty  of  nature,  the  greatness  and  might  of  divine 
mercy.  Let  not  man,  however,  imagine  that  in  his 
life  on  earth  he  can  ever  dispense  with  prayer  and  effort. 
Man  is  and  remains  free ;  consequently,  he  is  never 
established  in  good.  Time  cannot  hold  eternity.  How- 
ever strong  be  our  link  with  God,  we  remain  in  the 
devil's  power.  Resistance  to  evil  is  our  condition 
in  this  world,  right  on  to  the  end.  If  we  grow 
remiss,  nature  once  more  lays  hold  upon  us  :  the  form 
in  which  the  spirit  is  manifested  binds  and  imprisons 
this  latter  as  soon  as  it  ceases  to  act.  Each  moment 
we  must  correct  ourselves,  revive  our  new  birth,  create 
God  anew  within  ourselves.  Only  when  life  comes  to 
an  end  does  the  tree  of  faith,  hope  and  love,  nurtured 
by  our  own  unremitting  efforts,  stand  erect  and  in- 
capable of  being  uprooted. 

And  so,  in  the  world  of  time,  there  is  being  prepared 
the  rapprochement  of  the  good  and  the  evil  principle, 
and  the  conscious,  definitive  reconstitution  of  primor- 
dial unity.  All  end  has  a  tendency  to  rejoin  its  beginning, 


228  STUDIES  IN  PHILOSOPHY 

though  on  a  higher  plane,  ascending  right  to  the  fixed 
point  on  which  this  beginning  depends.  As  long  as  man  is 
a  terrestrial  body,  he  can  and  ought  to  choose.  Along 
with  his  temporal  nature,  however,  disappears  the  con- 
tingency of  his  actions.  Death  introduces  him  to  eternity. 
The  fruit  of  his  free  determinations  is  now  ripe  :  he 
detaches  himself ;  that  which  he  is,  he  is  once  for  all. 
Man,  then,  according  to  the  nature  he  has  created 
within  himself,  henceforward  belongs  either  to  God  or 
to  the  devil.  His  free  will  has  become  changed  either 
into  freedom  and  love  or  into  caprice  and  violence. 

And  so  the  final  end  of  things  is  the  definitive 
dualism  of  good  and  evil,  so  far  as  they  are  the  products 
of  a  will  that  is  free.  In  the  beginning,  God  engendered 
good  and  evil  considered  as  possibilities,  i.e.  he  created 
the  conditions  and  materials  of  good  and  evil  actions. 
From  the  way  in  which  free  beings  acted,  there  resulted, 
in  fact,  the  realisation  of  the  two  possibilities  God  had 
formed.  On  both  sides,  being  has  passed  through 
three  phases  :  possibility,  the  contingent  fact,  definitive 
determination.  It  was  by  thwarting  conscious  will  that 
idea  became  thing  ;  and  possibility,  necessary.  The 
kingdom  of  God  is  the  harmony,  henceforth  inde- 
structible, between  spirit  and  nature.  Individuals  subsist 
therein,  and  continue  to  be  distinguished  from  one 
another,  otherwise  there  would  be  no  more  nature  ;  but 
they  live  without  strife,  each  according  to  his  character  : 
they  subsist  by  love  alone  and  have  nothing  to  do  with 
hatred.  They  have  attained  true  unity  which  is  not 
an  exterior  rapprochement  practised  with  a  view  to  the 
satisfaction  of  egoistic  interests,  but  rather  the  common 
participation  of  individual  souls  in  divine  personality. 
In  the  kingdom  of  the  devil,  on  the  other  hand,  the 
will  to  live  has  definitively  thrown  off  all  law  and 


JACOB  BOEHME  229 

direction.  It  has  what  it  willed :  life  as  the  sole  end 
of  life.  Henceforward,  there  is  no  harmony,  goodness 
or  love.  Ego'fsm  and  anarchy  reign  without  a  rival. 
The  individual  is  his  own  master  ;  and  this  sovereignty, 
which  rests  on  rebellion,  not  on  obedience,  is  the  endless 
struggle,  infinite  torment. 


VI 

Boehme's  doctrine  concludes  with  an  exposition  of 
the  final  ends  of  all  things.  This  doctrine  presents 
itself  to  us  as  the  metaphysical  history  of  Being,  apper- 
ceived  by  intuition  deep  within  its  physical  history. 
Starting  from  the  eternal,  we  have  come  back,  through 
time,  to  the  eternal.  The  circle  is  closed  again  : 
revelation  is  accomplished. 

Now,  what  is  this  doctrine  which  is  called  by  its 
author  Aurora  or  the  Morning  Redness,  the  explanation 
of  the  celestial  and  terrestrial  mystery,  the  setting  forth 
of  the  genesis  of  God  and  of  all  things,  and,  speaking 
generally,  Christianity  interpreted  after  the  spirit  ? 

There  can  be  no  doubt  but  that  it  is,  first  of  all,  a 
religious  doctrine,  and  it  is  only  natural  that  Boehme's 
disciples  should  mainly  be  found  amongst  theologians. 
But  would  it  be  legitimate  to  abide  by  the  letter  of  the 
doctrine  in  judging  one  who  ever  affirmed  that  truth 
is  in  the  spirit,  not  in  the  letter,  and  that  it  is  the 
characteristic  of  the  spirit  to  be  for  ever  impossible  of 
expression?  Evidently,  by  this  theory  alone,  Boehme 
relegates  to  a  second  place,  religion  properly  so  called, 
religion  that  is  inconceivable  without  some  given  reve- 
lation, some  positive  fact,  and  puts  in  the  first  place 
philosophy,  or  rather,  religion  so  far  as  it  is  allied  with 
philosophy.  Indeed,  whoever  reads  Boehme's  works 


230  STUDIES  IN  PHILOSOPHY 

in  the  way  he  himself  recommends  us  to  read  them, 
trying  to  discover  the  spiritual  meaning  in  sensible  and 
intellectual  images,  finds  that  doctrines  of  a  philosophical 
character  appear  at  each  step  beneath  his  religious 
outpourings. 

The  theological  mysteries  of  the  Trinity,  the  Fall 
and  the  Redemption  are,  of  a  certainty,  the  promptings 
that  cause  him  to  reflect.  But  beneath  these  mysteries 
he  sees  the  problem  of  the  reconciliation  of  evil  and  the 
finite,  as  positive  realities,  with  infinite  personality  as 
the  first  and  only  source  of  being.  And  the  way  in 
which  he  solves  this  problem  is  certainly  metaphysics 
under  the  cloak  of  theology.  From  the  finite  and  evil, 
to  whose  existence  our  senses  testify,  the  suprasensible 
conditions  of  finite  nature  and  evil  action  are  distin- 
guished, and  these  conditions  are  deduced  from  the 
divine  will,  in  so  far  as  that  will  wills  to  be  manifested 
and  posited  as  a  person.  No  manifestation  without 
opposition.  And  so  God  posits  his  contrary  in  order 
to  lay  hold  upon  himself,  by  distinguishing  himself  from 
this  contrary  and  imposing  on  it  his  law.  This  contrary, 
or  eternal  nature  bound  to  the  very  existence  of  God, 
without  itself  being  the  finite  and  evil,  is  the  foundation 
of  their  reality.  The  finite  is  the  dissemination — freely 
effected  by  God,  by  means  of  time — of  the  essences 
contained  in  the  divine  nature.  Evil  is  nature,  which 
is  only  a  part,  posited  as  the  all  by  the  untrammelled 
will  of  created  beings.  The  finite  and  evil,  after  all,  as 
regards  their  matter,  are  deduced  from  the  conditions 
of  existence  of  the  personality,  whereas  in  their  sensible 
form  and  realisation  they  result  from  the  free  initiative 
of  the  will.  Consequently  the  world  is  something  quite 
different  from  mere  non-being  or  the  unstable  effect 
of  an  act  of  arbitrary  will :  it  possesses  reality,  a  true, 


JACOB  BOEHME  231 

internal  existence  :  though  founded  on  God  it  is  not 
God  :  it  is  based  upon  the  very  nature  God  needs  in 
order  to  become  manifest. 

It  cannot  be  denied  that  in  these  ideas,  clearly  ex- 
pressed by  Boehme  in  all  his  metaphors,  are  the  germs 
of  a  philosophic  system.  But  what  is  the  value  and 
signification  of  this  system  ?  Is  it  not  an  isolated  work, 
without  any  important  relation  to  the  general  history  of 
philosophy  ? 

It  must  be  confessed  that — with  the  exception  of 
Louis  Claude  de  Saint-Martin  (the  "  Unknown  Philo- 
sopher"), Baader  the  Catholic  theologian,  and  Schelling 
in  the  final  phase  of  his  philosophy — the  philosophers 
by  profession,  after  reading  and  forming  an  opinion  on 
Boehme,  are  rather  inclined  to  bestow  on  him  vague 
encomium  than  to  attempt  to  assimilate  his  doctrines. 
Saint-Martin's  ideas  have  scarcely  been  mentioned  in 
France  except  by  historians  ;  and  the  Germans  have 
developed  more  especially  the  intellectualist  philosophy 
born  of  Leibnitz,  Kant  and  Spinoza,  which  rejects  the 
absolute  reality  of  nature  and  the  freedom  of  the  will, 
those  essential  elements  in  Boehme's  system. 

On  this  point,  nevertheless,  we  must  guard  against 
judging  by  appearances  or  details.  Two  traits,  in  a 
word,  mainly  characterise  the  speculations  of  our  theo- 
sophist :  spiritualism,  posited  as  a  fundamental  truth  ; 
and  realism,  admitted  on  the  faith  of  experience  and 
connected  by  way  of  deduction  with  the  spiritualist 
principle.  On  the  one  hand,  Boehme  holds  that  spirit 
alone  is  the  first  and  true  being  :  spirit,  i.e.  infinite 
freedom,  that  creates  for  itself  objects  and  forms,  and 
remains  infinitely  superior  to  all  its  creations,  imper- 
ceptible being  that  is  everywhere  in  action  and  itself 
incapable  of  being  realised  and  becoming  an  object  of 


232  STUDIES  IN  PHILOSOPHY 

experience  ;  the  perfect  person,  in  word,  living  and 
truly  metaphysical  existence,  of  which  all  given,  deter- 
minate existence  can  be  nothing  but  an  imperfect  mani- 
festation. But,  on  the  other  hand,  Boehme  is  a  realist. 
He  does  not  admit  that  the  multiple  and  the  diverse 
may  be  a  vain  image  of  the  imagination,  or  the  purely 
phenomenal  effect  of  a  transcendent  cause  ;  he  does  not 
acknowledge  that  evil  may  only  be  a  lesser  good. 
Nature  has  her  own  principle  of  existence,  contrary  to 
that  of  spiritual  existence.  Evil  is  a  living  force  that 
tends  to  destroy  good.  To  posit  spiritualism  as  a  thesis 
and  realism  as  an  antithesis,  and,  in  a  synthesis,  to 
reconcile  the  reality  of  the  objects  of  experience  with 
the  supremacy  of  spirit :  such  is  Boehme's  task. 

Such,  too,  in  fine,  is  the  ground  of  the  principal 
German  systems.  With  Leibnitz,  Kant,  Fichte,  Schelling 
and  Hegel  it  is  spirit  that  is  being,  and  spirit  is  the 
living  infinite  that  no  form  can  contain.  For  all  these 
philosophers,  however,  the  world  has  a  reality  of  its 
own,  a  reality  that  is  a  stumbling-block  to  spirit  and 
yet  must  be  deduced  from  the  nature  of  spirit.  It  is 
in  this  antinomy  of  spirit  as  principle,  and  matter  as 
reality,  that  German  philosophy  flounders  ;  and  monad- 
ology,  transcendental  idealism,  the  philosophy  of  the 
absolute,  and  absolute  idealism  are  only  different  solu- 
tions of  one  and  the  same  problem.  Nor  is  this  all. 
Idealism,  realism,  and  the  search  after  the  reconciliation 
of  the  latter  with  the  former,  are  traits  of  German 
philosophy  that  are,  it  would  seem,  to  be  met  with  in 
the  nation  itself;  so,  at  all  events,  historians  have 
observed.  Thus,  whatever  may  have  been  the  exterior 
link  between  the  German  philosophers  and  Jacob  Boehme, 
they  are  united  to  him  by  a  stronger  and  closer  bond 
than  mere  influence,  they  are  his  brothers,  at  least,  if 


JACOB  BOEHME  233 

not  his  sons,  children  of  one  and  the  same  genius, 
expressions  of  one  and  the  same  aspect  of  the  human 
mind.  Was  he,  then,  a  false  prophet,  who,  in  1620, 
after  reading  the  Psychologia  vera  of  Jacob  Boehme, 
greeted  its  author  by  the  unexpected  name  of  "Philoso- 
phus  teutonicus  "  ? J 

1  "He  is  known,"  says  Hegel,  "as  the  Philosophus  Teutonicus,  and 
in  reality  through  him  for  the  first  time  did  philosophy  in  Germany  come 
forward  with  a  characteristic  stamp.  The  kernel  of  his  philosophizing  is 
purely  German"  (Gesch.  Ph.  iii.  1836,  p.  300)  (Translator's  note). 


DESCARTES 

REGARDING  things  only  from  the  historical  standpoint, 
Cartesianism  dominates  the  entire  development  of  modern 
philosophy.  Amongst  others,  the  German  savants, 
intent  on  finding  the  internal  principles  of  historical 
developments,  took  delight  in  discovering,  in  Cartesian 
problems,  the  starting-point  of  all  the  great  questions 
that  have  stirred  the  minds  of  modern  philosophers. 
More  particularly,  they  saw,  in  the  Cogito,  the  living 
germ  from  which,  by  immanent  dialectic,  all  the  great 
systems  that  have  so  far  appeared  were  to  blossom 
forth.  Thus,  Kuno  Fischer  distinctly  regarded  Cartes- 
ianism, and  the  antinomies  into  which  it  enters  as  it 
develops,  as  the  origin  or  necessary  condition  of  the 
occasionalism  of  Malebranche,  the  monism  of  Spinoza, 
the  monadology  of  Leibnitz,  the  sensualism  of  Locke, 
the  materialism  of  La  Mettrie,  the  idealism  of  Berkeley 
and  the  criticism  of  Kant.  In  most  of  the  German 
historians  of  philosophy  similar  deductions  may  be 
found. 

Speaking  generally,  it  may  be  said  that  the  central 
problem  of  Cartesian  metaphysics  was  the  transition 
from  thought  to  existence.  Thought  alone  is  indis- 
solubly  inherent  in  itself:  how,  then,  by  what  right 
and  in  what  way,  can  we,  in  our  judgments,  affirm 
existences?  There  is  one  case,  and  only  one,  wherein 
existence  is  immediately  connected. with  thought  in  the 

234 


DESCARTES  235 

intuition  of  the  understanding  :  and  that  is  when  we 
say  :  "  Cogito,  ergo  sum."  How  and  in  what  way  can 
we  extend  to  other  existences  the  certainty  we  directly 
attribute  to  that  of  thought  ?  This  is  the  knotty  point 
in  the  Cartesian  philosophy.  Now,  this  problem  of 
existence  controlled  the  investigations  of  Locke,  Hume, 
Reid  and  Kant,  as  it  did  those  of  Malebranche,  Spinoza 
and  Leibnitz.  Existence,  which,  to  the  ancients,  was 
a  thing  given  and  immediately  apprehensible,  and  that 
had  only  to  be  analysed,  is  here  something  far  away, 
which  has  to  be  attained  to,  if  that  be  possible.  There 
we  find  the  distinctive  characteristic  of  modern  as 
compared  with  ancient  philosophy,  and  this  characteristic 
is  the  mark  of  Cartesianism  itself. 

Not  only  does  Cartesianism  thus  control  the  progress 
of  modern  philosophy,  it  is  also  of  considerable  im- 
portance in  the  general  history  of  the  human  mind. 
Doubtless,  our  seventeenth  century  in  France  largely 
drew  upon  Christian  and  classic  sources,  but  science 
developed  alongside  of  literature  ;  and  science,  in  those 
days,  was  the  Cartesian  conception  of  the  world  :  it  was 
the  control  of  the  mathematical  mechanism  over  all  that 
was  not  thought  strictly  so  called,  the  condition  of  this 
very  mechanism.  As  Huyghens  wrote  on  the  occasion 
of  the  death  of  Descartes  : 

Nature,  prends  le  deuil,  et  pleure  la  premiere 
Le  grand  Descartes  !   .  .  . 
Quand  il  perdit  le  jour,  tu  perdis  la  lumiere  : 
Ce  n'cst  qu'a  sa  clarte  que  nous  t'avons  su  voir. 

And  when  Newton  reformed  Cartesianism,  did  he  not 
do  so  by  adopting  this  very  basis  of  natural  philosophy, 
treated  mathematically,  which  Descartes  had  discovered 
and  assured  ? 

Nor  is  this  all  :   as  Descartes  is  a  dualist  and  looks 


236  STUDIES  IN  PHILOSOPHY 

upon  all  blending  of  philosophy  with  religion,  corporeal 
with  spiritual  philosophy,  as  spurious,  so  too  the 
seventeenth  century  is  simultaneously  religious  and 
rationalistic,  partaking  both  of  the  moralist  and  of  the 
scientist,  without  these  various  disciplines  interpenetrat- 
ing or  being  weakened  by  one  another.  Pascal  the 
mystic  does  no  harm  to  Pascal  the  physicist,  and  vice 
versa. 

^  In  a  word,  Descartes  regards  thought  as  without  an 
equal ;  he  sees  in  it  alone  the  principle  of  certainty. 
The  seventeenth  century,  likewise,  considers  that  in 
thought  lies  human  dignity,  that  by  it,  and  not  by 
material  greatness,  can  we  rise  to  our  true  stature.  The 
conviction  of  the  power  of  reason  creeps  into  the  minds 
of  men  to  such  a  degree  that  the  obstacles,  both 
provisional  and  even  definitive,  which  Descartes  had  set 
up,  are  speedily  overthrown.  Social  and  political  ques- 
tions which  could  not,  for  a  long  time,  in  his  opinion, 
be  accessible  to  science  ;  religious  questions  which  went 
altogether  beyond  it,  were  submitted  to  the  examination 
of  reason.  The  eighteenth  century  dedicated  itself  to 
this  work ;  it  has  even  been  said  that  the  French 
Revolution  had  its  origin  in  the  Discours  de  la  Methode. 
A  false  statement,  if  it  means  that  Cartesianism  contained 
such  a  consequence  ;  and  yet  an  assertion  capable  of 
being  upheld,  if  the  statement  is  taken  as  signifying 
that  it  was  in  the  name  of  the  Cartesian  principle  of 
rational  evidence  that  society  was  revived  in  1789. 

And  so  we  see  that  Cartesianism  is  an  essential 
element  in  the  philosophical  and  moral  history  of  modern 
times.  But  does  it  belong  only  to  history  ?  Has  it  no 
longer  anything  to  teach  us  ? 

Huxley,  the  English  philosopher  and  scientist, 
affirmed  that  Descartes'  system,  far  from  being  a 


DESCARTES  237 

subject  of  scholarly  curiosity,  was  the  very  soul  both  of 
contemporary  philosophy  and  science.  Our  philosophy 
is  idealistic,  and  it  is  the  Cogito  of  Descartes  that  is  the 
principle  of  this  idealism.  Our  science  is  mechanistic, 
and  it  is  the  Cartesian  reduction  to  extent  of  all  that  is 
not  spirit,  which  has  founded  this  mechanism. 

Independently  of  these  general  tendencies,  many 
questions  more  especially  connected  with  contemporary 
speculation  have  been  bequeathed  to  us  by  Descartes' 
philosophy. 

Such,  in  metaphysics,  are  the  problem  of  existence, 
that  of  the  relations  between  will  and  understanding, 
that  of  certainty,  that  of  the  relations  between  science 
and  metaphysics,  and  that  of  the  relations  between 
spirit  and  matter.  The  philosophy  of  science  is  specially 
concerned  nowadays  with  the  question  of  the  relation 
between  mathematics  and  experience.  How  and  in 
what  sense  can  that  which  is  proved  by  demonstration 
agree  with  that  which  is  known  by  perception  ?  How 
comes  it  to  pass  that  physics  can  be  treated  mathe- 
matically ?  Now,  this  is  the  very  question  Descartes 
first  asked  himself,  and  he  may  be  said  to  have  con- 
structed his  system  of  metaphysics  for  the  purpose  of 
answering  it. 

As  regards  science,  the  alliance  between  geometry 
and  analysis,  the  mechanical  interpretation  of  phenomena, 
the  exclusion  of  final  causes,  mathematical  mechanism 
applied  not  only  to  the  systematisation  of  phenomena 
but  also  to  the  explanation  of  the  genesis  of  the  world  ; 
not  only  to  the  study  of  inorganic  bodies  but  also  to 
that  of  life  itself,  are  all  to  be  found,  as  so  many  essential 
elements,  in  the  Cartesian  philosophy.  It  is  also  the 
Cartesian  spirit  that  has  brought  into  existence  certain 
special  modern  sciences,  such  as  experimental  psychology 


23 8  STUDIES  IN  PHILOSOPHY 

and  positive  sociology,  which  attempt  to  examine 
psychical  or  social  facts  in  their  elements  or  mathematic- 
ally measurable  equivalents. 

Moreover,  let  it  not  be  said  that,  in  order  to  possess 
these  leading  ideas,  it  is  enough  to  receive  them  from 
present-day  savants  in  the  form  they  have  assumed  as 
the  result  of  two  centuries  of  discussion.  It  is  not  the 
same  with  ideas  as  with  facts,  the  knowledge  of  which 
almost  inevitably  becomes  more  and  more  perfect. 
What  advantage  is  it  for  a  man  to  acquire  a  rough 
measurement  of  some  phenomenon  when  he  can 
become  acquainted  with  an  exact  one  thereof?  An 
idea,  however,  is  a  mysterious  plant  which  does  not 
always  develop  in  another  in  the  same  way  as  it  does 
in  its  originator,  without  counting  the  fact  that  it  may 
have  long  to  wait  before  encountering  soil  favourable 
for  its  perfect  fruition.  This  is  the  reason  it  is  so 
important  to  consider  ideas  as  they  appeared  to  the 
genius  who  gave  birth  to  them.  How  often  have  they 
thus  shown  themselves  to  be  greater  and  more  fertile  than 
they  had  seemed  as  interpreted  by  disciples  incapable  of 
thoroughly  understanding  them  !  "  Philosophia  duce 
regredimur  "  was  a  profound  motto  of  the  Renaissance. 

Is  it  necessary  to  call  to  mind  Descartes'  excellence 
as  a  writer  ?  From  this  standpoint,  too,  his  importance 
could  not  be  exaggerated.  As  regards  the  part  he 
played  in  history,  Desire  Nisard  has  shown  that  he  was 
the  first  to  offer  a  perfect  model  of  French  prose.  The 
language  of  Descartes  is  the  fabric  on  which  the 
style  of  our  great  writers  is  woven.  Considered  in 
itself,  this  language,  stamped  with  the  philosopher's 
method,  possesses  in  the  highest  degree  the  noblest 
qualities  of  every  language  :  propriety  of  terms,  and  the 
expression  of  order  in  ideas.  Cartesian  intuition  and 


DESCARTES  239 

deduction  have  left  their  impress  on  the  style  of  the 
Discours  de  la  Mtthode.  Not  that  this  language  is 
abstract  or  impersonal.  Descartes'  reason  is  a  living, 
enthusiastic  reason  ;  it  does  not  merely  put  acquired 
truths  in  the  form  of  syllogisms,  but  rather  endeavours 
to  discover  and  create,  to  communicate  its  creative 
activity  to  men's  intellects.  This  life  of  thought 
animates  the  style  itself,  which,  in  a  surprising  way, 
unites  to  precision  and  demonstrative  order,  motion, 
accent,  originality,  colour,  wit,  and  even  charm,  or 
irony  or  pride,  according  to  the  intellectual  passion 
which  is  pouring  into  the  soul  of  this  lover  of  truth. 
Whatever  impression  may  at  first  be  felt,  when  one 
at  times  becomes  bewildered  with  those  long  sentences 
which  demand  an  alert  reader,  capable  of  making  his 
own  deductions,  one  speedily  comes  under  the  charm 
and  power  of  this  masterly  style.  Even  nowadays,  if 
an  author's  manner  merely  suggests  that  of  Descartes 
in  some  respect  or  other,  people  vie  with  each  other  in 
praising  its  superiority  and  austere  seductiveness. 

In  a  word,  why  should  we  not  call  to  mind  the 
special  motives  which  cause  us  to  desire  that  the  works 
of  Descartes  should  be  read  by  as  large  a  circle  of 
readers  as  possible,  both  in  France  and  abroad  ? 

Descartes  is  one  of  the  purest  and  finest  expressions 
of  the  genius  of  our  race  :  the  diffusion  of  his  thoughts 
represents  our  life  and  influence. 

We  love  reason,  a  middle  path  between  the  spirit 
of  positivism,  which  contents  itself  with  facts,  properly 
so  called,  and  the  spirit  of  mysticism,  which  tends  to 
believe  without  demanding  proof.  Of  all  intellectual 
qualities,  the  one  we  most  prize  is  the  faculty  of 
judgment,  in  whose  sight  even  experience  and  reasoning 
are  sources  of  truth  only  on  condition  they  have  been 


24o  STUDIES  IN  PHILOSOPHY 

submitted  to  the  control  of  the  mind.  It  is  in  this 
direction  that  we  seek  after  clarity  and  order  in  ideas. 
For  a  system  to  be  well  constructed  and  consistent  is 
not  sufficient  for  us,  we  want  every  part  of  it,  taken 
separately,  to  be  intelligible  and  true,  and  we  would 
rather  hold  separately  the  two  ends  of  the  chain  of 
reasoning  without  apperceiving  the  intermediary  links, 
than  let  slip  the  truths  we  have  won  in  order  to 
grasp  the  hypothetical  connection  between  them.  Qne 
•ofL_ih£^dences.-  in.  which-JBce -~exceL  is... mathematics. 
Our  sense  of  clarity  and  logic  is  here  afforded  un- 
restrained activity.  In  the  moral  order  of  things, 
we  love  reason  with  an  ardent,  enthusiastic  love, 
that  has  at  times  gone  astray  or  formed  a  striking 
contrast  with  the  very  object  of  that  love  ;  but  through 
all  our  fluctuations  the  goal  of  our  endeavours  is 
clearly  a  harmonious  blending  of  individual  freedom 
and  rational  law,  in  which  neither  would  be  sacrificed. 
And  whilst  seeking,  in  a  practical  spirit,  for  what  suits  our 
own  nation,  it  is  impossible  for  us  to  separate  in  thought 
the  happiness  of  others  from  our  own,  or  to  desire  good 
in  any  other  than  the  universal  form  which  reason  ordains. 

Now,  we  find  in  Descartes  these  different  traits, 
which  are  amongst  the  principal  ones  in  our  nature. 
A  clear-headed  and  profound  philosopher  and  mathe- 
matician, excelling  in  finesse  and  in  geometrical  precision 
alike,  jealous  of  independence  though  obedient  to  reason, 
solicitous  of  the  practical  ends  of  life  and  ambitious  to 
work  for  the  happiness  of  all  mankind,  he  offers  us,  pre- 
eminently, the  model,  and,  as  it  were,  the  archetype  of 
the  qualities  we  aspire  to  show  forth. 

To  study  Descartes  and  make  him  better  known  is 
to  work  for  the  fulfilment  of  the  scientific  and  civilising 
mission  of  France. 


DESCARTES  241 

ON  THE  CONNECTION  BETWEEN  MORALS 
AND  SCIENCE  IN  THE  PHILOSOPHY  OF 
DESCARTES 

Mirum  mihi  videtur,  plerosque  homines  plantarum  vires,  siderum 
motus,  mctallorum  transmutationes,  similiumque  disciplinarum 
objecta  diligentissime  perscrutari,  atquc  interim  fere  nullos  de  bona 
mente  .  .  .  cogitare,  quum  tamen  alia  omnia  non  tarn  propter  se 
quam  quia  ad  hanc  aliquid  conferunt,  sint  aestimanda. — DESCARTES, 
Reg.  ad  dir.  ing.  Reg.  i . 

That  portion  of  Descartes'  writings  referring  to 
morals  is  not  insignificant,  though  neither  in  form  nor 
content  does  it,  at  first  glance,  appear  to  belong  to  his 
philosophical  work,  properly  so  called.  It  consists 
mainly  of  the  letters  to  the  Princess  Elizabeth  and  the 
Queen  of  Sweden  ;  in  them  Descartes  visibly  adapts 
himself  to  the  needs  and  desires  of  his  illustrious 
correspondents.  True,  a  sketch  of  practical  morals 
forms  part  of  the  Discours  de  la  Mtthode.  According 
to  a  document  published  in  1896,  by  Ch.  Adam,  it 
would  appear  that  Descartes  added  these  rules  some- 
what against  his  will,  because  of  the  pedagogues  and 
others  of  the  same  type  who  were  quite  ready  to 
accuse  him  of  having  neither  religion  nor  faith,  and  of 
wishing  to  destroy  both  the  one  and  the  other  by  his 
method.  As  regards  the  contents  of  these  writings  on 
morals,  they  are  certainly  very  dignified  and  lofty  in 
thought  and  admirable  in  form,  though  evidently 
possessed  of  little  in  common  with  the  philosopher's 
doctrine  itself.  Borrowed  from  St.  Thomas,  as  Baillet 
says,  and  intended,  according  to  Descartes  himself,  to 
reconcile  the  teachings  of  Aristotle,  Zeno  and  Epicurus1 
with  one  another,  they  seem  to  have  been  particularly 

1  OEuvres  philos.  de  Descartes,  edit.  Gamier,  iii.  184-5. 

R 


242  STUDIES  IN  PHILOSOPHY 

stamped  with  the  impress  of  stoicism.  Now,  stoicism 
was  then  a  well-known  and  popular  philosophy.  Des- 
cartes is  a  stoic,  as  the  heroes  of  Corneille  are  stoics. 
His  mathematics  has  nothing  to  do  with  his  stoicism. 
It  would  therefore  seem  either  that  Descartes,  so  far  as 
he  personally  was  concerned,  had  no  interest  in  moral 
research,  or  that,  if  he  did  make  profession  of  moral 
maxims,  they  resulted  rather  from  individual  feelings  or 
outer  influences  than  from  the  logical  development  of 
his  philosophy. 

I 

It  is  worthy  of  note  that  this  appreciation,  which  the 
first  rapid  examination  of  Descartes'  moral  writings 
induces  us  to  make,  by  no  means  conforms  with  the 
continually  repeated  declarations  of  the  philosopher 
concerning  the  object  of  philosophy. 

What,  according  to  the  Regulae?  is  the  earnest  way 
of  seeking  after  truth  ?  It  is  to  think  solely  of  increasing 
the  natural  light  of  reason,  not  for  the  solving  of  any 
particular  scholastic  difficulty,  but  rather,  at  every 
conjuncture  in  life,  for  the  purpose  of  making  the 
understanding  capable  of  prescribing  to  the  will  the  line 
of  action  it  ought  to  choose.  The  reason  Descartes  is 
keenly  desirous  of  learning  to  distinguish  the  true  from 
the  false  is,  as  he  tells  us  in  the  Discours  de  la  Mhhode? 
because  he  knows  this  to  be  the  means  of  seeing  clearly 
into  his  own  actions  and  going  through  life  with  calm 
assurance.  Again,  in  the  Preface  to  the  Principes?  he 
defines  philosophy  as  the  study  of  wisdom,  which,  he 
says,  consists  of  a  perfect  knowledge  of  everything  a 
man  is  capable  of  knowing,  both  in  the  conduct  in  life, 
the  preservation  of  health  and  the  invention  of  all  the 

1  i.  i.  2  i.  14.  3  Baillet,  La  Vie  de  M.  Descartes,  i.  115. 


DESCARTES  243 

arts.  This  study,  he  adds,  is  more  necessary  for  the 
regulation  of  our  morality  than  the  use  of  our  eyes  for 
the  guidance  of  our  steps. 

And  indeed,  according  to  Clerselier,  who  appears  to 
have  known  him  most  intimately,  morals  was  the  object 
of  his  most  frequent  meditation.1  True,  he  did  not 
like  writing  about  such  matters,  but  that  is  from  a 
feeling  of  prudence,  as  he  himself  explains.2  In  physics, 
also,  he  more  than  once  preferred  silence  to  the  risk  of 
persecution. 

All  the  same,  we  may  ask  ourselves  whether,  in  the 
work  he  has  left  us,  the  moral  ideas  and  the  physical 
doctrines  really  form  part  of  one  and  the  same  system, 
or  whether  they  are  not  like  two  streams  which  flow 
parallel  to  each  other,  without  their  waters  ever  mingling. 
Certainly  Descartes  offers  us  the  rules  of  his  provisional 
system  of  morals  as  being  deduced  from  his  method. 
But  then,  of  what  use  is  it  to  affirm  this,  if  he  intro- 
duced these  rules  only  to  throw  pedagogues  off  the 
scent  ?  In  themselves,  they  appear  anything  but  part 
and  parcel  of  his  philosophy.  It  is  also  true  that,  in 
the  Preface  to  the  Prmcipes5  he  speaks  of  a  definitive 
system  of  morals  which  presupposes  a  complete  know- 
ledge of  all  other  sciences.  Many,  however,  consider 
that  he  did  not  even  outline  this  system  of  morals, 
and  that  it  is  his  provisional  system  which  is  in  reality 
his  final  one.4 

The  question  is  a  puzzling  one.  It  would  be  unfair 
to  judge  Descartes  solely  by  those  portions  of  his  work 
which  his  prematurely  curtailed  life  enabled  him  to 
complete.  In  creations  of  thought  the  inner  tendency 
and  the  living  principle  of  development  are  frequently 

1  Baillet,  La  Vie  de  M.  Descartes,  i.  115.  2  ii.  282. 

3  Edition  Gamier,  i.  592.  4  Cf.  e"d.  Gamier,  iii.  179. 


244  STUDIES  IN  PHILOSOPHY 

more  important  than  the  immediately  observable  results. 
The  reality  of  a  Cartesian  system  of  morals  would  be 
satisfactorily  proved  if  it  were  shown  that  the  philo- 
sophy of  Descartes  contained  within  itself  the  germs  of 
such  a  system. 

II 

There  can  be  no  doubt  but  that  this  philosophy, 
speaking    generally,    deals   with    practical    experience. 
Although  fond  of  withdrawing  into  solitude  and  seclu- 
sion for  the  purpose  of  meditation,  Descartes  is  anything 
but   an    armchair    philosopher.      He  possesses   in  the 
highest  degree  the  sense  of  reality,  interests  himself  in 
the  doings  of  his  times,  converses  with  men  of  divers 
stations  and  temperaments,  and  listens  attentively  to 
the  remarks  of  each  on  his  own  particular  subject.     He 
considers  that  our  highest  duty  is  to  bring  about  the 
general  good  of  all  men,  to  the  best  of  our  ability >-j 
Consequently  his  chief  grievance  against  the  scholastic    I 
philosophy  is  that  it  is  purely  speculative  and  gives  no^l 
results.     Instead   of  this   philosophy    of  arguers   and 
disputants,  he  seeks  after  a  practical  system  of  philo- 
sophy calculated  to  place  at  man's  disposal  the  power 
and  action  of  fire,  water,  air  and  all  the  other  bodies 
around  us,  and  which  will  make  him,  as  it  were,  master 
and  owner  of  nature.1     It  is  his  dream   to  preserve 
mankind  from  illness  and  disease,  perhaps  even  from 
the  debility  of  old  age.     His  death  was  announced  in 
the   Gazette  d'  Anvers  in   the  following   terms  : 2    "In 
Sweden  there  has  just  died  a  fool,  who  said  that  he 
could  live  as  long  as  he  wished." 

Descartes,   like  Bacon,  following  the   traditions  of 

1  Mfth.  vi.  2. 

2  Adam,    GOttingen    MS.    (Revue    bourgnignonne    de    I'Enseignement 
suptneur,  1896). 


DESCARTES  245 

the  magicians  and  alchemists,  was  inspired  with  the 
ambition  to  dominate  that  nature  which  the  ancients 
had  contented  themselves  with  gazing  upon. 

The  alchemists,  however,  believed  that  in  order  to 
make  nature  act  as  they  pleased,  all  that  was  needed 
was  to  set  it  going  by  an  altogether  external  and 
empirical  imitation  of  its  processes.  The  magicians 
regarded  nature  as  a  mysterious,  perhaps  diabolical, 
power,  whose  will  had  to  be  chained  down  by  means 
of  formulas.  Bacon  himself,  in  his  immediate  search 
after  an  active  philosophy,  can  find  no  reason  for 
admitting  that  nature  will  respond  to  human  promptings, 
except  that  such  response  is  necessary  in  order  that  man 
may  be  able  to  act  upon  her.  His  science  remains 
blind  because,  confounding  the  means  with  the  end,  it 
recognises  no  other  principles  than  the  rules  that  admit 
of  being  applied,  just  as  they  stand,  to  practice.1 

Descartes'  originality  consisted  in  regarding  the 
legitimacy  of  the  problem  of  man's  rule  over  nature  as 
uncertain,  and  its  solution  doubtful,  as  long  as  no 
attempt  was  made  to  discover  by  what  internal  mode  of 
working,  nature  really  brings  about  any  particular  effect  - 
from  any  particular  cause.  He  considered  that  practice 
implied  theory  in  the  real  sense  of  the  word  ;  the 
knowledge  of  the  interior  of  things.  To  his  mind  it 
was  from  this  standpoint  that  nature  must  be  considered, 
if  we  would  succeed  in  becoming  master  of  her.  So, 
too,  in  the  past,  dealing  with  the  moral  order  of  things, 
Socrates  had  taught  that  the  practical  skill,  quite  legiti- 
mately sought  after  by  the  Sophists,  could  only  be 
attained  in  a  roundabout  way,  viz.  by  a  rational  know- 
ledge of  the  essence  of  virtue.  And  since,  to  Descartes, 
the  very  type  of  theory,  the  king  of  sciences,  was 

1  Nov.  Org.  i.  4.;  ii.  1-5. 


246  STUDIES  IN  PHILOSOPHY 

mathematics,  he  made  it  his  object  to  demonstrate  that 
everything  in  nature  is  brought  about  mathematically  ; 
hence  his  metaphysical  speculations.  He  proves,  both  by 
the  perfections  of  God  and  by  the  clear,  distinct  nature 
of  the  idea  of  extent,  that  we  are  entitled  to  consider 
mathematical  qualities  as  the  essence  of  material  things. 

Consequently,  he  studies  mathematics,  and  his  whole 
work  is  dominated  by  this  science ;  only,  however, 
because,  according  to  him,  it  is  by  considering  things 
from  this  point  of  view  that  we  can  really  make  them 
our  own.1  And  it  is  this  practical  end,  ever  present 
in  his  mind,  that  determines  the  general  line  of  his 
investigations.  He  does  not  dwell  upon  the  develop- 
ments of  science,  which  would  be  merely  of  speculative 
interest.  He  is  content  with  setting  up,  in  mathematics, 
the  few  general  principles  which  will  enable  him  to  base 
mechanics  and  physics  on  this  discipline.  These  two 
sciences,  in  turn,  need  to  be  developed  only  in  the 
direction  and  degree  necessary  to  make  possible  the 
science  of  life.  His  object  is  to  prove  that  life  itself 
is  nothing  but  a  mechanism,  and  is,  consequently,  subject 
to  our  control.  Whilst  studying  one  science,  Descartes 
is  thinking  of  the  science  which,  in  the  nature  of  things, 
is  to  come  after,  and  bring  him  nearer  to  practice. 
The  idea  of  the  goal  in  view,  which  never  leaves  him, 
controls  and  restrains  his  efforts.  Semper  ad  eventum 
festmat. 

This  method  enabled  him  to  conceive  the  possibility 
of  carrying  through,  alone,  his  project  of  instituting  a 
universal  science.  In  1637  ne  came  to  the  conclusion 
that  the  truths  he  had  found  in  the  various  sciences 
were  nothing  but  the  consequences  of  five  or  six  main 
difficulties  he  had  overcome,  on  which  they  depended  ; 

1  Baillet,  ii.  227. 


DESCARTES  247 

and  he  imagined  he  needed  only  to  win  two  or  three 
other  like  battles  to  bring  his  plans  to  a  successful  issue.1 

Here,  too,  is  the  explanation  of  his  apparently 
capricious  passing  from  one  science  to  another.  From 
1623  he  began  to  neglect  geometry,2  and  six  years 
afterwards  plunged  into  metaphysical  meditation,  to 
which,  however,  he  devoted  only  nine  months.  A  year 
afterwards  he  reminds  Mersenne  that  he  has  long  ago 
abandoned  the  study  of  mathematics,  anxious  not  to 
waste  his  time  any  longer  in  unproductive  effort.  From 
1629  to  1633  he  is  mainly  occupied  with  physics.  At 
the  end  of  the  Discours  de  la  Methode  he  announces 
his  intention  to  spend  the  rest  of  his  life  in  wresting 
from  physics  a  more  certain  medicine,  or  art  of  curing 
disease,  than  the  one  in  vogue. 

This,  in  short,  is  the  explanation  of  that  particularity 
of  his  system  for  which  Newton  reproached  him  so 
strongly,  viz.  hypothesis,  regarded  in  certain  cases  as  a 
sufficient  explanation  of  phenomena.  A  strict  adherent 
of  the  principle  of  following  the  line  of  least  resistance 
in  his  own  method  of  work,  Descartes  contents  himself, 
in  his  theories,  with  what  is  indispensable  for  practice. 
Now,  from  this  point  of  view,  provided  it  be  possible 
to  make  a  forecast  of  the  result,  it  matters  little  that 
the  mechanism  of  nature,  in  detail,  should  be  in  every 
respect  what  it  has  been  conceived  to  be.  Well  aware 
that  in  mathematics  several  solutions  are  often  possible, 
Descartes  comes  to  regard  it  as  sufficient,  even  in  physics, 
if  he  obtains  one.  He  believes  he  has  done  everything 
necessary  if  the  causes  he  has  explained  are  such  that 
all  the  effects  they  are  capable  of  producing  are  similar 
to  those  we  find  in  the  manifested  world.  He  considers 
it  useless  to  inquire  whether  the  effects  are  really  brought 

1  Mfth.  iv.  4.  2  Baillet,  i.  in. 


248  STUDIES  IN  PHILOSOPHY 

about  by  these  causes  or  by  others.  He  thinks  it  is  as 
useful  in  life  to  know  causes  thus  imagined  as  to  know 
the  real  ones.1  On  this  point  he  is  satisfied  with  moral 
certainty.2 

In  the  progress  of  knowledge,  as  thus  understood, 
morals  cannot  fail  to  find  a  place,  all  the  more  so 
because,  according  to  Descartes,  the  root  and  trunk  of 
a  tree  are  mainly  held  in  esteem  on  account  of  the  fruit 
they  should  produce,  and  it  is  mostly  on  the  sciences 
which  should  come  last,  medicine,  mechanics  and  morals, 
that  the  primary  utility  of  philosophy  depends.3  And 
Descartes  does  not  despair  of  satisfying  himself  as 
regards  these  ultimate  objects,  in  spite  of  the  shortness 
of  human  life  and  the  limits  of  our  intelligence,  for  the 
very  reason  that  he  is  able  to  economise  his  strength 
and  demand  of  each  science  only  what  it  can  and  should 
give  him  for  the  carrying  out  of  his  plans.  The  pro- 
ductiveness of  knowledge  lies  not  in  its  extent,  but 
rather  in  its  clearness  and  precision. 

Ill 

But  now,  what  is  the  nature  of  the  morals  to  which 
this  progress  will  lead  ?  Does  it  not  merely  tend  to 
enable  us  to  dispose  of  human  nature,  by  means  of  the 
science  of  man,  just  as  we  dispose  of  the  corporal  nature, 
by  means  of  the  science  of  the  body  ?  Is  not  psychical 
mechanics  all  that  Descartes  has  in  view  ? 

As  a  matter  of  fact,  Descartes  laid  the  foundations 
of  some  such  morals  in  his  Traite  des  Passions,  in  which, 
expounding  the  principle  governing  these  mental 
activities,  he  teaches  us  to  modify  and  control  them. 

1  Prindpes,  iv.  104.  2  Baillet,  ii.  227-8. 

3  Pref.  of  the  Prin.  Gamier,  i.  192. 


DESCARTES  249 

As,  moreover,  this  study  shows  us  how  far  the  mind 
depends  on  the  temperament  and  arrangement  of  the 
organs  of  the  body,  Descartes  distinctly  concludes  that, 
if  it  is  possible  to  find  any  means  whereby  men,  gener- 
ally, may  be  made  wiser  and  more  skilful,  this  means 
must  be  sought  for  in  medicine. 

Thus  would  appear  to  be  completed  the  edifice 
planned  by  our  philosopher.  Its  culminating  point  is 
a  system  of  morals,  though  how  different  from  that 
indicated  in  the  Discours  de  la  Methode  and  the  Lettres  \ 
This  latter,  instinct  with  the  spirit  of  antiquity  or 
with  Christian  influences,  was  either  an  exhortation, 
a  metaphysic  or  a  religion.  That  of  the  Principcs  and 
the  Traite  des  Passions  was  only  the  final  and  most 
immediately  practical  application  of  modern  science. 
According  to  the  Lettres,  man  ought  to  seek  outside 
of  the  world,  in  those  perfections  that  depend  solely 
on  free-will,  resignation,  constancy,  and  the  mystical 
love  of  God  and  men,  for  those  things  to  which  he  is 
to  bend  his  will.  According  to  the  Traite  des  Passions, 
man,  a  mere  part  of  nature,  could  aim  at  nothing 
else  than  maintaining  the  integrity  of  his  existence 
by  utilising  the  mechanism  of  the  universe  for  his 
own  advantage.  Now,  it  is  easy  to  see  how  these 
scientific  morals  are  the  fruit  of  the  Cartesian  philo- 
sophy, whereas  the  former  seem  to  remain  outside  the 
logical  development  of  this  philosophy. 

And  yet,  is  it  right  to  content  oneself  with  this 
result,  and  declare  that  Descartes,  as  a  philosopher, 
knows  no  other  morals  than  applied  science  ? 

It  is  unnecessary  to  have  recourse  to  such  of 
Descartes'  writings  as  deal  specially  with  morals  in 
order  to  see  how  narrow  and  incomplete  such  an  inter- 
pretation would  be.  Speaking  generally,  it  is  not 


250  STUDIES  IN  PHILOSOPHY 

science  that  is  the  centre  of  the  Cartesian  philosophy  ; 
it  is  man,  or  rather  the  reason  within  man.  Even  when 
studying  the  sciences  of  nature,  it  is  not  science  itself 
that  our  philosopher  has  in  view,  it  is  the  formation 
of  the  judgment  by  science.  Judgment  is  the  power 
to  distinguish  the  true  from  the  false  in  all  things 
without  hesitation  or  uncertainty.  To  do  this  we 
must  develop  within  ourselves  a  kind  of  sense  of  truth. 
Mathematics,  especially  algebra,  is  a  wonderful  help  in 
this  respect.1  By  accustoming  the  mind  to  feed  upon 
truth  and  never  be  satisfied  with  false  reasons,  mathe- 
matics compels  it  to  quit  its  natural  indifference  and 
leads  it  in  the  direction  of  its  own  perfection.  It  is 
this  mental  culture,  not  the  knowledge  of  particular 
truths,  that  forms  the  real  utility  of  the  sciences.2 
They  cannot  be  detached  from  reason,  as  the  fruit  is 
detached  from  the  tree,  for  it  is  in  reason  that  they 
have  both  their  principle  and  their  end. 

Descartes,  however,  does  more  than  train  his  reason 
mechanically  by  exercise  and  habit ;  he  uses  the  intel- 
lectual force  thus  gained  in  studying  the  nature  of 
reason  itself,  analysing  its  content,  gauging  its  power 
and  trying  to  discover  its  purpose.  He  rises  above 
science  to  metaphysics.  Not  that  this  makes  it  neces- 
sary that  he  should  free  himself  from  the  requirements 
of  science.  Rather  is  it  science  which,  properly  inter- 
preted, opens  up  the  path  leading  to  this  higher  know- 
ledge. He  remarks  that  the  mathematical  method, 
however  perfect  it  be,  is  nothing  but  the  outer  cover 
of  the  true  method.3  The  latter,  apart  from  the  parti- 
cular form  given  to  it  by  geometricians,  is  of  universal 
import,  and  allows  of  the  truths  contained  in  any  sub- 
ject being  obtained.  By  the  use  of  this  method,  then, 

1  Regulae,  i.  2  M/th.  iii.  5.  3  Regulae,  iv.  20. 


DESCARTES  251 

one  may  succeed  in  strictly  demonstrating  the  truths 
of  metaphysics  as  well  as  those  of  geometry.  To 
attempt  thus  to  know  God,  oneself  and  the  first  prin- 
ciples of  the  science  of  nature,  is  the  main  use  man 
ought  to  make  of  his  reason.1 

If,  therefore,  it  is  conceived  that  a  purely  natural 
philosophy  has  for  its  ultimate  object  the  supremacy  of 
man  over  nature,  a  more  complete  philosophy  sees  in 
this  very  supremacy  only  a  means  at  the  service  of  a 
loftier  end.  No  longer  is  it  merely  a  question  of 
governing,  but  of  doing  so  in  the  name  of,  and  with 
a  view  to,  reason.  To  moderate  the  influence  of  the 
body  by  medicine  is  indeed  the  most  practical  external 
means  of  helping  men  to  become  wise  ;  but  medicine  is 
not  wisdom,  any  more  than  the  tool  is  the  work  upon 
which  it  is  used.2  In  the  same  way,  to  control  one's 
passions,  owing  to  our  knowledge  of  their  mechanism, 
is  not  the  same  as  directing  them  to  their  true  use. 
Not  any  thought  we  please  should  we  attempt  to  sub- 
stitute for  those  which  passion  suggests,  but  rather  the 
thoughts  which  really  free  the  soul,  those  of  which  the 
reason  approves.  For  it  is  the  duty  of  reason  to 
examine  the  correct  value  of  the  various  benefits,  the 
attaining  of  which  depends  on  ourselves.3  And  even 
above  the  right  use  of  the  passions,  which  concerns  the 
soul  from  the  standpoint  of  its  union  with  the  body, 
Descartes  places  the  benefits  of  the  soul  itself  from  the 
standpoint  of  its  own  life.  There  is  a  joy  that  is  purely 
intellectual.4  The  soul  can  have  its  own  pleasures 
apart  from  all  else.5  The  practice  of  virtue,  to  which 

1  Letter  to  Mersenne,  i5th  April  1630,  Garn.  iv.  303. 

2  Baillet,  ii.  11-12. 

3  Passions,  art.  144.     Cf.  Letter  to  the  Princess  Elizabeth,  ist  June  1645, 
Gamier,  iii.  189. 

4  Passions,  art.  91.  5  Ibid.  art.  212. 


252  STUDIES  IN  PHILOSOPHY 

these  pleasures  are  linked,  is  not  only  a  sovereign 
remedy  against  the  passions,1  it  is  also  the  greatest 
perfection  to  which  one  can  lay  claim,  for  it  is  the 
genuine  action  of  a  will  that  is  free.2 

Above  the  morals  of  means,  then,  which  is  hardly 
anything  but  applied  physics,  Descartes  conceives  of 
a  morals  of  ends  which  is  founded  directly  on  the 
loftiest  elements  of  metaphysics.  Both  of  these  morals 
are  based  on  science,  if  this  word  is  taken  in  its  Cartesian 
sense,  i.e.  as  signifying  the  clear,  distinct  knowledge 
both  of  corporeal  and  of  spiritual  things.  The  second, 
however,  cannot  be  derived  solely  from  the  science  of 
nature,  whose  domain  does  not  include  reason  and  will. 

Now,  when  Descartes  undertakes  to  define  this 
superior  morals,  it  is  only  natural  that  he  should  again 
come  into  touch  with  the  Stoics  and  other  philosophers 
of  antiquity,  to  whom  the  culture  of  reason  formed 
the  main  interest  in  life.  Human  reason  has  not 
changed  its  nature,  from  the  time  of  Aristotle  to  that 
of  Descartes.  The  most  perfect  expressions  it  has  met 
with,  ever  since  men  have  been  able  to  reflect,  thus  find 
their  place  in  the  Cartesian  system,  and  that  not  as 
mere  patchwork,  but  as  integral  parts  thereof. 

They  have  not,  however,  been  transferred  into  that 
system  just  as  they  were.  Stoic  morals,  in  particular, 
is  for  Descartes  nothing  but  a  provisional  system  of 
morals.  To  try  to  conquer  oneself  rather  than  fortune 
is  surely  the  wisest  decision  to  arrive  at,  as  long  as  we 
are  powerless  to  modify  the  outer  world.  But  it  is  this 
very  power  that  the  Cartesian  philosophy  confers  upon 
us  ;  therefore,  in  place  of  morals  inculcating  abstention 
it  substitutes  positive  and  active  morals.  Likewise, 
to  endeavour  to  find  the  rules  of  conduct  in  the  outer 

1  Passions,  art.  148.  2  Ibid.  art.  17-18. 


DESCARTES  253 

order  of  things  themselves  is  the  best  course  to  follow,  as 
long  as  we  are  ignorant  of  the  first  principles  of  which 
this  order  is  a  continuation.  But  when,  as  the  result 
of  a  methodical  culture  of  reason,  man  has  come  to 
know  the  principal  truths  from  which  the  laws  of  nature 
are  derived,  he  substitutes — -and  that  in  a  precise  and 
positive  sense  of  which  the  ancients  knew  nothing — 
for  the  maxim  :  "  Follow  nature,"  that  other  maxim  : 
"  Follow  true  reason."  * 

The  doctrine  of  a  proper  content  of  reason,  and  of 
man's  possibility  to  conform  things  thereto,  gives  an 
original  stamp  to  Cartesian  morals.  When  brought  in 
contact  with  a  mysterious,  inflexible  nature,  the  ancients 
could  only  contemplate  and  acquiesce  in  it,  or  else 
retire  within  themselves.  In  the  case  of  Descartes, 
reason,  grounded  on  a  science  which  opens  out  things 
for  its  consideration,  becomes  an  efficient  power,  a 
natural  force  ;  it  assumes  the  task  of  employing  in  its 
own  perfecting  the  mechanism  of  external  things. 
And  so,  whereas  Socrates  regarded  the  claim  to  investi- 
gate the  causes  of  physical  phenomena  as  vain  and 
sacrilegious,  and  the  Stoics  looked  upon  resignation  and 
detachment  from  the  world  as  the  principle  and  the 
goal  of  all  felicity,  Descartes  can  see  no  limit  to  the 
conquests  that  science — and  by  means  of  science,  human 
reason — will  achieve  over  the  world.  Whereas  the 
Stoics  only  condemned  passion, — in  which  they  recog- 
nised the  violence  and  indiscipline  of  brute  nature, — 
Descartes,  by  the  aid  of  a  science  which  penetrates  to 
the  causes  of  passion,  subjugates  and  converts  it  into 
an  auxiliary  of  reason.  Man  is  no  longer  crushed  by 
nature,  he  makes  use  of  her.  The  soul,  no  longer  a 

1  Letters  to  the  Princess  Elizabeth,  ist  and  i5th  May  1645,  Gamier, 
Hi.  181,  183. 


254  STUDIES  IN  PHILOSOPHY 

prisoner  of  the  body,  guides  and  controls  it.  Morals 
is  no  longer  the  art  of  retiring  from  the  world  and 
being  sufficient  unto  oneself;  it  is  the  command  to 
make  of  reason — which  is  our  very  essence — a  living, 
sovereign  reality,  the  queen  of  nature. 

And  this  very  sovereignty  of  reason  over  things  is, 
to  Descartes,  nothing  but  the  means  it  has  for  pursuing 
the  ends  proper  to  itself,  such  as  the  love  of  God, 
and  interest  in  the  all,  of  which  one  forms  part.1  Car- 
tesian metaphysics,  by  its  method,  enables  us  to  know, 
with  certainty  these  ultimate  truths  which  are  the 
indispensable  illumination  of  the  will.  This  gives 
us  another  originality  of  Descartes'  morals.  Most 
certainly  the  ancients  raised  the  virtues  to  a  lofty  pitch  ; 
but  as  they  were  ignorant  of  true  metaphysics  they 
could  not  possibly  become  well  acquainted  with  the 
virtues,  and  what  was  called  by  so  fine  a  name  was 
frequently  nothing  more  than  an  aberration  of  the  will.2 

Thus  it  is  really  for  its  close  union  with  science  that 
Cartesian  morals  is  distinguished  throughout.  Still, 
the  pure  and  simple  statement  that  it  is  derived  from 
science,  especially  the  science  of  natural  things,  could 
not  be  made.  In  all  its  phases  it  makes  use  of  science 
for  the  attainment  of  its  object :  the  complete  deter- 
mination of  will  by  reason.  The  full  realisation  of 
reason  is  the  end  :  all  else  is  but  the  means.  In  all 
things,  said  Descartes,3  what  we  must  seek  after  is 
the  bona  mens ;  nothing  else  deserves  to  be  taken 
into  consideration  except  in  so  far  as  it  contributes 
thereto. 

1  Letter  to  the  Princess  Elizabeth,  i5th  June  1645,  Gamier,  iii.  192-3. 

2  Mtth.  i.  10.  3  Reg.  i. 


KANT 

"Was  uns  zu  thun  gebtihrt,  dess  sind  wir  nur  gewiss." — KANT  (1782). 

THE  philosophy  of  Kant  is  one  of  the  most  important 
facts  in  the  history  of  the  human  mind.  Kuno  Fischer, 
the  well-known  historian  of  modern  philosophy,  affirmed 
that  it  represents  nothing  less  than  a  revolution  of  like 
nature  to  that  wrought  by  Socrates  when  he  brought 
mankind  back  from  the  study  of  the  world  to  the  study 
of  self ;  indeed,  it  sets  before  the  human  mind  the  task, 
not  of  discovering  the  principles  of  being  and  forming 
a  conception  of  the  universe,  but  rather  of  looking  for 
the  conditions  of  knowledge  itself,  the  origin  of  our 
representations  and  judgments,  and  their  importance. 
Windelband  shrewdly  said  that  the  rationalism  of  Kant 
is  the  concentration  in  a  living  unity  of  all  the  motor 
principles  of  modern  thought. 

Kant's  philosophy,  in  fact,  was  the  beginning  of  the 
development  of  German  philosophy,  strictly  so  called. 
From  Fichte  or  Schelling  on  to  Wundt  or  Riehl,  there 
is  not  a  single  German  philosopher  who  does  not  either 
continue  or  elaborate  Kantian  ideas.  But  even  outside  of 
Germany,  Kantianism  exercises  an  influence  that  grows 
greater  and  greater  the  better  it  is  known.  Though 
refuted  by  some,  it  is  accepted  by  the  rest  and  is  one 
of  the  essential  factors  of  contemporary  philosophic 
thought.  In  France,  particularly,  it  attracts  not  merely 
a  keen  historical  interest,  but  a  theoretical  one  as  well. 

255 


256  STUDIES  IN  PHILOSOPHY 

There  exists  a  very  flourishing  French  neo-criticism  and 
scarcely  a  single  philosophic  dissertation  appears  in  which 
Kant's  point  of  view  is  not  discussed,  whilst  its  action 
makes  itself  felt  even  in  literature  and  social  life. 

It  is  no  easy  task  to  set  forth  the  true  nature  of  a 
doctrine  dealing  with  present-day  preoccupations  and 
controversies.  The  safest  course  to  pursue  will  be  to 
leave  on  one  side  the  many  developments  it  may  have 
undergone,  and  look  upon  it,  as  far  as  possible,  from 
the  philosopher's  own  standpoint. 

I. — BIOGRAPHY 

Kant  was  a  contemporary  of  Frederick  the  Second  and 
the  French  Revolution.  His  principal  works  appeared 
between  1770  and  1797.  Though  he  valued  the 
triumphs  of  right  more  highly  than  those  of  might, 
yet  he  would  never  agree  to  separate  freedom  from 
order  and  discipline.  The  moral  environment,  in  which 
his  thought  developed,  was  Pietism  on  the  one  hand, 
and  the  philosophy  of  the  eighteenth  century  on  the  other. 
Pietism,  which  is  opposed  to  abstract,  theological  Pro- 
testantism, set  practice  before  dogma ;  it  extolled  feeling 
and  the  spirit  of  devotion,  interior  piety  and  the  private, 
individual  interpretation  of  the  Scriptures.  The  philo- 
sophy of  the  eighteenth  century,  the  philosophy  of  en- 
lightenment (Aufklarungsphilosophie),  as  it  was  called  in 
Germany,  teaches  that  all  the  evil  from  which  mankind 
suffers,  is  the  result  of  ignorance  and  of  the  bondage  that 
succeeds  it,  and  also  that  the  progress  of  enlightenment,  in 
itself  alone,  procures  happiness  and  its  ensuing  liberation. 

The  life  of  Kant  may  be  divided  into  three  main 
periods,  that  correspond  to  the  different  phases  of  his 
philosophic  development  :  ist,  childhood  and  youth 


KANT  257 

from  1724  to  1755,  a  Peri°d  °f  study  and  preliminary 
essays  ;  2nd,  the  years  he  spent  as  Privatdozent,  from 
1755  to  1770,  immediately  preceding  his  critical  work; 
3rd,  his  professorship,  from  1770  to  1797,  devoted  to 
criticism  and  the  development  of  his  teachings. 

Immanuel  Kant  was  born  in  KOnigsberg  on  the 
22nd  of  April  1724.  This  town,  in  which  the  whole 
of  his  life  was  destined  to  be  spent  almost  without  a 
break,  was  a  large  commercial  centre  to  which  there 
flocked  a  considerable  number  of  Jews,  Poles,  English 
and  Dutch.  Here  the  philosopher  found  ample  material 
for  psychological  and  moral  observations.  KOnigsberg, 
a  university  town,  was  likewise  the  centre  of  intellectual 
and  political  life  in  the  Duchy  of  Prussia. 

The  family  of  Kant  was  of  Scotch  origin.  His  name 
was  spelt  Cant,  which,  as  it  was  pronounced  tsant  in 
German,  he  changed  to  Kant.  His  father  was  a  saddler, 
poor  and  of  stern  morality.  His  mother,  Anna  Regina 
Reuter,  says  our  philosopher,  was  a  woman  of  consider- 
able intelligence  and  lofty  ideals  ;  she  was  an  earnest 
and  devout  Pietist,  though  her  religion  was  free  from 
both  mysticism  and  fanaticism.  Kant  was  the  fourth  in 
a  family  of  eleven  children.  The  importance  of  and 
respect  for  everything  that  was  religious  and  moral  was 
inculcated  on  him  from  his  earliest  years.  He  quietly 
acquiesced  in  this  influence  and  retained  a  keen  and 
pious  memory  thereof  throughout  life. 

At  the  age  of  nine  he  entered  the  Collegium 
Fredericanum,  the  master  of  which  was  Franz  Albert 
Schulz,  professor  of  theology  at  the  University  of 
KOnigsberg.  Schulz  was  Kant's  first  master.  An 
ardent  Pietist,  he  put  his  entire  soul  into  his  teachings. 
From  him  Kant  learnt  to  regard  interior  piety  as 
superior  to  reasoning,  and  practice  as  more  important 

s 


258  STUDIES  IN  PHILOSOPHY 

than  dogma.  It  may  be  noted  that  he  invariably  spoke 
with  respect  and  gratitude  of  his  Pietist  masters.  Was 
it  the  philosopher,  or  was  it  the  former  Pietist,  who,  in 
1782,  wrote  in  the  epitaph  of  Lilienthal,  the  minister 
who  had  married  his  parents,  the  line  : 

Was  uns  zu  thun  gebiihrt,  des  sind  wir  nur  gewiss  ? l 

Kant  spent  seven  years  at  the  Collegium  Fredericanum. 
He  was  devoted  to  Latin  and  to  Roman  stoicism,  which 
he  looked  upon  as  the  religion  of  discipline.  Right  to 
the  end  of  his  life  he  adopted  as  his  motto  these  lines 
of  Juvenal : 

Summum  crede  nefas  animam  praeferre  pudori 
Et  propter  vitam  vivendi  perdere  causas. 

In  1740,  at  the  age  of  seventeen,  he  entered  the 
University  of  Konigsberg,  intending  to  study  theology. 
His  idea  at  the  time  was  to  become  a  minister,  but  he 
quickly  changed  his  mind.  At  first  he  attended  the 
classes  of  Martin  Knutzen,  professor  of  mathematics 
and  philosophy.  Knutzen  was  his  second  master,  and 
he  too  was  a  Pietist.  Although  a  disciple  of  Wolf  in 
philosophy,  he  was  opposed  to  dualism,  and  came  round 
to  the  genuine  teaching  of  Leibnitz,  according  to  which 
representative  and  motor  force  participate  in  and  imply 
each  other. 

Kant  was  indebted  to  Knutzen  for  his  acquaintance 
with  the  works  of  Newton,  who  may  be  called  his 
third,  and  perhaps  his  principal  master.  The  New- 
tonian philosophy  was  to  Kant  an  experimental  proof 
of  the  possibility  of  an  a  priori  knowledge  of  nature. 
Henceforth  it  was  his  object  to  explain  this  possibility, 
and  along  that  line  to  become,  in  a  way,  the  Newton 
of  metaphysics. 

1  What  duty  calls  upon  us  to  do,  of  that  alone  are  we  certain. 


KANT  259 

Knutzen  did  much  to  turn  Kant  from  theology  to 
philosophy.  And,  by  degrees,  he  dropped  the  strict 
orthodoxy  of  his  Pietism,  retaining  nothing  but  its 
moral  rigidity. 

Unable  to  obtain  a  living  on  the  fees  from  his 
lessons,  Kant  became  a  private  tutor  in  1746,  in  which 
capacity  he  remained  for  nine  years.  He  was  thus 
brought  into  connection  with  foreigners  and  the 
nobility,  and  began  to  take  considerable  interest  in 
foreign  literature  and  politics.  He  went  into  society, 
anxious  to  show  himself  a  worthy  citizen. 

This  the  first  period  of  his  life  concluded  with  the 
anonymous  publication  of  his  Universal  History  of  Nature 
and  Theory  of  the  Heavens  (1755),  a  work  that  prepared 
the  way  for  the  theory  of  Laplace  on  the  formation  of 
the  heavenly  bodies. 

After  obtaining  promotion  by  the  writing  of  a  dis- 
sertation on  fire,  and  habitation  by  one  on  the  first 
principles  of  metaphysical  knowledge,  he  was  appointed 
Pr'matdozent.  He  taught  mathematics,  physics,  the 
theory  of  fortification,  pyrotechnics,  logic,  morals,  and 
philosophical  Encyclopedism.  His  teaching  was  full  of 
life.  Whatever  his  subject,  he  spoke  as  one  possessed 
of  special  knowledge,  the  result  being  that  he  met  with 
considerable  success.  Between  1760  and  1769  he  also 
lectured  on  natural  theology,  anthropology,  criticism  of 
the  proofs  of  the  existence  of  God,  and  the  doctrine 
of  the  beautiful  and  the  sublime. 

Here  we  find  the  influence  of  Rousseau,  whose 
works  were  then  becoming  known  and  being  consider- 
ably discussed.  Kant  devoured  Rousseau's  books  and 
was  thus  brought  to  take  a  passionate  interest  in  moral 
problems,  the  combat  against  prejudice,  and  the  return 
to  nature  and  reason.  Rousseau  taught  him,  he  tells 


260  STUDIES  IN  PHILOSOPHY 

us,  not  to  despise  man's  natural  tendencies.  Physical 
science  a  priori  as  a  fact  was  what  he  had  found  in 
Newton  ;  Rousseau  now  made  him  see  morality  as  a 
fact.  These  facts  he  purposed  to  analyse. 

With  the  object  of  thoroughly  investigating  moral 
questions,  he  read  the  English  moralists  :  Shaftesbury, 
Hutcheson    and    Hume.      Shortly   afterwards,    about 
1762,  he  became  acquainted  not  only  with  the  moral 
but  also  with  the  metaphysical  theories  of  Hume.    This 
initiation  proved  to  be  a  psychological  moment  in  the 
development  of  his  thought.     "  Hume  was  the  first," 
he  says,  "  to  shake  me  out  of  my  dogmatic  indolence 
and  start  me  on  a  fresh  line  of  investigation  in  the 
domain  of  speculative  philosophy."     He  adds  immedi- 
ately afterwards  :    "  Of  course,    I  was  careful  not  to 
accept  his  conclusions."     To  his  mind,  Hume's  skepti- 
cism was  adequately  refuted  by  the  reality  of  moral 
action.       His   object   was   to   do  justice   to    Hume's 
criticisms  in  so  far  as  they  were  well  founded  without 
agreeing  with  his  conclusions,  to  steer  his  course  safely 
between  the  Scylla  of  skepticism  and  the  Charybdis  of 
dogmatism.     A  slight  clue  which  he  found  in  Locke 
(book  4,  chap.  3,  §  9,  etc.)  proved  the  starting  point 
of  his  own  theory.     And  so  Hume's  influence,  though 
certainly  considerable,  manifested  itself  in  Kant  as  a  note 
of  warning  or  a  stimulus  for  reflection.    There  is  no  proof 
that  Kant  passed  through  a  phase  of  skepticism  ;  it  was 
to  escape  from  Hume's  skepticism  that  he  sought  to 
take  a  stand  outside  traditional  dogmatism. 

It  may  be  that  his  transcendental  idealism  drew  its 
inspiration  from  the  teaching  of  Leibnitz,  now  set  forth 
in  all  its  purity  in  the  New  Essays,  which  appeared  in 
1765.  Leibnitz  demonstrated  how  the  principle  of 
innateness  may  be  held,  whilst  considering  experience 


KANT  261 

as  indispensable  to  the  formation  of  knowledge.  Kant's 
forms  and  categories,  however,  are  quite  different  from 
the  Leibnitzian  virtualities. 

To  become  an  ordinary  professor,  Kant  wrote  and 
defended  a  dissertation  in  Latin  on  the  form  and  prin- 
ciples of  both  the  sensible  and  the  intelligible  worlds 
(1770).  He  was  appointed  to  the  University  of 
KOnigsberg  by  Frederick  II.,  at  a  salary  of  400  thalers 
(60  pounds  sterling).  From  that  time  he  refused  all 
invitations  from  other  Universities.  He  now  lectured 
publicly  only  on  logic  and  metaphysics,  and  privately 
on  natural  law,  morals,  natural  theology,  anthropology 
and  physical  geography.  His  ability  as  a  professor  was 
wonderful ;  he  did  not  teach  his  pupils  philosophy,  he 
rather  trained  them  to  become  philosophers.  His  lessons 
were  simple,  clear  and  attractive ;  he  reserved  all 
abstruse  deductions  and  special  terminology  for  his 
books  intended  for  scholars.  On  moral  subjects  he 
spoke  with  warmth  and  conviction  ;  his  eloquence  was 
virile,  leaving  a  profound  impression  on  the  souls  of  his 
hearers. 

The  problem  of  the  criticism  of  human  knowledge  was 
not  long  before  it  captivated  him.  How  can  we  explain 
why  ideas,  conceived  of  a  priori,  conform  with  things 
that  exist  outside  of  ourselves  ?  At  first  he  thought  he 
would  be  able  to  answer  the  question  in  a  few  months  : 
he  spent  twelve  years  on  it.  Even  then  he  allowed 
himself  only  four  or  five  months  to  put  his  thoughts 
into  words,  for  fear  of  delaying  the  solution  too  long. 
It  was  at  Riga,  in  the  beginning  of  178 1,  that  the  Critique 
of  Pure  Reason  appeared,  one  of  the  pillars  of  human 
thought.  Kant  was  then  fifty-seven  years  of  age.  The 
originality  and  purport  of  his  book  were  not  at  first 


262  STUDIES  IN  PHILOSOPHY 

understood.  No  one  cared  to  regard  him  as  anything 
else  than  a  Platonic  dreamer  or  a  Cartesian  idealist; 
Hamann  called  him  a  Prussian  Hume.  Kant  stoutly 
explained  his  position  in  a  treatise  entitled  :  Prolegomena 
to  all  Future  Metaphysic  that  may  present  itself  as  Science 
(1783),  and  also  in  the  preface  to  the  second  edition  of 
the  Critique  (1787).  Sure  of  his  principles,  he  con- 
centrated his  efforts  more  and  more  exclusively  on 
developing  their  consequences,  finishing  his  work  of 
criticism,  and  establishing  on  this  basis  a  complete 
doctrine  of  speculative  and  moral  philosophy.  His 
writings  devoted  to  this  task  appeared  between  1785 
and  1797. 

His  reputation  began  to  increase.  In  1790,  Fichte, 
then  quite  a  young  man,  forwarded  him  his  Aphorisms 
on  Religion  and  Deism,  along  with  an  enthusiastic  letter. 
Schiller  studied  his  teachings  on  esthetics  and  induced 
Goethe  to  do  the  same.  J.  P.  Richter  recorded  his 
opinion  that  Kant  was  not  so  much  a  light  of  the 
world  as  an  entire  system  of  dazzling  suns.  His  fame 
spread  to  Holland  and  England.  His  dissertation  on 
eternal  peace,  published  in  1795,  was  translated  into 
French. 

The  government  accorded  him  its  esteem  and  pro- 
tection. Once  only  was  he  near  receiving  a  check  in 
the  promulgation  of  his  doctrines  :  when  writing  on 
religious  subjects.  In  1792  he  had  sent  an  article  on 
the  root  evil  in  human  nature  to  the  Berlin  Monthly 
Review,  and  the  Board  of  Censors  had  authorized  its 
insertion.  A  second  article,  however,  on  the  struggle 
between  the  good  and  the  evil  principles,  was  not 
accepted.  Now,  Kant  had  still  two  more  to  bring  out. 
Refused  by  the  Board,  he  applied  to  the  Faculty  of 
Theology,  who  granted  the  imprimatur.  The  four 


KANT  263 

dissertations  appeared  under  the  title,  Religion  within 
the  Limits  of  Reason  alone  (1793).  The  government 
grew  alarmed  at  the  success  of  the  book,  and  on 
the  ist  of  October  1794  Kant  received  a  letter  asking 
for  an  explanation  and  commanding  him  never  again 
to  write  on  the  subject  of  religion.  Outwardly  Kant 
yielded,  and  gave  a  written  promise  not  to  teach  or 
write  on  religion  "  as  a  loyal  subject  of  His  Royal 
Majesty."  When  the  king,  however,  died  in  1797, 
he  regarded  himself  as  released  from  his  promise. 

In  other  respects  he  lived  quietly  enough,  though  he 
was  very  sympathetic  towards  the  French  Revolution. 
This  sympathy  is  a  special  characteristic  of  his  moral 
make-up.  He  looked  upon  the  Revolution  as  an  effort 
to  establish  the  organisation  of  human  societies  on 
reason.  Even  after  1794  he  persevered  in  his  political 
convictions  though  he  despaired  of  a  favourable  issue  to 
events  in  France  itself.  To  the  very  end  he  believed 
in  the  justice  and  practical  value  of  theory  ;  in  right,  as 
a  principle  ;  and  in  eternal  peace,  as  the  practicable  goal 
of  politics.  Behind  personal  disputings  he  saw  the 
conflict  between  history  and  philosophy,  between  the 
positive  and  the  rational ;  in  all  things  he  relied  on  the 
triumph  of  reason. 

After  the  year  1790  his  intellectual  powers  began  to 
decline,  and  in  1797  he  resigned  his  professorship.  All 
the  same,  he  continued  to  work  right  on  to  the  end. 
The  book  on  which  he  was  engaged  was  to  be  his 
chef-d" ceuvre,  his  object  being  to  explain  the  transition 
from  the  metaphysics  of  the  science  of  nature  to  physics. 
This  work,  which  he  left  unfinished,  was  lost ;  it  has 
been  found  recently.  Kant's  last  year  of  life  was 
marked  by  ever -increasing  feebleness  of  body.  He 
died  on  the  I2th  of  February  1804.  His  last  words 


264  STUDIES  IN  PHILOSOPHY 

were  :  Es  ist  gut  (It  is  well).  His  funeral  took  place 
amid  universal  homage  and  admiration.  The  body  was 
interred  beneath  the  arcades  of  Konigsberg  Cathedral. 
Several  statues  were  erected  in  his  honour,  the  most 
famous  being  the  one  by  Rauch  in  Konigsberg.  Kant 
was  a  man  of  small,  short  stature,  only  a  little  over  five 
feet  in  height,  with  poorly  developed  bones  and  muscles, 
a  narrow,  almost  concave  chest,  the  right  shoulder  joint 
slightly  displaced,  high  forehead  and  fine  blue  eyes.  A 
cast  of  his  head  was  taken  by  Knorr,  and  his  remains 
were  exhumed  in  1880. 

Kant  lived  for  philosophy  alone.  He  held  no 
political  office  and  never  married.  All  the  same,  he 
did. not  consider  it  possible  to  be  a  philosopher,  without 
at  the  same  time  being  a  man,  and  so  regarded  it  as 
necessary  to  come  into  contact  with  the  realities  of  life 
before  attempting  to  understand  and  regulate  them. 
In  his  loftiest  aspirations  he  was  careful  not  to  overstep 
the  limits  of  this  terrestrial  world  of  ours.  His  object 
was  to  live  here  below  in  accordance  with  his  own 
principles,  which  he  looked  upon  as  absolute  and  followed 
out  to  the  letter.  To  his  mind,  the  reconciliation  of 
law  and  independence  was  to  be  found  in  reason  ;  by 
it  he  determined  to  form  his  opinions  and  control  his 
life.  In  politics  he  professed  liberalism,  but  would 
not  admit  of  any  separation  between  liberty  and  order, 
whilst  he  maintained  a  conscientious  respect  for  estab- 
lished power.  In  religion  he  was  a  rationalist,  though 
he  deemed  it  right  to  uphold  the  spirit  of  Christianity, 
and  valued  the  work  done  by  the  positive  religions.  In 
philosophy  he  attacked  dogmatism,  though  rejecting 
skepticism.  In  morals  he  repudiated  all  exterior  laws, 
though  obeying  an  interior  command  of  greater  severity 


KANT  265 

than  the  laws  he  rejected.  Boldness  in  speculation, 
respect  towards  practical  life  and  the  material  order  of 
things  :  such  were  his  distinguishing  traits. 

Kant  was  a  thinker  more  than  a  writer.  Some  of 
his  earlier  works  such  as  the  Observations  on  the  Beautiful 
and  the  Sublime ',  the  Methodology  of  the  Critique  of  Pure 
Reason,  and,  speaking  generally,  the  passages  in  which 
he  expresses  his  moral  convictions,  manifest  a  facile, 
pleasing  and  vigorous  style.  In  metaphysical  analysis, 
however,  his  style  is  complicated,  laboured  and  redun- 
dant, and  often  only  the  more  obscure  from  the  fact  that 
the  author  has  made  every  effort  to  be  clear.  Kant's 
work  is  a  thought  seeking  its  form.  In  more  finished 
shape,  would  it  have  stirred  the  human  intellect  to  the 
same  degree  ? 

The  following  is  a  chronological  list  of  Kant's 
principal  works,  written,  for  the  most  part,  in  German  : 

Thoughts  on  the  True  Estimate  of  Living  Force  (Vis  Fiva}^ 
and  an  Investigation  into  the  Proofs  of  Leibnitz  and  other 
Mechanical  Philosophers  thereon  (1747).  Kant,  in  this 
work,  reconciles  the  doctrines  of  Descartes  and  Leibnitz 
with  each  other,  as  regards  the  measurement  of  the  force 
of  a  moving  body. 

Has  the  Earth^  from  its  Origin^  undergone  any  Modifications  in 
its  Rotatory  Motion  ?  (magazine  article,  1754).  Relying 
on  Newton's  principles  Kant  clearly  shows  that  the  speed 
of  the  earth's  rotation  must  have  diminished. 

Is  the  Earth  growing  Old  ?  A  research  made  from  the  physical 
standpoint  (article,  1754). 

A  Universal  History  of  Nature  and  Theory  of  the  Heavens^ 
dealing  with  the  System  and  Mechanical  Origin  of  the 
Universe^  in  accordance  with  Newton's  Principles  (1755),  a 
famous  work  that  appeared  anonymously,  with  a  dedica- 
tion to  Frederick  II.,  and  serving  as  a  kind  of  prelude  to  the 
exposition  of  the  world-system,  published  by  Laplace  in  1 796. 


266  STUDIES  IN  PHILOSOPHY 

Brief  Account  of  some  "Thoughts  on  Fire  (in  Latin)  (1755). 
Heat,  like  light,  is  a  vibratory  movement  of  the  ether. 

A  New  Explanation  of  the  First  Principles  of  Metaphysical 
Knowledge  (1755),  a  thesis  in  Latin;  written  to  obtain 
the  right  to  be  appointed  privatdocent.  It  deals  with  the 
principles  of  contradiction  and  determinative  reason. 

Three  dissertations  On  Earthquakes  that  took  place  at  £>uito  and 
Lisbon  in  1755. 

Physical  Monadology  (1756),  Latin  thesis;  Kant  defended  this 
thesis  with  a  view  to  a  professorship  which  he  did  not 
obtain.  In  it  he  transformed  the  monad  of  Leibnitz  into 
a  physical  atom. 

Explanatory  Remarks  on  the  Theory  of  the  Winds  (1756),  a 
precise  explanation  of  periodical  winds. 

A  New  Conception  of  Motion  and  Rest  (1758). 

A  Few  Thoughts  on  Optimism  (1759).  Kant  claims  that 
everything  is  good  if  we  regard  things  as  a  whole.  At 
the  end  of  his  life  he  repudiated  this  work,  inspired  by 
Leibnitz. 

The  False  Subtlety  of  the  Four  Syllogistic  Figures  (1762).  The 
first  figure  alone,  he  affirmed,  was  pure  and  primitive. 

An  Attempt  to  Introduce  into  Philosophy  the  Notion  of  Negative 
Quantities  (1763).  Real  opposition,  in  which  the  two 
terms  are  equally  positive  in  themselves  cannot  be  reduced 
to  logical  opposition,  in  which  one  of  the  terms  contradicts 
the  other. 

The  only  possible  Foundation  for  a  Demonstration  of  the  Exist- 
ence of  God  (1763).  The  possible,  regarded  not  in  its 
form  but  in  its  matter  or  data,  presupposes  existence,  and, 
in  the  final  analysis,  the  existence  of  a  necessary  being. 

An  Essay  on  the  Evidence  of  the  Fundamental  Propositions  of 
Natural  Theology  and  Ethics  (1764),  a  work  written  for  a 
competition  inaugurated  by  the  Berlin  Academy.  Kant 
obtained  only  the  accessit^  the  prize  being  awarded  to 
Mendelssohn.  Both  contrast  philosophy  with  mathe- 
matics, and  Kant  concludes  that  the  methods  employed  in 
the  latter  are  not  suitable  for  the  former. 

Observations  on  the  Sentiment  of  the  Beautiful  and  the  Sublime 
(1764) ;  a  work  on  morals  and  criticism. 


KANT  267 

Programme  of  Classes  for  the  Winter  Session  (1765-17  66).  The 
education  of  the  various  faculties  of  the  mind  should 
precede  the  acquiring  of  knowledge.  In  this  treatise  a 
critical  propensity  begins  to  show  itself. 

Dreams  of  a  Spirit-seer  (or  Clairvoyant]  explained  by  the  Dreams 
of  Metaphysic  (i  766,  anonymous).  This  work  was  inspired 
by  Swedenborg's  visions.  Kant  here  appears  in  a  skeptical 
and  somewhat  inconsiderate  vein,  a  la  Voltaire.  The 
only  difference  between  illuminism  and  metaphysics,  to  his 
mind,  is  that  the  former  is  the  dream  of  sentiment,  the 
latter  that  of  reason  ;  one  is  no  better  than  the  other. 
Let  us  not  claim  to  know  the  unknowable. 

Grounds  for  distinguishing  Positions  in  Space  (1768).  A  refuta- 
tion of  the  Leibnitzian  theory  which  posits  things  before 
space,  this  latter  being  reduced  to  nothing  but  a  concept. 
According  to  Kant,  we  must  admit  the  existence  of 
universal,  absolute  space. 

Form  and  Principles  of  the  Sensible  and  the  Intelligible  Worlds 
(1770),  a  dissertation  in  Latin,  written  in  order  to 
obtain  the  right  of  being  appointed  professor  of  logic 
and  metaphysics.  Kant  breaks  away  from  dogmatism  as 
regards  sensible — though  not  intelligible — knowledge. 

Letters  to  Marcus  Herz^  from  1770  to  1781.  Kant  endeavours 
to  find  some  mean  between  idealism  and  realism. 

The  Different  Human  Races.  The  races  are  varieties  that 
have  become  stable.  A  true  history  of  natural  beings 
would  doubtless  reduce  many  so-called  species  to  the 
position  of  simple  races,  the  offshoot  of  one  common 
species. 

The  Critique  of  Pure  Reason  (1781).  Theoretical  knowledge 
implies  both  intuition  and  necessary  connection.  As  we 
can  realise  the  first  condition  only  with  regard  to  sensible 
things,  these  latter  are  the  only  ones  we  can  know 
theoretically.  In  1787,  Kant  published  a  second  edition 
of  the  Critique.  Whether  the  changes  in  this  second 
edition  refer  to  the  substance  or  only  to  the  form  is  a 
much-disputed  question.  Rosenkranz,  Schopenhauer  and 
Kuno  Fischer  agree  that  a  thorough  modification  took 
place,  tending  to  restore  the  "  thing-in-itself,"  which, 


268  STUDIES  IN  PHILOSOPHY 

they  alleged,  the  first  edition  had  abolished.  According 
to  Kant  himself,  the  second  edition  merely  emphasises 
the  realistic  side  of  the  doctrine,  an  aspect  that  had  been 
disregarded  by  certain  readers.  Kant's  affirmation  may 
very  well  be  maintained.  The  first  edition  did  not  abolish 
the  "  thing-in-itself,"  but  rather  the  theoretical  knowledge 
of  the  u  thing-in-itself,"  a  very  different  matter. 

Prolegomena  to  all  Future  Metaphysics  which  may  present  itself 
as  Science  (1783).  This  short  work  gives  an  analytical 
exposition  of  the  doctrine  which  the  Critique  of  Pure 
Reason  had  set  forth  synthetically,  and  rectifies  the  mistakes 
made  with  reference  to  certain  points  in  this  doctrine. 

Notion  of  a  Universal  History  in  a  Cosmopolitan  Sense  (magazine 
article,  1784). 

Answer  to  the  question :  What  is  enlightenment  ?  (magazine 
article,  1784).  Enlightenment,  says  Kant,  is  the  emanci- 
pation of  the  intelligence. 

An  Account  of  Herder  s  work  entitled :  Ideas  on  the  Philosophy  of 
the  History  of  Mankind  (magazine  article,  1785).  Kant 
rejects  the  doctrine  of  the  essential  unity  of  nature  and 
freedom. 

Foundations  of  the  Metaphysics  of  Morals  (1785;  4th  edition, 
1797).  Here  Kant  determines  and  affirms  the  funda- 
mental principle  of  morality. 

Metaphysical  Elements  of  Natural  Science  (1786  ;  3rd  edition, 
1800).  In  this  work  the  axioms  of  pure  physics  are  given. 

Conjectures  on  the  Beginning  of  Human  History  (1786). 

Corporeal  Medicine  in  so  far  as  it  comes  under  Philosophy ',  a 
discourse  in  Latin  (1786  or  1788). 

The  Employment  of  Teleological  Principles  in  Philosophy  (article, 
1788). 

Critique  of  Practical  Reason  (1788;  6th  edition,  1827).  A 
determination  of  the  nature  of  the  moral  law,  and  of  the 
kind  of  adhesion  that  practical  principles  allow  of. 

Critique  of  Judgment  (1790;  3rd  edition,  1799).  Here  Kant 
deals  with  the  basis  and  value  of  the  ideas  of  beauty  and 
finality. 

Illuminism  and  the  Remedies  against  it  (1790).  Dissertation  on 
Cagliostro. 


KANT  269 

The  Failure  of  all  Philosophic  Effort  in  Theodicy  (1791). 

Religion  within  the  Bounds  of  Reason  only  (1793  ;  2nd  edition, 
1794).  The  deduction  or  legitimation  of  religion.  Only 
what  relates  to  morals  is  founded  on  religion.  We  must 
tend  to  make  religion  purely  rational. 

The  Commonplace  Remark :  "  That  is  all  Right  in  Theory  but 
Worthless  in  Practice"  (magazine  article,  1793).  Kant 
rejects  this  well-known  aphorism  not  only  as  regards 
morality,  but  also  with  reference  to  political  and  human 
right. 

The  Influence  of  the  Moon  upon  the  Weather  (article,  1 794). 

Eternal  Peace,  a  Philosophical  Essay  (1795).  Eternal  peace 
Kant  regards  as  the  goal  of  the  historic  development  of 
mankind,  and  that,  not  from  sentiment,  but  from  the  idea 
of  justice. 

Metaphysical  Principles  of  the  Theory  of  Right  (1797  ;  2nd 
edition,  1798).  The  theory  of  right  or  legality  as  deduced 
from  the  criticism  of  practical  reason. 

Metaphysical  Principles  of  the  Theory  of  Virtue  (1797  ;  2nd 
edition,  1803).  The  theory  of  morality,  also  as  the  result 
of  criticism.  These  two  latter  works  bear  the  title  : 
Metaphysics  of  Morals. 

Contest  of  the  Faculties  (to  this  work  is  added  an  article  that 
appeared  in  1797  :  The  Power  of  the  Mind  to  Master  its 
Morbid  Feelings  by  Will  alone]  (i  798).  This  was  the  conflict 
of  the  Faculty  of  Philosophy,  representing  rational  truth, 
with  the  three  other  Faculties :  Theology,  Law,  and 
Medicine,  representing  the  positive  disciplines. 

Anthropology  Treated  from  the  Pragmatic  Point  of  View  (1798  ; 
2nd  edition,  1800).  Pragmatic  anthropology  is  the  art  of 
using  men  for  one's  own  purposes. 

Logic,  a  work  published  by  Jasche  (1800). 

Physical  Geography,  published  by  Rink  (1802-1803). 

Paedagogics,   published  by  Rink  (1803).      Notes   taken  from 

several  lectures  delivered  by  Kant  on  this  subject. 
Transition  from  the  Metaphysical  Principles  of  the  Science  of 
Nature  to  Physics,  an  unfinished  work,  written  between 
1783  and   1803,  first  published  by  Reicke  between  1882 
and   1884  in  the  Altpreussische  Monatsschriften,  and  then, 


270  STUDIES  IN  PHILOSOPHY 

more  completely,  by  Albrecht  Krause  (1888).  Here  we 
have  the  progress  of  deduction  proceeding  from  the  meta- 
physics of  material  nature  to  experimental  physics  regarded 
as  a  science,  i.e.  as  a  system. 

Kant's  Reflections  on  Critical  Philosophy^  published  by  Benno 
Erdmann  (1882-1884). 

Letters.  About  a  hundred,  nineteen  of  which  were  addressed 
to  Marcus  Herz. 


II. — THE  ANTECRITICAL  PERIOD 

On  the  2Oth  of  August  1777  Kant  left  it  on  record 
that  his  investigations,  hitherto  professional  and  frag- 
mentary, had  finally  taken  systematic  form  and  brought 
him  to  the  idea  of  the  whole.  Thus,  in  the  first  place, 
the  development  of  Kantian  thought  shows  a  long 
period  of  formation,  during  which  works  of  different 
kinds  were  undertaken  for  themselves  alone  without 
reference  to  a  general  standpoint,  and  afterwards 
brought  together  with  a  view  to  being  reconciled  with 
each  other.  And  so  Kant's  thought  progresses  from 
the  parts  to  the  whole.  His  main  idea  is  arrived  at  by 
a  process  of  synthesis.  This  first  period  extends  to 
the  time  of  his  elaboration  of  criticism,  i.e.  up  to  and 
including  the  year  1770. 

The  starting-point  of  Kantian  thought  is,  on  the 
one  hand,  a  substratum  of  Christian,  and  more  especi- 
ally of  Pietistic  beliefs,  faith  in  duty,  the  cult  of  moral 
intention,  conviction  of  the  superiority  of  practice  over 
dogmatism  ;  on  the  other,  a  very  clear,  keen  sense  of 
science,  the  determination  to  be  guided,  so  far  as  a 
knowledge  of  nature  is  concerned,  only  by  the  evidence 
of  experience  and  mathematical  reasoning.  Hence- 
forth Kant  is  principally  concerned  with  the  connec- 
tion between  science  and  religion  ;  this,  too,  after  both 


KANT  271 

have  been  developed  in  his  mind  independently,  each 
according  to  its  own  method. 

During  the  antecritical  period  Kant  meditates  in 
turn  on  the  divers  objects  presented  both  by  his  studies 
and  by  the  circumstances  of  life. 

From  1747  to  1755  he  is  a  Leibnitzo-Wolfian, 
though  with  a  tendency  to  accentuate  the  difference 
between  the  mathematical  and  the  real. 

With  Newton,  he  studies  the  mechanism  of  the 
heavens,  from  1754  to  1763.  Like  him,  he  determines 
to  employ  experience  only  in  conjunction  with  mathe- 
matics. Newton,  however,  did  not  state  the  problem 
of  origins.  Kant  believes  that  the  method  which  has 
established  the  present  system  is  capable  of  going 
back  to  the  genesis  of  this  very  system  :  the  forces  that 
preserve  must  also  be  the  forces  that  have  created. 
He  undertakes  to  trace  not  only  the  possible  but  the 
real,  the  actual  history  of  the  formation  of  the  world. 

In  the  beginning  was  one  homogeneous,  elementary 
matter,  moved  by  forces  of  attraction  and  repulsion  :  a 
gaseous  chaos.  This  matter  was  maintained  in  a  state 
of  extreme  tenuity  by  being  kept  at  a  very  high 
temperature.  In  obedience  to  the  forces  contained 
within  itself  this  chaos  is  subjected  to  a  movement  of 
rotation.  Purely  as  the  result  of  these  physical  condi- 
tions the  homogeneous  becomes  differentiated.  Rota- 
tion occasions  the  formation  of  nebulae,  which  themselves 
acquire  a  rotatory  movement.  In  turn  these  nebulae, 
as  the  result  of  the  centrifugal  force,  produce  rings,  and 
these  rings  represent  the  orbits  of  future  planets.  Then 
the  rings  break,  and  collect  together  again  in  planets. 
Satellites  are  formed  in  the  same  way. 

The  scientific  value  of  this  theory  is  now  recognised 
even  by  such  men  as  Helmholtz  (Mdmoire  sur  la  conser- 


272  STUDIES  IN  PHILOSOPHY 

vation  de  la  force •,  1847)  anc^  Fave  (Revue  scienttfique^ 
1884). 

The  theory  was  the  result  of  purely  scientific  con- 
siderations. Kant,  however,  at  once  confronts  it  with 
the  teachings  of  religion.  Religion,  he  says,  has 
nothing  to  fear  from  a  doctrine  which,  though  dis- 
missing accidental  and  extrinsic  finality,  as  met  with  in 
the  works  of  men,  implies,  on  the  other  hand,  a  fruitful 
and  essential  finality,  which  alone  is  truly  worthy  of 
God.  Besides,  who  will  ever  be  able  to  say,  "  Give 
me  matter  and  motion  and  I  will  make  a  snail  "  ?  At 
its  very  lowest  stage  life  is  immeasurably  superior  to 
mechanism  ;  it  is  a  witness  to  God. 

Following  on  Wolf,  Kant  studies  the  relations 
between  possibility  and  existence  (1755).  The  prin- 
ciple of  contradiction  is  the  law  of  the  former  ;  the 
principle  of  determining  reason,  irreducible  to  that  of 
contradiction,  is  the  law  of  the  latter.  Determining 
reason  is  either  antecedently  determining  and  a  reason 
of  existence,  or  subsequently  determining  and  a  reason 
of  knowledge.  Antecedently  determining  reason  alone 
gives  us  complete  science.  From  these  principles  Kant 
deduces  the  impossibility  of,  explaining,  solely  by  the 
analysis  of  their  distinctive  essence,  either  change  or 
the  real  connection  between  substances*  All  rela- 
tions between  substances  must  come  from  without. 
Thus,  succession  has  its  foundation  in  an  external 
action  which  constitutes  the  reality  of  the  world,  whilst 
coexistence  is  based  on  an  extrinsic  connection  which 
implies  the  existence  of  God.  And  so,  speculating 
on  Wolr's  metaphysics,  Kant  ends  in  a  deduction  of  the 
principles  of  Newtonianism.  His  system  at  this  period 
may  be  defined  as  realistic  mechanism  dependent  on 
natural  theology. 


KANT  273 

Dealing  with  the  relations  between  philosophy  and 
mathematics,  as  did  his  contemporaries  (1756-1764), 
Kant  neither  admits  that  the  concepts  of  the  mathema- 
ticians, infinite  divisibility,  an  absolute  plenum,  the 
exclusive  mechanism  of  all  notion  of  force,  are  intel- 
ligible to  the  understanding,  nor  that  these  concepts 
are  meaningless  and  devoid  of  real  value.  Though  a 
stumbling-block  to  the  logician,  mathematics  is  none 
the  less  the  key  of  the  science  of  nature,  as  Newton 
proved.  The  problem  is  to  reconcile  mathematics 
with  transcendental  philosophy,  not  to  sacrifice  the  one 
to  the  other.  Now,  if  we  analyse  the  conditions  of 
mathematical  speculation,  and  those  of  philosophic 
speculation,  we  find  that  in  both  cases  the  object  is  a 
synthesis,  but  that  in  the  former  it  is  built  up  by  the 
mind,  and  in  the  latter  it  is  given  to  the  mind.  Hence, 
the  method  that  suits  the  one  is  useless  for  the  other. 
Everything  referring  to  dimension  will  be  dealt  with 
mathematically,  but  if  we  would  know  qualities  and 
existences,  we  shall  use  experience  and  metaphysical 
systematisation,  along  with  Newton.  There  are  two 
certainties,  two  outlooks  upon  nature  :  that  of  mathe- 
matical proof  and  that  of  experience.  These  two 
paths  of  knowledge,  starting  from  opposite  poles,  can 
never  meet. 

Yielding  to  the  influence  of  the  aesthetician  Baum- 
garten,  Rousseau  and  the  English  philosophers,  Kant 
takes  up  the  questions  of  taste  and  morals  (1763-1766). 
His  method  consists  in  taking,  as  his  starting-point, 
an  impartial  observation  of  human  nature.  We  must 
proceed,  he  says,  from  what  is  to  what  ought  to  be. 
His  observation,  however,  in  spite  of  himself,  is  tinged 
with  metaphysical  analysis.  In  the  given  he  is  about 
to  discover  the  absolute.  What  he  thinks  he  ought 

T 


274  STUDIES  IN  PHILOSOPHY 

to  observe  is  not  so  much  ideas  and  things  as  the  inner 
movements  of  sensibility.  Taking  this  point  of  view, 
he  is  led  to  make  a  profound  distinction  between  the 
beautiful  and  the  sublime.  This  distinction  introduces 
enlightenment  and  precision  into  literature  and  art. 
Thus,  it  is  the  province  of  tragedy  to  be  sublime,  that 
of  comedy  to  be  beautiful.  The  distinction  likewise 
applies  to  morals.  True  virtue  is  sublime  ;  good  quali- 
ties :  a  kind  heart,  the  sense  of  honour,  modesty,  are  only 
beautiful.  The  spring  of  virtue  is  the  sentiment  of  the 
beauty  and  dignity  of  human  nature,  regarded  as  a 
motive  of  action.  This  principle  must  be  understood 
in  a  formal  sense :  it  consists  essentially  of  an  obligatory 
rule.  This  principle,  too,  is  impossible  of  demonstra- 
tion ;  and  it  is  good  that  it  should  be  so.  Providence 
has  not  willed  that  knowledge  indispensable  to  our 
happiness  should  depend  on  subtle  reasoning  ;  it  has 
entrusted  such  knowledge  to  natural  common  sense. 

Swedenborg's  claim  that  he  held  direct  communica- 
tion with  spirits,  affords  Kant  an  opportunity  to  examine 
the  value  of  metaphysics,  so  far  as  it  also  affirms  the 
possibility  of  becoming  acquainted  with  suprasensible 
existences  (1763-1766).  Metaphysics  seems  to  meet 
with  unexpected  confirmation  in  the  facts  affirmed  by 
illuminism.  It  is  apparently  justified  by  the  theory  it 
advances  thereof,  as  Newtonianism  is  justified  by  the 
explanation  it  affords  of  the  experimental  laws  of  motion. 
Unfortunately,  illuminism  can  be  explained  in  a  far 
simpler  and  more  satisfactory  manner,  as  hallucination 
caused  by  certain  organic  disturbances.  Might  it  not 
then  possibly  happen  that  metaphysics  had  a  like  origin  ? 
What  if  it  were,  after  all,  a  mere  hallucination  of  the 
understanding,  endowing  the  phantoms  of  sensible 
hallucination  with  an  apparently  logical  existence  ?  All 


KANT  275 

the  same,  we  must  beware  of  leaping  to  the  conclusion 
that  metaphysics  is  altogether  inane.  In  one  scale  of 
the  balance  it  places  the  hope  of  a  future  life.  Now, 
we  could  not  will  that  this  weight  remain  actionless  on 
our  mind.  What  we  do  know,  is  that  we  can  expect 
nothing  from  experience  calculated  to  confirm  our  moral 
and  religious  beliefs.  But  these  beliefs  need  no  experi- 
mental confirmation  ;  they  both  will  and  ought  to  be 
free.  In  a  word,  the  result  of  our  examination  is  that 
we  must  offer  the  following  new  definition  of  meta- 
physics, one  alike  favourable  to  practice  and  imposed 
upon  theory  :  metaphysics  is  the  science  of  the  limits 
of  human  reason. 

Kant,  like  Leibnitz,  studies  the  nature  of  space  and 
time  (1768-1770).  Several  facts  of  experience,  includ- 
ing the  real  existence  of  symmetrical  figures,  prove  that 
geometrical  space  is  no  mere  consequence  of  the  relations 
between  things  in  point  of  position,  but  rather  the  basis 
of  the  possibility  of  these  relations.  The  reality  of 
absolute  space  being  thus  established,  Kant  asks  himself 
how  space  is  possible,  i.e.  conceivable  without  contra- 
diction. Space  and  time  are  known  a  priori ;  at  the 
same  time  they  are  intuitions.  How  can  these  two 
characteristics  be  reconciled  ?  The  only  way  of  doing 
so,  is  to  regard  space  and  time  as  the  conditions  imposed 
on  the  human  mind,  by  its  very  nature,  for  the  percep- 
tion of  sensible  objects.  Space  and  time  do  not  concern 
things  as  they  are  in  themselves,  but  only  as  they 
appear  to  our  sensibility.  The  "  critique "  idea  has 
come  to  birth,  but  Kant  applies  it,  so  far,  only  to 
sensible  or  mathematical  knowledge. 

It  was  through  Hume's  influence  that  Kant's  thought, 
which  had  hitherto  wandered  over  all  kinds  of  subjects, 
was  finally  to  become  concentrated  and  steadied  (1762- 


276  STUDIES  IN  PHILOSOPHY 

1780).  Hume's  dialectic  made  such  an  impression  on 
Kant's  mind  that  he  soon  thought  of  nothing  else  than 
solving  the  difficulties  raised  by  the  famous  skeptic.  In 
this  task  his  true  originality  was  shown,  and  there 
blossomed  forth  the  idea  which  was  to  be  the  soul  of  his 
philosophy.  Kant  had  long  ago  pondered  on  the  rela- 
tion between  cause  and  effect,  he  soon  saw  the  element 
of  strangeness  in  a  connection  which  could  not  be 
analytical,  and  yet  was  necessary.  Still,  he  did  not 
think  of  criticising  its  legitimacy.  It  was  Hume  who 
roused  him  out  of  his  dogmatic  calmness,  proclaiming 
that  the  concept  of  causality — a  concept  foreign  to 
reason,  formed  by  nothing  but  imagination  on  the  occa- 
sion of  a  mere  habit  and  under  the  influence  of  some 
obscure  instinct — could  have  no  object  outside  of  our- 
selves. Kant  refused  to  follow  Hume  in  the  deductions 
the  latter  claimed  to  found  on  this  analysis  of  his. 
Indeed,  what  would  become  of  the  freedom  of  the  will, 
the  condition  of  moral  determination,  if  there  existed 
for  us  nothing  but  phenomena ;  and  what  would  become 
of  science  itself,  the  knowledge  of  things  as  necessary, 
if  causality  were  nothing  more  than  a  subjective  con- 
nection ?  In  Kant's  mind,  science  and  morals  are  given, 
as  are  also  the  characteristics  peculiar  to  them  ;  it  is 
the  part  of  philosophy  to  explain  their  possibility  or 
conditions,  not  to  discuss  their  reality. 

And  so  Hume's  thesis  was  not  a  doctrine,  but  rather 
a  problem,  a  starting-point  for  Kant.  How  comes  it 
to  pass  that  a  relation,  the  terms  of  which  are  hetero- 
geneous, should  also  be  posited  as  necessary  as  valid 
for  things  ?  This  was  the  question  to  be  answered. 

First,  he  had  to  satisfy  himself  that  the  principle  of 
causality  did  not  proceed  from  experience,  for  in  that 
event  its  necessity  would  have  been  radically  unintelli- 


KANT  277 

gible.  Having  noticed,  however,  that  many  other 
concepts,  such  as  those  of  substance,  mutual  action,  etc., 
held  the  same  position  as  the  one  with  which  Hume  had 
grappled,  and  having  succeeded  in  determining  the  exact 
number  of  these  concepts  by  means  of  a  single  principle, 
a  thing  impossible  in  the  case  of  concepts  of  experience, 
Kant  from  that  time  regarded  it  as  established  that  the 
concept  of  cause  may  be  acknowledged  to  have  an  origin 
a  priori.  And  yet,  can  there  conceivably  be  concepts 
that  are  at  once  a  priori  and  synthetical !  Have  we 
not  here  two  incompatible  characteristics?  This  was 
Hume's  idea,  and  so  he  gave  up  the  problem,  discarding 
causality  in  favour  of  experience.  The  reason  was  that 
he  shared  a  prevailing  error  of  the  age  upon  an  im- 
portant point  closely  connected  with  the  question  :  the 
nature  of  mathematical  judgments.  These  latter  he 
regarded  as  analytical,  and  so  refused  to  consider  them. 
In  reality,  they  are  synthetical  ;  and  as  their  character 
of  necessity  and  apriority  is  indisputable,  and  even  un- 
disputed, they  afford  an  instance  of  the  effective  union, 
within  our  knowledge,  of  apriority  and  synthetic  con- 
nection. There  is  nothing,  therefore,  to  prevent  the 
judgment  of  causality  from  being  both  synthetic  and 
necessary. 

Nevertheless,  it  is  not  enough  for  it  to  be  necessary 
in  the  sense  that  mathematical  judgments  are  necessary. 
Necessary,  in  the  sense  of  causal  connection,  means  : 
applicable  a  priori  to  real  things.  How  is  such  a 
property  possible  ?  If  objects  were  produced  by  the 
understanding,  or  ideas  by  objects,  the  agreement  between 
concepts  and  things  would  afford  no  difficulty ;  but  such 
is  not  the  case  :  mind  and  things  form  two  distinct 
worlds.  Then  how  does  the  mind  come  to  have  the 
right  to  dictate  laws  for  things  ?  It  acquires  this  right, 


278  STUDIES  IN  PHILOSOPHY 

answers  Kant,  from  the  conditions  of  experience  itself, 
both  inner  and  outer  :  no  other  explanation  is  possible. 
This  view,  the  origin  of  transcendental  deduction,  is 
the  goal  of  the  regressive  movement  occasioned  by 
Hume's  criticism.  It  includes  the  formula  of  Kant's 
criticism,  and  the  central  idea  of  the  system  he  is  now 
to  build  up. 

III. — CRITICISM 

The  Kantian  criticism  of  pure  reason  is  strictly  a 
theory  of  science.  As  Newton  sought  for  the  principle 
governing  the  system  of  celestial  bodies,  so  Kant  seeks 
for  the  principle  governing  the  system  of  our  knowledge. 
Science  is  given,  just  as  the  universe  is  given  ;  philosophy 
does  not  ask  whether  it  is  possible  or  not,  but  how  it 
is  possible,  i.e.  conceivable  without  contradiction. 

Science  consists  of  two  disciplines,  mathematics  and 
physics,  and  the  union  of  the  two  ;  we  must  take  these 
facts  into  account.  Mathematics  consists  of  a  priori 
synthetic  judgments,  i.e.  judgments  in  which  the  subject 
is  attached  a  priori  to  a  predicate  not  contained  in  it. 
It  is  the  same  with  physics.  Ever  since  the  time  of 
Newton,  the  certainty  of  physics,  which  deals  with 
things  themselves,  is  in  no  way  inferior  to  that  of 
mathematics,  which  deals  only  with  relations  of  dimen- 
sion. How  are  these  characteristics  intelligible,  whence 
do  they  proceed,  and  what  is  science,  considered  in  its 
generating  principles  ?  It  is  the  object  of  Kant's  in- 
vestigations to  answer  these  questions. 

And  it  is  the  province  of  philosophy  to  institute 
these  investigations.  Now,  the  inviolable  principle 
philosophy  gives  us  in  this  matter,  is  the  following  : 
all  our  knowledge  has  its  starting-point  in  experience. 
We  have  to  discover  if,  from  this  principle,  there  can 


KANT  279 

be  deduced  the  theory  of  science,  as  given  to  us. 
Thus,  the  problem  resolves  itself  into  the  following 
question  :  "  What  is  experience  ?  Is  it  an  irreducible 
unity,  or  can  analysis  discern  different  elements  in  it  ? 
Of  these  elements,  are  there  any  a  'priori  ?  Will  these 
a  priori  elements  account  for  the  necessity  proper  to 
the  judgments  of  science,  and  in  what  way  ? " 

In  experience,  an  object  is  first  given,  secondly 
thought.  How  is  that  possible  ? 

For  an  object  to  be  given  to  us,  it  must  be  presented 
in  space  and  time.  Are  the  notions  of  space  and  time 
supplied  by  experience  ?  No,  for  before  experience, 
we  know  that  the  objects  given  will  be  given  in  space 
and  time.  Consequently,  they  are  a  priori  elements. 
What  is  their  nature  ?  Are  they  concepts  ?  No,  for 
space  and  time  are  objects  that  are  integral,  homo- 
geneous, and  infinite,  characteristics  opposed  to  those 
offered  by  the  objects  of  the  concepts.  Space  and 
time  are  substrata  of  the  things  and  objects  of  intuition. 
Then,  are  they  suprasensible  realities  outside  of  our- 
selves ?  No,  for  the  conception  of  two  infinite  non- 
beings  as  substances  is  impossible.  After  all,  the 
representation  of  space  and  time  can  only  be  an  intuition 
resting  on  the  form  of  our  sensibility.  Space  and  time 
are  our  way  of  seeing  things. 

But  then,  if  such  is  the  case,  are  not  our  ideas  of 
place  and  duration  purely  subjective  ?  With  such 
a  doctrine,  what  is  to  become  of  the  truth  of  mathe- 
matics ? 

The  objection  is  groundless,  for,  as  a  matter  of  fact, 
it  is  in  dogmatic  theories,  isolating  the  sensible  from 
the  mathematical,  that  the  agreement  of  the  one  with 
the  other  is  undemonstrable.  Mathematics  is  justified 
if  regarded  in  its  true  nature  as  a  system  of  a  priori 


280  STUDIES  IN  PHILOSOPHY 

synthetic  judgments,  when  once  objects  are  capable  of 
affecting  us  only  by  becoming  subject  to  the  laws  of  time 
and  space.  Doubtless  we  cannot  say  that  things,  in 
themselves,  possess  modes  of  being  that  we  can  only 
explain  as  forms  of  our  power  to  feel.  But  we  know, 
a  priori^  that  every  object  of  our  sensibility  will  con- 
form with  mathematics,  and  that  is  sufficient  to  insure 
the  objectivity  of  this  science.  Transcendental  ideality 
and  empirical  reality  are  the  two  characteristics  of  time 
and  space.  They  explain  and  determine  the  possibility 
of  mathematics. 

This  is  the  explanation  of  the  first  condition  of 
experience  :  there  is  a  second.  For  an  object  to  be 
given  is  not  sufficient,  it  must  also  be  thought.  Does 
thought  imply  a  priori  elements  ? 

Thought  consists  in  setting  up  between  two  terms 
an  objective  relationship  of  subject  and  predicate,  i.e.  in 
affirming  that  the  one,  really  and  of  necessity,  belongs 
to  the  other.  This  is  what  takes  place,  for  instance, 
when  we  say  that  one  thing  is  the  cause  or  substance 
of  another.  Such  a  connection  cannot  be  supplied  by 
experience,  which  gives  nothing  necessary.  There- 
fore, it  is  known  a  priori,  though  in  what  way  ?  If  we 
consider  logic  as  it  has  been  conceived  of  ever  since  the 
time  of  Aristotle,  we  note  that  it  supplies  many  necessary 
connections,  but  yet  is  unable  to  determine  one  term  as 
being  a  real  subject  regarding  the  other.  In  every 
declaration  relative  to  existence,  there  is  something 
more  than  simple  logic.  To  affirm  of  an  object  that  it 
is  a  cause,  is  to  go  beyond  the  limits  of  its  concept. 
Now,  we  are  without  that  intellectual  intuition  of  the 
whole,  which  alone  would  enable  its  parts  to  be  dis- 
closed by  a  process  of  analysis.  We  proceed,  in  discur- 
sive fashion,  from  the  parts  to  the  whole.  On  what 


KANT  281 

principle,  then,  do  the  different  relations  that  constitute 
thought,  depend  ? 

Apart  from  those  we  have  had  to  reject,  there 
remains  only  the  understanding  itself,  or  the  faculty 
of  judging.  As  relations  of  dimension  are,  at  bottom, 
only  the  forms  of  our  sensibility,  so  qualitative  relations 
of  things  cannot  be  anything  else  than  the  categories  of 
our  understanding. 

If  this  is  so,  the  logical  function  of  the  under- 
standing will  enable  us  to  detect  and  systematise  all 
the  concepts  that  control  judgments  of  existence.  For, 
on  both  sides,  it  is  the  province  of  the  understanding 
to  unify  ;  the  extent  of  this  unification  alone  causes 
difference.  The  table  of  the  modes  of  logical  unification 
thus  supplies  a  model  for  the  table  of  categories. 

The  following  is  the  logical  table  of  judgments  : 
ist,  from  the  standpoint  of  quantity  :  universal,  par- 
ticular and  individual  propositions ;  2nd,  from  the 
standpoint  of  quality  :  affirmative,  negative  and  in- 
determinate propositions ;  3rd,  from  the  standpoint 
of  relation  :  categorical,  hypothetical  and  disjunctive 
propositions  ;  4th,  from  the  standpoint  of  modality  : 
problematical,  assertorial  and  apodeictic  propositions, 

The  following  is  the  transcendental  table  of  the 
concepts  of  understanding  :  ist,  from  the  standpoint  of 
quantity  :  unity,  multiplicity,  universality  ;  2nd,  from 
the  standpoint  of  quality  :  reality,  negation,  limitation  ; 
3rd,  from  the  standpoint  of  relation:  inherence  and 
subsistence,  causality  and  dependence,  reciprocal  action  ; 
4th,  from  the  standpoint  of  modality  :  possibility  or 
impossibility,  existence  or  non-existence,  necessity  or 
contingency. 

This  is  the  system  of  concepts  or  categories  by  whose 
aid  we  unite  our  representations  of  things.  As  these 


282  STUDIES  IN  PHILOSOPHY 

concepts  are  only  the  modes  of  action  of  our  under- 
standing, in  themselves  they  are  devoid  of  all  content. 
They  can  find  a  use  only  if  they  are  supplied  with 
matter,  and  the  only  matter  at  our  disposal  is  sensible 
intuition.  Have  concepts,  then,  only  a  subjective 
value ;  and  whereas  transcendental  esthetics  or  the 
analysis  of  sensibility  may  have  pronounced  for  mathe- 
matical realism,  will  the  analysis  of  the  understanding 
or  transcendental  logic  have  to  confine  itself  to  that 
logical  idealism  which  resolves  things  into  modes  of 
thought  ? 

Here  we  have  the  famous  transcendental  deduction 
whose  object  is  to  establish  the  objective  value  of  the 
categories,  i.e.  the  possibility  of  obtaining,  by  means 
of  the  categories  as  they  have  been  determined,  the 
knowledge  not  only  of  our  way  of  thinking,  but  of  the 
things  themselves.  This  possibility  will  be  demon- 
strated if  it  can  be  proved  that  the  categories  are  them- 
selves the  condition  of  the  existence  of  realities,  from 
our  standpoint.  Categories  apply  to  things,  if  things, 
to  us,  are  possible  only  by  their  means. 

According  to  our  condition,  in  order  that  there  may 
be  knowledge  of  a  thing,  there  must  be  distinction 
between  subject  and  object :  "  I  think  "  should  accom- 
pany all  our  representations.  For  such  a  condition  to 
be  possible,  however,  there  must  exist  between  the 
two  terms  a  relation  analogous  to  that  between  positive 
and  negative  quantities  in  mathematics,  a  relation  of 
opposition  on  a  common  ground.  The  subject  being 
a  unifying  action,  the  object  must  be  a  unified  multiple. 
And  so  it  is  because  things  are  unified,  and  unified  for 
the  subject,  that  they  can  be  presented  as  an  object. 

Now,  how  could  this  condition  be  satisfied,  were 
not  the  multiple  unified  by  the  subject  itself?  Doubt- 


KANT  283 

less  the  empirical  consciousness  does  not  perceive  this 
formation  of  the  object.  The  operation  takes  place  in 
the  depths  of  transcendental  apperception  implied  by  the 
empirical  consciousness  ;  and  when  the  particular  I  is 
posited,  it  finds  the  object  ready  formed  before  it,  and 
takes  it  for  a  brute  thing.  This  thing,  however,  is  the 
work  of  thought,  therefore  thought,  in  each  of  us, 
recognises  its  own  laws  therein.  Thus  the  categories 
are  necessarily  applied  to  the  things  themselves,  so  far 
as  they  exist  for  us ;  consequently  they  have  an 
objective  value. 

Again,  as  the  only  intuitions  at  the  disposal  of  our 
understanding,  for  the  forming  of  objects  thereof,  are 
our  sensible  intuitions,  and  as  the  latter  do  not  represent 
things  as  they  are  in  themselves,  but  only  the  exigencies 
of  our  sensibility,  one  consequence  of  our  human  con- 
dition is  that  even  our  intellectual  knowledge  is  unable 
to  attain  to  the  absolute,  it  remains  confined  to  the 
world  of  experience.  Empirical  realism,  and  trans- 
cendental idealism  are  allied  and  correlative  terms. 

Thus,  on  the  other  hand,  we  find  a  place  reserved 
for  the  suprasensible  itself.  Indeed,  the  concept  of  the 
"  thing-in-itself,"  whilst  limitative  of  the  claims  of  our 
science,  enables  us  to  conceive  of  a  world  other  than 
the  one  with  which  we  are  acquainted,  and  there- 
fore susceptible  of  being  freed  from  the  conditions 
of  our  knowledge,  and  especially  from  the  necessary 
connection  which  is  opposed  to  freedom.^  We  are 
permitted  to  superpose  the  noumenon  on  to  the  pheno- 
menon. 

It  is  essentially  this  doctrine  that  contains  the  philo- 
sophic revolution  wrought  by  Kant.  Instead  of 
admitting,  as  appearances  would  seem  to  indicate,  that 
thought  gravitates  around  things,  Kant,  like  a  modern 


284  STUDIES  IN  PHILOSOPHY 

Copernicus,  causes  things  to  gravitate  around  thought. 
From  this  point  of  view,  he  says,  the  disorderly  and 
the  inexplicable  give  way  to  the  orderly  and  the  intel- 
ligible. The  agreement  between  the  laws  of  nature 
and  the  laws  of  our  mind  is  no  longer  either  an  in- 
soluble problem  or  an  object  of  faith  :  it  is  a  scientifi- 
cally demonstrated  truth.  And  this  revolution,  which 
guarantees  the  objective  value  of  science,  is  equally 
favourable  to  morals,  which  latter,  in  the  field  opened 
up  by  criticism,  can  now  be  developed  unhindered,  in 
conformity  with  the  laws  proper  to  itself.  "  It  was 
only  by  abolishing  learning,"  says  Kant,  with  reference 
to  the  so-called  knowledge  of  the  suprasensible,  "  that 
I  could  find  room  for  belief." 

It  is  not  enough,  however,  to  lay  down  that,  in 
order  to  be  thought  of  and  to  become  objects,  the 
divers  elements  of  intuition  must  be  brought  under  the 
concepts  of  the  understanding.  How  will  concept,  the 
one  and  universal,  unite  with  phenomenon,  the  diverse 
and  particular  ?  How  shall  we  be  brought  to  apply 
to  intuition  any  one  category  rather  than  any  other  ? 
A  middle  term  is  here  necessary. 

This  middle  term  is  supplied  by  a  faculty  inter- 
mediary between  understanding  and  sensibility  :  viz. 
imagination.  In  the  form  of  the  inner  sense,  i.e.  in 
temporal  intuition,  imagination  traces  out,  a  priori^ 
frames  into  which  phenomena  are  capable  of  fitting, 
and  which  indicate  under  what  category  they  are  to 
be  brought.  Kant  calls  these  frames  schemata  of  the 
concepts  of  pure  understanding.  Each  category  has  its 
own  schema.  The  schema  of  quantity  is  number,  that 
of  substance  is  the  permanence  of  the  real  in  time  ; 
that  of  causality,  the  regular  succession  of  phenomena, 
and  so  on.  The  observance  of  regular  succession,  for 


KANT  285 

instance,  is  a  signal  to  us  that  the  category  of  cause  is 
being  employed. 

Still,  the  schemata  are  not  yet  sufficient  to  objectivise 
phenomena,  because  they  only  call  forth  the  employ- 
ment of  a  given  category,  without  justifying  this 
operation.  But  they  make  possible  a  priori  synthetic 
judgments  which  complete  the  elimination  of  the 
subjective.  These  judgments  are  the  principles  of  pure 
understanding.  Understanding  forms  them  a  priori^ 
by  determining  the  conditions  of  an  objective  employ- 
ment of  the  schemata.  They  are  :  the  principle  of 
quantity  :  "  All  intuitions  are  extensive  dimensions  "  ; 
the  principle  of  quality  :  "  In  all  phenomena,  sensation, 
as  well  as  the  real  which  corresponds  thereto  in  the 
object,  possesses  an  intensive  dimension,  a  degree  "  ; 
the  principle  of  relation :  "  All  phenomena  have  a 
necessary  connection  in  time."  The  principle  of 
modality  indicates  the  way  in  which  a  thing  should 
agree  with  the  conditions  of  experience,  in  order  to  be 
possible,  real,  or  necessary.  The  proof  of  these  prin- 
ciples consists  in  showing  that,  without  them,  the 
meaning  of  the  schemata  remains  indeterminate  ;  that 
the  sensible  can  be  determined  and  objecti vised  only  by 
the  intellectual.  Thus,  succession,  for  instance,  instead 
of  itself  founding  causality,  can  be  regarded  as  objective 
only  if  it  is  founded  thereon. 

On  reaching  this  stage,  Kant  was  enabled  to  accom- 
plish the  second  of  the  two  tasks  he  had  set  himself : 
that  of  justifying  physics  and  its  alliance  with  mathe- 
matics. The  first  two  principles — so-called  mathema- 
tical— establish  the  application  of  mathematics  to  the 
science  of  nature.  The  second  two — so-called  dynamic 
— establish  the  physical  laws  strictly  so  called.  In 
their  entirety,  the  principles  of  pure  understanding 


286  STUDIES  IN  PHILOSOPHY 

constitute  the  first  distinctive  features  of  natural 
philosophy.  This  theory,  whilst  being  the  meta- 
physical justification  of  Newtonian  science,  was  the 
starting-point  of  that  speculation  which,  with  Schelling, 
enjoyed  so  dangerous  a  renown,  under  the  name  of  the 
philosophy  of  nature. 

Up  to  this  point,  Kant  has  analysed  sensibility  and 
understanding.  There  remains  reason,  properly  so 
called,  the  object  of  which  faculty  is  the  complete 
unification  of  knowledge.  Its  syllogisms  infer  the 
unconditioned  as  their  starting-point.  So  we  see  that 
reason  is  the  faculty  of  the  ideas,  or  concepts,  of  the 
total  synthesis  of  the  conditions. 

From  what  precedes,  we  find  that  the  ideas  of 
reason  have  no  real  object.  Going  beyond  all  possible 
experience,  they  can  be  nothing  but  regulative,  non- 
constitutive  principles  of  knowledge.  The  illusion, 
however,  which  makes  us  believe  in  their  objectivity,  is 
natural,  as  is  that  of  the  man  who  believes  the  moon  to 
be  larger  at  its  rise  than  at  its  meridian.  To  destroy 
this  illusion  it  is  not  enough  to  demonstrate  the  falsity 
of  our  opinion  ;  its  origin  must  be  disclosed,  it  must 
be  shown  that,  in  this  domain,  in  contradistinction  to 
what  takes  place  when  dealing  with  objects  of  possible 
experience,  it  is  wholly  illegitimate  to  pass  from  the 
logical  to  the  real ;  and  that  the  dialectic  which  lies  deep 
hidden  in  dogmatic  metaphysics  must  be  denounced. 

Reason  thinks  it  can  build  up :  ist,  a  rational 
psychology,  on  the  idea  of  the  soul-substance  ;  2nd,  a 
rational  cosmology,  on  the  idea  of  the  world  as  absolute 
reality  ;  3rd,  a  rational  theology,  on  the  idea  of  God  as 
the  absolute  basis  of  the  possibility  of  being  in  general. 
In  each  of  these  domains  it  is  mistaken  regarding  its 
own  power. 


KANT  287 

When  inferring  the  existence  of  an  absolute  subject 
from  the  reality  of  the  thinking  being,  it  illegitimately 
passes  from  a  unity  of  form  to  one  of  substance,  and 
commits  a  paralogism. 

When  attempting  to  determine  the  absolute  existence 
it  attributes  to  the  world,  reason  becomes  involved  in 
insurmountable  antinomies.  Indeed,  it  proves,  with 
like  rigour,  by  the  absurdity  of  the  contradictory 
proposition,  that  the  world  both  has,  and  has  not, 
limits  ;  that  it  consists  of  simple  parts,  and  is  divided 
ad  infinitum  ;  that  freedom  exists  and  that  nothing  free 
exists  ;  that  there  is  a  necessary  being,  and  that  there 
exist  only  contingent  beings.  The  very  production  of 
these  antinomies  proves  the  illegitimacy  of  the  point  of 
view  that  gives  birth  to  them,  that  is,  of  the  supposition 
of  a  world  existing  in  itself.  In  the  first  two  anti- 
nomies, thesis  and  antithesis  are  alike  false.  In  the 
latter  two,  they  become  true  of  each  other  if  we  have 
recourse  to  that  distinction  between  phenomenon  and 
noumenon  called  forth  by  analysis  of  the  understanding. 
The  free  and  the  absolute  are  possible  in  the  world  of 
noumena,  whereas  natural  causality  and  contingency 
hold  sway  in  that  of  phenomena. 

Finally,  when  reason  speculates  on  perfect  being, 
it  gratuitously  converts  into  a  reality,  a  substance,  a 
person,  the  ideal  in  which  it  unites  all  the  modes  of 
being  possessed  by  finite  things.  Consequently,  the 
reasonings  it  forms  to  prove  the  existence  of  this 
supreme  person  will  not  hold  together.  The  onto- 
logical  argument,  the  basis  of  all  the  rest,  wrongly 
considers  existence  as  a  predicate,  which  can  be  obtained 
from  a  concept  by  analysis  :  existence  is  the  position  of 
a  thing  outside  of  thought  and  is  absolutely  inaccessible 
to  analysis.  To  this  error  the  cosmological  argument 


288  STUDIES  IN  PHILOSOPHY 

adds  the  affirmation  of  a  first  cause  in  the  name  of 
the  principle  of  causality,  and  this  principle,  just  in 
so  far  as  it  is  vouched  for,  excludes  the  possibility  of 
a  first  cause.  Lastly  the  physico-theological  argument, 
or  the  argument  of  final  causes,  adds  to  the  defects  of 
the  first  two  the  false  comparison  of  the  world  to  a 
work  of  man,  and  the  arbitrary  transition  from  an 
"  architect "  God  to  a  perfect  "  creator  "  God. 

The  general  cause  of  this  dialectic  of  our  reason  is 
our  natural  disposition  to  believe  that  the  conditions  of 
our  thought  are  also  the  conditions  of  being,  that  the 
laws  of  our  knowledge  are  the  laws  of  reality.  Criti- 
cism alone  can  dispel  this  illusion  ;  but  the  necessity 
of  criticism  is  seen  only  in  the  consequences  of  this 
very  illusion.  The  ideas  of  our  reason  correspond  to 
nothing  real :  none  the  less  are  they  useful  as  excitative 
and  regulative  principles.  They  forbid  our  halting  in 
our  search  after  causes.  We  cannot  begin  with  God, 
but  our  efforts  should  tend  in  his  direction. 

And  so  criticism  is  established,  wherein  Kant  sees 
the  goal  of  the  education  of  reason.  The  human 
mind  began,  and  was  compelled  to  begin,  with  dog- 
matism, or  a  blind  belief  in  the  absolute  existence  of 
the  objects  of  our  thoughts  :  Leibnitzo-Wolfianism  is 
the  complete  expression  thereof.  Then  came  skepticism, 
excellently  represented  by  Hume,  who  inferred,  from 
the  vices  of  dogmatism,  the  impossibility  of  knowing 
reality  and  the  absolute  subjectivity  of  knowledge.  But 
skepticism  is  only  a  warning  to  mistrust  dogmatism. 
Criticism,  or  the  science  of  our  ignorance,  forbids  us  to 
speculate  on  the  nature  of  things  as  they  are  in  them- 
selves ;  at  the  same  time,  it  withdraws  experience  from 
imagination  and  the  individual  sense,  to  make  it  an 
object  common  to  all  human  intelligences  and  conse- 


KANT  289 

quently  substantial  and  real  to  ourselves.  At  the  same 
time,  criticism  frees  being  in  itself  from  the  fatum 
which  the  presumption  of  the  understanding  caused  to 
lie  heavy  on  it ;  it  makes  conceivable  a  world  wherein 
freedom  and  the  moral  laws  would  hold  undivided 
sway.  The  advantage  is  twofold,  being  both  practical 
and  speculative  ;  it  attests  to  the  providential  harmony 
of  our  needs  with  our  powers  of  knowing. 

The  "  critique "  of  pure  reason  has  explained  the 
possibility  of  science  ;  in  the  same  way,  the  possibility 
of  morals  must  now  be  explained.  We  are  not  trying 
to  find  out  if  morality  is  possible,  since  it  exists,  but 
rather  on  what  it  rests  and  what  its  meaning  is.  Here, 
too,  a  sane  philosophy  can  recognise  no  other  starting- 
point  for  knowledge  than  experience,  but  this  ex- 
perience must  be  analysed. 

The  general  idea  afforded,  in  this  connection,  by 
common  reason,  is  the  concept  of  good  will.  Is  this 
concept  altogether  empirical  ? 

When  examined,  it  is  found  to  imply  the  idea  of  a 
law  which  ought  to  be  observed  for  itself,  without 
regard  for  the  consequences  of  the  actions  it  enforces. 
This  law  is  not  a  hypothetical  imperative,  dependent 
on  such  or  such  an  end  to  be  attained  :  it  is  a  categori- 
cal imperative.  It  can  be  formulated  only  in  the 
following  terms  :  act  in  such  a  way  that  you  would 
wish  the  maxim  of  your  action  to  be  set  up  as  a  uni- 
versal Jaw.  Now,  such  a  principle  does  not  proceed 
from  experience,  it  is  known  a  priori. 

Can  we  find  its  origin  ?  If  we  try  to  discover  under 
what  conditions  a  practical  principle  may  be  universally 
obligatory  upon  us,  we  shall  find  that  it  ought  to  imply 
no  object  or  matter  as  a  mobile  of  the  will.  Indeed, 

u 


290  STUDIES  IN  PHILOSOPHY 

given  the  faculties  we  possess,  there  are  none  other  than 
empirical  objects  as  far  as  we  are  concerned,  the  only 
matter  at  our  disposal  in  the  practical  order  of  things 
is  pleasure  or  the  satisfaction  of  the  love  of  self ;  and 
pleasure  cannot  supply  a  universal,  obligatory  prin- 
ciple. The  intention  of  our  will,  alone,  depends  entirely 
on  ourselves  and  fulfils  the  requisite  conditions.  Law, 
then,  is  a  purely  formal  principle  which  implies  nothing 
else  than  itself  and  a  will  free  to  accomplish  it.  It  has 
its  root  in  the  autonomy  of  the  will. 

But  even  in  this,  is  it  not  illusory  ?  Detached  from 
things  and  referred  to  the  subject,  is  it  not  purely  sub- 
jective ?  Can  we  escape  from  idealism  in  the  practical, 
as  we  have  done  in  the  theoretical  order  of  things  ? 

To  deduce  the  moral  law  from  the  conditions  of 
experience  is  impossible,  since  every  object  of  experience 
ought  to  be  separated  from  moral  determination  ;  but, 
on  the  other  hand,  the  moral  law  itself  establishes  a 
deduction  from  freedom.  If  I  ought  to,  it  is  because 
I  can.  Moreover,  if  speculative  reason  has  had  to 
be  debarred  from  knowing  freedom,  it  has  none  the 
less  regarded  it  as  possible,  even  theoretically  ;  and 
thus  the  moral  law  has  a  basis  in  the  reality  of  things, 
as  this  reality  is  theoretically  known  to  us,  viz.  in  that 
region  of  existence  to  which  the  knowledge  of  things  as 
phenomena  refers  us.  If  the  moral  law  is  the  ratio 
cognoscenti  of  freedom,  the  latter  supplies  the  former 
with  its  ratio  essendi. 

So  far,  however,  we  have  only  reached  a  principle,  a 
formal  law.  Now,  morality  also  offers  us  concepts,  the 
two  principal  of  which  are  those  of  good  and  evil.  Can 
we  search  into  and  understand  these  concepts  ?  After 
eliminating  all  empirical  matter,  we  have  to  deduce  fresh 
matter  from  a  principal  posited  as  purely  formal. 


KANT  291 

The  course  we  must  take  is  apparently  paradoxical. 
Is  it  not  duty  that  is  deduced  from  good,  and  not  good 
that  is  determined  by  duty?  The  ancients,  in  their 
search  after  the  sovereign  good,  constantly  followed  the 
first,  the  dogmatic  course.  Now,  willingly  or  unwillingly, 
it  came  about  that  they  founded  morals  on  empirical 
data.  It  could  not  be  otherwise.  From  good,  one 
cannot  deduce  duty,  unless  this  good  is  already  moral 
good,  and  it  is  only  moral  if  there  has  previously  been 
instilled  into  it  the  very  duty  it  is  desired  to  deduce 
therefrom.  On  the  other  hand,  it  is  possible,  by  means 
of  duty,  to  determine  good  ;  it  is  possible  for  law 
posited  as  primary,  to  find  a  suitable  object  in  the 
sensible  world  itself,  the  only  one  we  can  affect.  For 
this  sensible  world  not  only  does  not  clash  with  the 
universality  characterising  the  moral  law,  but  is  itself 
subject  to  universal  laws.  Good,  therefore,  is  the  realisa- 
tion in  the  sensible  world  of  a  form  of  universality 
capable  of  being  the  symbol  of  practical  reason. 

This  doctrine  of  Kant's  rejects  mysticism  as  well  as 
empiricism.  Though  the  principle  of  determination 
ought  to  be  obtained  from  the  world  of  noumena,  it  is 
in  the  world  of  phenomena  that  morality  will  be  realised 
and  practised.  And  the  very  principle  of  determina- 
tion will  not  remain  unrelated  to  nature.  There  exists 
a  feeling  which  is  within  nature  and  which  likewise 
goes  beyond  it,  and  that  is  respect,  a  special  affection 
aroused  by  the  idea  of  law  in  a  soul  endowed  both  with 
sensible  tendencies  and  with  reason.  Respect  is  the 
moral  mobile.  The  inclination  it  enshrouds,  and  which 
comes  from  the  will,  does  no  harm  to  the  disinterested 
practice  of  duty. 

And  so  the'given  morality  is  explained  and  defined 
in  all  its  elements  :  mobiles,  concepts,  and  principles. 


292  STUDIES  IN  PHILOSOPHY 

Here,  too,  we  had  only  to  go  back  from  experience  to 
its  conditions,  in  order  to  explain  whatever  is  absolute 
in  our  knowledge,  without  detracting  from  the  general 
principle  of  modern  science  and  philosophy. 

And  not  only  does  criticism  thus  insure  the  founda- 
tions of  morals,  it  also  discloses  the  spring  and  reason 
of  religious  beliefs  from  the  very  point  to  which  this 
investigation  has  brought  it.  Reason  requires  the  full 
performance  of  duty,  it  exacts  the  union  of  virtue  with 
happiness.  How  can  such  an  object  be  realised  ? 

The  necessity  of  answering  this  question  leads  us  on 
to  theoretical  propositions  that  cannot  be  demonstrated 
as  such,  but  are  inseparably  bound  to  practical  truths 
of  an  absolute  character.  These  propositions  Kant 
calls  postulates.  They  are  three  in  number  : 

1.  Freedom  :    necessary    in    order   that    man    may 
determine   himself,   apart  from   all  sensible   attraction, 
in   accordance   with   the  laws   of  a  purely   intelligible 
world.     Doubtless,  freedom  does  not  intervene  in  the 
course  of  phenomena,  which  would  cease  to  be  objects 
of  possible  experience  if  the  law  of  cause  and  effect 
were  violated  in  them.     It  is  complete  and  entire,  how- 
ever, in  the  world  of  noumena,  in  which  it  establishes 
personality  and  creates  within  each  of  us  an  intelligible 
character,  of  which  our  empirical  character  is  the  symbol. 

2.  Immortality  :   necessary  in  order  that   indefinite 
progress  may   be  realised,   without  which  the   perfect 
adaptation  of  our  will  to  the  moral  law  is  inconceivable. 

3.  God  :  necessary  in  order  that  we  may  establish 
that  agreement  which  reason  demands,  between  morality 
and  happiness,  and  the  principle  of  which  is  contained 
in  neither  the  one  nor  the  other. 

Thus,  morality  leads  to  religion,  not  as  to  some 
theoretical  science  explaining  the  nature  of  things,  but 


KANT  293 

rather  as  to  the  knowledge  of  our  duties  in  so  far  as  they 
are  divine  commands. 

And  so  criticism,  continuing  its  progress,  gradually 
re-establishes  all  the  suprasensible  existences  it  had 
overthrown.  Is  it  self-contradictory  in  doing  this? 
By  no  means  ;  since  it  no  longer  regards  these  exist- 
ences in  the  same  manner.  The  criticism  of  pure 
reason  has  demonstrated  that  such  objects  cannot  be 
known  theoretically,  i.e.  by  the  aid  of  intuitions  which 
determine  them.  This  result  subsists.  The  criticism 
of  pure  reason,  however,  did  not  prevent  our  conceiving 
of  objects  above  experience,  on  the  contrary  it  allowed 
and  invited  this.  On  the  other  hand,  the  criticism  of 
practical  reason  in  no  way  shows  us  the  world  shut  out 
from  us  by  the  criticism  of  pure  reason,  it  does  not 
give  us  an  intuition  thereof,  but  offers  us,  as  connected 
with  the  existence  of  duty,  the  objects  on  which  theo- 
retical reason  could  not  declare  itself.  It  brings  us  to 
say,  not  :  It  is  certain  there  is  a  God  and  immortality  ; 
but  rather  :  I  will  there  to  be  a  God,  I  will  my  being, 
in  one  aspect,  to  be  free  and  immortal.  That  is  not 
a  matter  of  science,  it  is  a  practical,  pure,  and  rational 
belief.  We  can  neither  see  the  object  nor  deduce  it 
from  what  we  see  ;  we  can  only  conceive  of  it.  How 
fortunate  this  inability  !  For  were  we  in  possession  of 
the  missing  faculty,  instead  of  duty  tempering  and 
ennobling  our  will,  God  and  eternity,  with  all  their 
awful  majesty,  would  ever  be  before  our  eyes,  and 
would  reduce  us,  through  fear,  to  the  condition  of 
marionettes,  making  the  proper  movements  but  devoid 
of  life  or  moral  worth.  "  The  mysterious  wisdom  by 
which  we  exist  is  no  less  admirable  in  the  gifts  it  has 
refused  than  in  those  it  has  granted  to  us"  (Critique  of 
Practical  Reason,  Part  i.  Book  ii.  Chapter  ix.). 


294  STUDIES  IN  PHILOSOPHY 

Criticism  has  explained  the  existence  of  science  and 
morals.  To  complete  the  different  orders  of  our  know- 
ledge, it  remains  for  us  to  examine  the  notions  of  taste 
and  finality.  Will  experience  be  able  to  supply  us 
with  their  principle  and  their  limits  ? 

The  experimental  datum  here  considered  is  judg- 
ment ;  not  determining  judgment,  which  proceeds  from 
the  general  to  the  particular,  but  reflecting  judgment, 
which  rises  from  the  particular  to  the  general.  This 
judgment  is  that  which  affirms  the  existence  in  nature, 
not  only  of  laws  in  general,  but  of  certain  determinate 
laws.  It  calls  for  a  special  principle  which  can  be  only 
the  following  :  just  as  the  universal  laws  of  nature  are 
based  in  our  understanding,  which  prescribes  them  to 
nature,  so,  as  regards  empirical  and  particular  laws, 
everything  takes  place  as  though  they  also  had  been 
dictated  by  an  understanding  that  purposed  to  make 
intelligible  and  objective  the  very  details  of  the  pheno- 
mena. This  reason  of  particular  laws  may  be  sought 
for,  either  in  the  agreement  of  things  with  our  faculty 
of  knowing,  i.e.  in  the  beautiful,  or  in  the  agreement 
of  things  with  themselves,  i.e.  in  finality. 

Appreciation  of  the  beautiful  cannot  be  explained 
by  sensation  alone,  as  Burke  would  have  it.  The 
beautiful  is  not  the  agreeable  ;  it  is  disinterested,  the 
object  of  a  real  judgment.  Nor  is  it  explained  by 
reason  alone,  according  to  Baumgarten  the  Wolfian. 
The  beautiful  is  not  the  perfect  :  it  dwells  only  in  the 
form,  not  in  the  matter,  of  the  object,  and  it  pleases 
without  aiming  at  pleasing  but  solely  by  reason  of  its 
harmony,  by  a  kind  of  endless  finality  :  in  a  word,  it 
has  something  of  feeling  in  its  nature.  Formed  a 
priori  and  being  subjective  at  the  same  time,  what  is 
the  origin  of  the  judgment  of  taste  ? 


KANT  295 

It  can  only  be  explained  as  the  working  of  an 
aesthetic  common  sense,  or  the  faculty  of  perceiving 
some  agreement  between  our  sensible  faculty  of  know- 
ing and  our  intellectual  faculty.  Those  objects  are 
beautiful,  before  which  our  imagination  finds  itself,  of 
its  own  accord,  satisfying  our  understanding.  The 
beautiful  is  the  feeling  that  our  faculties  are  at  play, 
somewhat  analogous  to  a  physical  pastime,  wherein  the 
spontaneous  observance  of  a  rule  freely  laid  down  in 
no  way  trammels  the  free  expression  of  activity.  Con- 
sequently, the  beautiful  dwells  only  within  ourselves  ; 
it  has  no  other  origin  or  rule  than  the  special  sense  in 
which  sensibility  and  understanding  meet  each  other. 

From  the  beautiful,  properly  so  called,  that  we  are 
now  analysing,  we  must  distinguish  the  sublime,  as 
being  another  species  of  the  same  genus.  Whereas  the 
beautiful  object  is  the  adequate  sensible  realisation  of 
the  idea,  the  sublime  object  utterly  routs  the  imagina- 
tion, which  spends  itself  in  vain  attempts  to  represent 
an  idea  transcending  it.  There  are  no  images,  but  only 
symbols,  of  the  infinite.  The  substratum  of  the  sub- 
lime and  the  beautiful  alike  can  therefore  be  nothing 
else  than  our  suprasensible  nature,  and  the  need  of 
agreement  between  that  nature  and  our  sensible  nature. 

But  then  does  not  this  analysis  result  in  the  judg- 
ment of  taste  being  denied  all  objective  value  ?  Such 
would  be  the  case,  had  the  objectivity  of  the  beautiful 
to  consist,  in  our  mind,  of  some  property  of  things  in 
themselves  :  such  an  objectivity,  however,  is  an  illusion. 
The  sense  of  taste  that  we  have  found,  has  an  objec- 
tive import,  in  so  far  as  it  alone  makes  intelligible  the 
characteristic  of  beauty  that  we  attribute  to  objects, 
and  in  so  far  as  this  very  sense  should  be  considered 
identical  in  all  beings  capable  of  sensibility  and  dis- 


296  STUDIES  IN  PHILOSOPHY 

cursive  understanding.  The  universality  of  the  faculty 
is  sufficient  to  establish  the  objectivity  of  the  operation. 

But  if  we  now  consider  things  of  taste,  especially  art, 
whose  existence  is  given,  our  doctrine  will  supply  die 
theory  thereof.  Art  is  a  product  of  intelligence,  and 
ought  to  appear  a  product  of  nature  ;  it  has  an  object 
and  ought  to  seem  not  to  have  one  ;  it  punctually 
observes  rules  and  does  this  without  manifesting  effort. 
All  these  characteristics  are  explained  as  soon  as  man 
possesses  a  faculty  wherein  the  understanding,  which 
thinks  and  rules,  coincides  with  the  imagination,  which 
sees,  feels,  and  invents.  The  spring  of  genius  is  dis- 
covered in  the  general  essence  of  man.  And  it  is  also 
seen  that  the  more  human  the  object  of  an  art,  the 
more  sublime  that  art  is. 

Moreover,  the  ideality  of  the  beautiful  is  the  only 
doctrine  that  enables  us  to  solve  the  antinomy  to  which 
the  judgment  of  taste  gives  rise.  We  discuss  about  the 
beautiful,  and  yet  we  cannot  account  for  it  by  demon- 
stration. This  would  be  incomprehensible,  did  the 
beautiful  belong  to  things  in  themselves.  But  then,  on 
the  other  hand,  the  beautiful  could  not,  like  time  and 
space,  be  enclosed  within  the  sensible  world.  We 
discuss  about  the  beautiful  and  yet  we  cannot  demon- 
strate anything,  because  the  judgment  of  taste  is  based 
on  a  principle  connected  both  with  concept  and  intuition, 
on  an  indeterminate  concept :  that  of  a  suprasensible 
substratum  of  phenomena.  The  beautiful  is  the  symbol 
of  moral  good,  and  it  is  towards  this  good  that  taste 
dimly  leads  us. 

The  second  principle  of  particular  natural  laws  is 
derived  from  finality.  Do  there  really  exist  in  nature 
harmonies  that  cannot  be  explained  by  mechanism  or 
the  system  of  causes  and  effects  ? 


KANT  297 

Wherever  finality  is  only  exterior,  consisting  only  in 
the  utility  of  one  being  with  reference  to  another,  the 
mechanical  explanation  is  sufficient,  for  this  agreement 
of  different  beings  with  one  another  is  far  from  being 
the  rule  in  nature.  But  there  is  one  case  in  which 
finality,  being  internal,  cannot  be  explained  by  the 
hazards  of  mechanism :  the  case  of  organised  beings. 
That  which  is  living  produces  itself,  both  as  species  and 
individual,  and  the  parts  thereof  are  conditioned  by  the 
very  ensemble  which  is  to  result  therefrom.  The  effect 
here  is  the  cause  of  its  cause  ;  the  cause  is  the  effect  of 
its  effect.  Such  a  relation  goes  beyond  mechanism,  such 
a  being  is  an  end,  as  well  as  a  product  of  nature.  How 
is  that  possible  ? 

In  vain  does  dogmatism  attempt  to  reply  either  by 
hylozoism,  which  looks  upon  nature  as  intelligent,  or 
by  theism,  which  weaves  the  action  of  intelligence  into 
the  tissue  of  phenomena :  the  former  attributes  to 
matter  qualities  opposed  to  its  essence  ;  the  latter  vainly 
claims  to  pierce  the  designs  of  God.  Organisation, 
the  internal  finality,  is  not  cognisable  in  its  cause. 
Finality,  to  us,  can  be  nothing  but  ideal :  it  is  our  way 
of  looking  upon  a  certain  class  of  phenomena. 

Is  such  a  doctrine  a  purely  negative  result  ?  By  no 
means. 

Some  knowledge  of  nature  is  implied  if  we  simply 
know  that,  in  certain  of  its  products,  nature  cannot  be 
known  by  us.  This  principle  is  instructive,  either  in  its 
restrictive  or  its  positive  bearing.  It  is  regulative,  not 
constitutive.  In  this  capacity  it  serves  science.  Though 
it  does  not  make  the  production  of  things  more 
intelligible,  all  the  same,  it  supplies  anticipations  by 
which  we  are  enabled  to  discover  the  particular  laws  of 
nature.  It  sets  up  beacon-lights  throughout  infinity. 


298  STUDIES  IN  PHILOSOPHY 

So  far  as  metaphysics  is  concerned,  only  such  a  con- 
ception of  finality  enables  one  to  escape  from  the 
traditional  antinomy  of  mechanism  and  teleology.  On 
the  ground  of  being  in  itself,  wherein  both  systems  are 
placed,  neither  the  first  is  able  to  explain  what  it  calls 
the  illusion  of  finality,  nor  the  second  to  prove  that  the 
transcendent  explanation  of  it  is  necessary.  On  the 
other  hand,  the  principle  of  final  causes  becomes  unas- 
sailable when  there  is  only  one  point  of  view  upon 
things. 

And  it  opens  up  to  our  conception,  if  not  to  our 
knowledge,  a  perspective  upon  the  absolute  itself. 
Indeed,  how  do  we  come  to  posit  the  idea  of  an  end 
as  the  cause  of  a  phenomenon  ?  The  impossibility  of 
deducing  the  particular  from  the  universal  comes  from 
the  fact  that  understanding  and  intuition  are  separate 
in  us  ;  our  concepts  are  void,  our  intuitions  powerless 
to  connect  themselves  into  laws.  Then  how  can  we 
affirm  the  existence  of  particular  laws  ?  The  problem 
is  solved  as  follows.  We  can  conceive  that  the  difficulty 
in  our  way  would  be  non-existent  to  a  mind  in  which 
understanding  were  one  with  sensibility  :  to  an  intuitive 
understanding.  Such  a  mind,  instead  of  proceeding 
from  the  parts  to  the  whole,  as  does  our  discursive 
understanding,  and,  consequently,  seeing  a  contingent 
result  in  the  whole,  would  proceed  from  the  whole  to 
the  parts,  and,  in  a  flash,  would  see  the  latter  in  their 
necessary  connection.  To  this  mind,  mechanism  and 
finality  would  coincide.  Now,  once  the  idea  of  such 
an  intelligence  is  conceived  of,  our  understanding,  in 
order  to  approach  it  in  its  own  way,  substitutes  for  the 
whole  the  idea  of  the  whole,  and  posits  this  idea  before 
its  intuitions  as  the  cause  of  the  special  relations  that 
unite  them.  To  the  employment  of  the  notion  of  an 


KANT  299 

end  is  thus  linked  on  the  conception  of  an  intuitive 
understanding,  as  a  possible  foundation,  in  the  absolute, 
of  the  sum  total  of  the  harmonies  of  nature. 

This  deduction  from  teleological  judgment  deter- 
mines the  use  we  ought  to  make  of  it. 

As  regards  the  explanation  of  the  phenomena  of 
nature,  we  have  the  right,  as  far  as  possible,  to  assume 
the  mechanical  point  of  view,  but  we  cannot  do  this  on 
all  occasions  with  like  success.  In  the  fact  of  life  we 
are  brought  in  opposition  against  an  invincible  barrier. 
We  cannot  picture  to  ourselves  living  bodies  as  capable 
of  coming  from  inorganic  matter.  Doubtless,  it  is  not 
inconceivable  that  from  one  common,  originally  organ- 
ised, matter,  all  living  bodies  might  have  issued  by 
purely  mechanical  changes.  In  this  way,  the  explana- 
tion of  things  would  be  the  province  of  mechanism  ; 
their  origin,  that  of  teleology.  Indeed,  the  comparison 
of  organic  forms  enables  us  to  conjecture  the  relation- 
ship of  all  that  lives,  and  encourages  us  to  hope,  however 
feebly,  that  it  will  be  possible  to  refer  them  to  one 
common  origin.  Then  one  could  picture  to  oneself 
the  womb  of  the  earth  as  giving  birth,  first  to  creatures 
ill-suited  to  the  conditions  of  their  existence,  and  then 
to  these  same  creatures  as  becoming  more  perfect,  gener- 
ation after  generation,  until  finally  the  creatress,  in  a 
state  of  congealed  ossification,  so  to  speak,  limited  her 
productions  to  a  certain  number  of  clearly  defined  and 
henceforth  immutable  species.  This  is  a  brilliant  hypo- 
thesis of  reason,  but  apart  from  the  fact  that  so  far 
experience  does  not  seem  to  warrant  it,  instead  of 
excluding,  it  would  imply  as  a  condition  of  its  con- 
sistency the  primordial  life  of  the  universal  womb. 

As  regards  the  general  conception  of  the  world,  we 
have  the  right  to  complete  by  thought  the  unification 


300  STUDIES  IN  PHILOSOPHY 

to  which  teleological  concepts  tend,  provided  we  place 
this  ultimate  end  outside  the  sphere  of  sensible  pheno- 
mena. And  as  this  end  can  be  only  a  being  that  has 
within  itself  the  object  of  its  activity,  and  consequently, 
is  capable  of  positing  ends  and  using  nature  as  a  means, 
man  alone,  not  as  a  part  of  nature,  but  as  intelligence 
and  will,  can  be  the  end  of  the  universe.  We  must 
not,  like  Rousseau,  expect  nature  to  satisfy  our  longings, 
to  give  us  happiness  ;  that  is  out  of  her  province,  and 
she  will  play  us  false.  But  she  will  not  belie  the 
expectations  of  the  man  who,  through  her,  endeavours 
to  realise  moral  good. 

Finally,  in  the  matter  of  our  conception  of  God  as 
the  principle  of  finality,  it  has  not  been  without  purpose 
that  men,  at  all  times,  have  been  influenced  by  the 
argument  of  final  causes.  This  argument  well  expresses 
man's  impression  when  he  sees  the  order  of  nature  : 
the  aspiration  towards  something  that  goes  beyond 
nature.  We  must  always  speak  of  this  argument  with 
respect,  for  it  is  the  most  persuasive,  popular,  and  potent 
one  of  all.  To  be  really  solid  and  sound,  however,  the 
argument  must  be  understood  in  its  true  meaning.  Not 
as  an  architect  is  God  revealed  to  us  by  the  world,  but 
rather  as  the  condition  of  an  agreement  between  nature 
and  morality.  In  trying  to  discover  the  attributes 
needed  to  play  this  part,  we  shape  for  ourselves  a  moral 
theology  which  leads  us  on  to  a  moral  religion. 

IV. — THE  METAPHYSICAL  DOCTRINE 

Criticism  is  not  the  abolition  of  metaphysics,  it  is 
the  introduction  to  metaphysics  as  a  science.  In  realising 
the  plan  it  here  marks  out,  the  method  to  be  followed  is 
the  one  inaugurated  by  the  famous  Wolf.  We  know  that 


KANT  301 

transcendental  logic  does  not  break  through  the  frame- 
work of  general  logic  :  it  fills  it  in.  We  shall  find  meta- 
physics changing  its  meaning  without  changing  its  form. 
Human  reason  is  legislative  in  two  ways  :  by  its 
understanding,  in  the  domain  of  nature  ;  and  by  its 
will,  in  the  domain  of  freedom.  Hence  the  idea  of  a 
double  metaphysics  :  of  nature  and  of  morals.  There 
are  no  others. 

Kant  deals  first  with  the  metaphysics  of  the  science 
of  nature. 

Corporeal  matter,  being  alone  lasting,  can  alone  give 
rise  to  metaphysics.  The  latter  seeks  amongst  the 
sensible  data  or  properties  of  matter,  for  some  object 
to  which  the  synthetic  laws  of  understanding  are 
applicable,  and  this  it  finds  in  motion.  Once  this 
single  result  has  been  obtained  from  experience,  meta- 
physics pursues  its  course,  proceeding  a  priori. 

Determined  solely  by  the  notion  of  quantity,  motion 
is  nothing  but  dimension  in  time  and  space :  it  does  not 
yet  imply  cause  of  production  or  of  modification.  In 
this  connection  it  gives  rise  to  phoronomics,  which  we 
now  call  kinematics. 

Determined,  besides,  by  the  notion  of  quality,  it 
envelops  an  intensive  dimension  or  force,  as  the  cause 
of  its  existence  and  of  our  sensible  affections.  The 
theory  of  force  is  dynamics :  the  essential  element  in 
this  portion  of  Kantian  metaphysics.  We  admit  as 
many  simple  forces  as  it  is  necessary  to  posit,  in  order 
to  distinguish  movements  in  a  straight  line,  conse- 
quently we  admit  a  force  of  repulsion  and  one  of 
attraction.  From  the  first,  there  results  divisibility 
ad  infinitum ;  from  the  second,  a  limitation  of  the  first. 
These  two  forces  are  solidary  :  solidity,  which  the 


302  STUDIES  IN  PHILOSOPHY 

Newtonians  found  themselves  compelled  to  add  on  to 
attraction,  unless  it  be  an  occult  quality,  implies  a 
repulsive  force.  Matter  results  from  the  equilibrium 
of  the  two. 

Determined  by  the  notion  of  relation,  matter 
assumes  properties  which  are  investigated  by  mechanics, 
properly  so  called.  Here,  Kant  establishes  the  law  of 
the  persistence  of  material  substance,  the  law  of  inertia 
and  that  of  action  and  reaction. 

Finally,  regarding  modality,  we  have  to  find  out 
the  rules  followed  by  the  mind  when  distinguishing 
possible,  real,  or  necessary  motion  :  this  is  phenomenology. 
Rectilinear  motion  is  only  possible,  it  appertains  to 
phoronomics  ;  curvilinear  motion  is  real,  and  appertains 
to  dynamics  ;  motion  conceived  of  as  communicated  by 
a  mover  to  something  movable  is  of  necessity  deter- 
mined as  regards  existence  and  speed  ;  it  appertains 
to  mechanics. 

From  these  metaphysical  principles  Kant  en- 
deavoured to  pass  on  to  physics  itself.  Physics 
would  evidently  be  constituted  as  a  science,  if  only 
we  could  determine  a  priori  the  forces  that  produce 
sensation.  Now,  we  see  from  the  Critique  that  these 
forces,  being  bound  to  the  life  of  the  mind,  must,  after 
all,  be  of  the  same  nature  as  the  mind.  They  can  be 
nothing  else  than  the  action  exerted  upon  our  empirical 
I  by  our  spontaneity,  i.e.  our  understanding.  And 
it  is  because  this  action  is  transcendental  that,  in  our 
endeavours  to  picture  to  ourselves  the  cause  of  our 
sensations,  we  imagine  things  outside  of  ourselves  in 
space.  Henceforth,  the  principle  of  the  deduction  of 
material  species  is  in  our  hands  :  it  is  none  other  than 
the  principle  of  the  functions  of  the  subject  itself. 
It  is  in  this  way  that  Kant,  by  the  light  of  the  cate- 


KANT  303 

gories,  undertakes  the  deduction  of  the  different  kinds 
of  forces,  of  first  matter  or  ether,  of  bases  or  specific 
matters.  In  all  probability,  he  would  have  reached  a 
rational  deduction  of  the  system  of  the  world  itself, 
such  as  Newton  had  constituted  it. 

The  second  and  last  part  of  metaphysics  is  the 
metaphysics  of  morals. 

In  the  moral  as  in  the  physical  order  of  things,  it  is 
the  task  of  method  to  bring  the  given  empirical  condi- 
tions under  the  laws  of  reason,  and  thence  deduce  the 
complete  system  of  fundamental  laws.  Moral  legisla- 
tion has  a  double  object  in  view  :  action  and  its  mobile. 
Harmony  of  action  with  the  law  is  legality ,  that  of  the 
mobile  is  morality.  This  distinction  results  in  the 
division  of  the  metaphysics  of  morals  into  the  theory  of 
right  and  the  theory  of  virtue. 

Right  consists  of  the  whole  of  those  conditions  that 
are  universally  required  in  order  that  the  free-will  of 
each  individual  may  be  reconciled  with  that  of  the  rest. 
External  free-will  commands  respect,  because  it  is  the 
form  of  moral  freedom,  the  latter  being  realised  only 
by  action  and  action  implying  a  connection  with  some- 
thing external.  Consequently  the  science  of  right  is 
distinct  from,  though  dependent  on,  morals. 

There  are  two  essential  principles  that  control  the 
development  of  the  theory  of  right  :  1st,  right  is  alto- 
gether based  on  the  suprasensible  nature  of  man  so  far 
as  it  is  manifested  in  time,  i.e.  on  personal  dignity  ; 
2nd,  legal  restraint  is  legitimate,  so  far  as  it  is  neces- 
sary for  suppressing  the  obstacles  that  one  will  may 
arbitrarily  set  up  against  the  development  of  the  rest. 
The  consequences  of  these  principles  are  as  follows  : — 

So  far  as  private  right  is  concerned,  there  belongs  of 
necessity  to  each  man  such  a  portion  of  freedom  as  is 


304  STUDIES  IN  PHILOSOPHY 

compatible  with  the  freedom  of  the  rest  of  mankind. 
But  here  we  can  deal  only  with  freedom  regarded  in  its 
external  existence.  This  external  expression  of  freedom 
is  what  is  called  possession. 

There  are  as  many  kinds  of  rights  as  there  are  of 
possessions. 

The  first  has  reference  to  things,  and  gives  rise  to 
real  right.  This  right  is  not  a  relation  between  the 
owner  and  the  thing,  it  is  rather  one  between  persons. 
How  can  its  realisation  be  legitimate  ?  On  the  one 
hand,  possession  in  common  is  the  primitive  right ;  on 
the  other,  the  given  fact  is  individual  property.  Here 
we  should  have  an  insoluble  antinomy,  if  possession  in 
common  were  regarded  as  a  fact  that  has  existed  histori- 
cally. It  is  not  a  fact,  however,  it  is  the  command  of 
reason.  Consequently,  the  actual  fact  does  not  go  against 
a  previous  realisation  of  justice.  Till  further  orders, 
it  is  the  only  effective  realisation  of  the  principle  that 
attributes  things  to  persons.  None  the  less  should  it 
be  sanctioned  by  a  contract  between  wills  for  it  to 
become  juridical ;  all  appropriation,  in  the  state  of 
nature,  is  only  provisional. 

The  second  kind  of  possession  refers  to  the  actions 
of  persons,  and  creates  personal  right.  This  is  realised 
by  contract,  the  value  of  which  lies  in  the  stability  and 
simultaneity  of  suprasensible  wills. 

The  third  kind  of  possession  refers  to  persons 
themselves  and  creates  real  personal  right.  Its  domain 
is  the  family.  How  can  a  person  become  a  thing  ? 
Here  we  should  have  an  intolerable  contradiction,  if 
the  owner  of  the  person  did  not  restore  that  person's 
dignity,  by  also  giving  himself,  re-establishing  by  an 
act  of  freedom  the  moral  order  which  is  threatened  by 
nature.  Thus,  marriage  is  the  only  legitimate  relation- 


KANT  305 

ship   between  the  sexes,   for    it   alone   safeguards   the 
dignity  of  the  woman. 

As  regards  public  or  civil  right^  Kant  lays  it  down 
as  a  principle  that,  since  the  natural  state  of  mankind 
is  war,  it  is  necessary  to  constitute  a  civil  society  in 
order  to  make  possible  a  regime  of  right.  The  laws 
that  create  such  a  regime  are  divided  into  -political  right^ 
the  right  of  nations  and  cosmopolitical  right. 

Political  right  rests  exclusively  on  the  idea  of 
justice.  Sovereignty  originally  belongs  to  the  people  ; 
the  State  can  only  be  the  result  of  a  contract,  by  which 
men  give  up  their  natural  freedom,  to  recover  it  intact 
under  a  legal  regime.  But  this  contract  is  not  an 
historical  fact,  it  is  an  idea  of  reason  :  this  is  the 
point  of  view  that  both  citizens  and  legislator  must 
adopt,  in  the  performance  of  their  respective  tasks. 
Consequently  power  must  be  obeyed  without  inquiring 
into  its  origin.  However  vicious  a  social  form  may 
be,  it  is  not  a  falling  away  from  a  primordial  state  of 
justice  :  it  is  the  degree  of  reality  that  the  idea  of  right 
has  been  able  to  reach  in  the  world  of  time.  To  amend 
it  by  reform  is  legitimate,  but  not  to  overthrow  it  by 
revolution. 

If  such  is  its  principle,  the  State  has,  for  its  mission, 
to  guarantee  the  natural  rights  of  man.  It  will  trouble 
itself  about  morals  only  in  so  far  as  they  interest  public 
order.  It  will  respect  religious  beliefs,  but  will  resist 
political  influence  on  the  part  of  the  Churches.  It  has 
the  right  to  abolish  all  privileges  which  are  only  facts 
devoid  of  rational  foundation. 

The  realisation  of  the  idea  of  the  State  requires  the 
division  of  power  into  legislative,  executive  and  judicial 
power.  The  most  important  of  these  is  the  legislative, 
which  ought  to  be  the  full  complete  expression  of  the 


306  STUDIES  IN  PHILOSOPHY 

collective  will.  Government  is  more  or  less  despotic 
in  proportion  as  it  departs  from  the  representative 
system.  The  republic,  an  ideal,  rational  form,  is  a 
government  that  is  representative  in  its  three  powers. 
In  practice,  Kant,  as  became  a  loyal  subject  of  Frederick 
II.,  recognises  an  autocratic  regime,  wherein  power, 
thanks  to  the  generosity  of  the  prince,  is  in  conformity 
with  the  philosophical  principles  of  right. 

Ever  relying  on  the  idea  of  justice,  Kant  regards 
penal  right  as  based  not  on  utility  but  on  reward  ;  he 
defends  the  death  penalty  against  the  sentimentality  of 
Beccaria. 

The  right  of  nations  extends  to  States — with  certain 
modifications — the  relations  which  public  right  sets  up 
between  individuals.  Their  original  condition  is  war, — 
not  a  regime  of  right.  In  order  that  juridical  relations 
may  be  established  between  them,  they  must  form  and 
maintain,  in  accordance  with  an  original  contract,  an 
alliance  or  federation,  by  which  they  undertake  not  to 
intervene  in  internal  discords,  and  also  to  unite  for 
mutual  protection  against  external  attacks. 

Finally,  cosmopolitical  right  insures  for  each  man 
the  power  to  enter  into  communication  with  all. 
Nations  should  allow  foreigners  access  to  their  terri- 
tories. Colonisation  is  a  right ;  all  the  same,  it  should 
not  violate  any  acquired  right  :  injustice  is  not  per- 
mitted, even  with  the  object  of  extending  the  domain 
of  justice. 

Right  comes  indefinitely  near  to  morals,  without 
being  able  to  attain  to  it.  It  requires  that  it  be 
possible  for  the  rule  of  our  external  actions  to  be 
set  up  as  a  universal  law  :  morals  puts  forward  the 
same  demand  as  concerns  the  maxim  itself,  the  internal 
principle  of  our  actions.  Thus,  the  duties  of  virtue 


KANT  307 

differ  from  those  of  right,  both  in  their  object, — for 
they  determine  the  intention,  not  the  act,  whereas  the 
duties  of  right  determine  the  act  and  not  the  intention  ; 
this  is  expressed  by  saying  that  the  latter  are  strict 
and  the  former  accommodating  ; — and  in  their  motive, 
for  the  subject  imposes  them  upon  himself,  whereas 
duties  of  right  are  imposed  by  external  compulsion. 

What  are  the  ends  that  are,  at  the  same  time, 
duties  ?  There  can  only  be  two  :  one's  own  perfection, 
and  the  happiness  of  others.  I  ought  to  aim  after  my 
own  perfection,  not  happiness  :  whereas,  I  ought  to  aim 
after  the  happiness,  not  the  perfection  of  others.  As  a 
matter  of  fact  I  can  neither  make  myself  happy,  nor 
can  I  work  out  the  will  of  others  ;  whereas  the  deter- 
mination of  my  will  does  concern  me,  as  also  the 
condition  of  the  rest  of  mankind. 

The  detailed  list  of  duties  will  comprise  nothing 
referring  to  family  or  State.  Kant  sees  in  these  com- 
munities only  juridical  relationships,  so  he  has  already 
said  all  he  wished  about  them,  in  the  theory  of  right. 
Morals  will  be  essentially  individual  and  social. 

We  have  duties  only  towards  ourselves  and  other 
men,  not  towards  God  or  the  animal  world.  For  we 
can  be  under  obligation  only  towards  persons  who  are 
objects  of  experience  to  us  :  and  one  or  the  other  of 
these  two  conditions  falls  through,  in  the  case  of  beings 
superior  or  inferior  to  ourselves. 

Respect  for  human  dignity,  in  oneself  and  in  others, 
is  the  one  preeminent  duty.  This  duty  admits  neither 
of  conditions  nor  of  temperament  :  it  is  absolute  and 
immutable.  Love  of  one's  neighbour,  and  benevolent 
feelings  in  general,  can  become  duties  only  in  so  far 
as  we  are  dealing  with  active  benevolence,  not  with  the 
sympathy  of  complaisance  or  pathological  love. 


3o8  STUDIES  IN  PHILOSOPHY 

From  these  principles  proceed  such  maxims  as  the 
following  : — Allow  no  one  with  impunity  to  trample 
your  right  under  foot.  Never  incur  a  debt,  without 
giving  security.  Lying,  whether  to  others,  or  more 
especially  to  oneself,  is  moral  suicide.  Meanness  is 
unworthy  of  man  ;  he  who  crawls  like  a  worm  cannot 
complain  if  he  is  trampled  upon.  The  violation  of  the 
duty  of  love  is  only  a  sin,  that  of  the  duties  of  respect 
is  a  vice  ;  for  in  the  latter  case  man  is  insulted,  in  the 
former  he  is  not.  Moral  gymnastics  is  not  mortifica- 
tion, it  is  the  will  practising  to  overcome  one's  inclina- 
tions so  as  not  to  be  hindered  by  them,  and  joyfully 
exulting  in  its  regained  freedom. 

Naturally  following  on  the  metaphysics  of  morals, 
comes  religion,  not  as  implied,  but  as  demanded  by 
morality.  Religion  consists  in  looking  upon  moral  laws 
as  though  they  were  divine  commandments.  It  cannot 
increase  our  knowledge  either  of  God  or  of  nature  ;  it 
ought  not  to  aim  at  this.  Its  sole  object  is  to  extend 
the  ascendency  of  the  moral  law  over  the  will. 

Thus  understood,  it  is  in  conformity  with  and 
sanctioned  by  reason.  But  the  positive  religions  add 
on  to  the  moral  postulates  and  the  law,  traditional  and 
statutory  elements  :  it  is  important  for  us  to  find  out 
how  far  this  addition  can  be  justified  by  reason. 

If  we  examine  the  Christian  religion  :  an  excellent 
form  of  religion,  we  find  four  essential  ideas  in  it  :  that 
of  original  sin,  that  of  Christ,  that  of  the  Church,  and 
that  of  worship.  What  value  have  these  ideas  ? 

In  the  dogma  of  original  sin  lies  concealed  a  phil- 
osophic truth.  There  are  two  characters  in  each  of  us : 
the  empirical  and  the  intelligible.  The  vices  of  the 
one,  whilst  attesting  an  innate  tendency  towards  evil, 
indicate  a  radical  failing  in  the  other.  This  failing 


KANT  309 

consists  in  reversing  the  order  which  ought  to  regulate 
the  relations  between  sensibility  and  reason  ;  in  placing 
the  latter  at  the  service  of  the  former.  Morality,  to 
the  one  who  has  been  guilty  of  this  failing,  cannot  from 
that  time  be  anything  else  than  conversion,  a  new  birth, 
as  it  is  called  in  Christian  theology.  In  this  sense,  dogma 
is  justified. 

The  idea  of  Christ,  too,  is  accepted  by  criticism,  if 
by  Christ  we  mean  the  ideal  of  the  human  person. 
This  ideal  descends  from  heaven  to  earth,  not  historic- 
ally, of  course,  but  in  the  sense  that,  whilst  belonging 
to  the  intelligible,  it  is  manifested  in  the  sensible  world. 
This  ideal  redeems  us,  for  whereas  punishment  affected 
the  guilty  man,  it  is  the  man  who  is  converted  by  the 
conception  of  the  ideal,  the  new  man,  who  suffers  and 
struggles  in  order  to  free  the  former  man  from  evil. 
The  good  man  takes  upon  himself  the  sins  of  the 
wicked,  and  stands  in  his  place  before  the  judge. 

The  Church,  also,  is  recognised  by  reason,  so  far  as 
it  is  an  association  whose  members  mutually  fortify 
themselves  in  the  practice  of  duty,  both  by  example 
and  by  the  declaration  of  a  common  moral  conviction. 
In  itself,  it  is  one,  like  rational  faith,  but  human  weak- 
ness demands  that  there  be  added  to  this  faith,  in 
order  to  make  it  sensible,  various  historical  dogmas 
that  claim  a  divine  origin.  Hence,  a  multiplicity  of 
churches,  and  antagonism  between  heretics  and  ortho- 
dox. The  history  of  the  Church  consists  entirely  of 
the  struggle  between  rational  and  positive  faith  ;  and 
the  goal  to  which  it  is  advancing  is  the  effacement  of 
the  latter  by  the  former. 

Finally,  worship  itself  is  a  rational  matter,  provided 
it  be  assigned  a  place  in  moral  intention  and  in  the 
realisation  of  that  intention.  All  that  man  thinks  he  can 


310  STUDIES  IN  PHILOSOPHY 

add  on  to  virtue  in  order  to  honour  God  is  but  false 
worship  and  vain  observance.  The  consequence  of  the 
illusory  value  attributed  to  this  false  worship  is  the 
subordination  of  the  laity  to  the  Church,  and  all  the 
evils  to  which  this  subordination  gives  rise,  such  as 
hypocrisy  and  fanaticism.  The  positive  faith  the  Church 
enjoins  has,  for  its  true  object,  to  make  itself  super- 
fluous. This  faith  has  in  the  past  been  necessary  as  a 
vehicle  ;  it  remains  useful  until  mankind  comes  of  age. 
Once  this  time  arrives,  however,  the  leading  strings 
of  tradition  become  mere  fetters.  The  very  ecclesiastic 
who,  as  a  minister  of  religion,  is  bound  down  to  symbols, 
as  a  scholar  has  the  right  to  examine  dogmas  :  to  decree 
the  unchangeableness  of  statutory  faith  would  be  an 
outrage  on  human  nature. 

V. — APPLICATIONS  OF  THE  METAPHYSICAL  DOCTRINE 

It  is  Kant's  constant  preoccupation  to  unite  concrete 
reality  to  practice.  His  principles,  obtained  by  meta- 
physical analysis  from  the  given  itself,  ought  rationally 
to  reconstitute  and  govern  the  given.  In  the  material 
order  of  things,  he  sought  the  transition  from  meta- 
physics to  physics  ;  so  also  in  the  moral  order  he  again 
descends  from  idea  to  action. 

In  this  connection,  the  history  of  mankind  is  his 
principal  theme.  He  purposes  deducing  its  main  phases, 
not  describing  them.  Here,  too,  he  makes  a  distinc- 
tion between  the  natural  and  the  moral  history  of  man  ; 
the  latter  having  its  beginning  in  the  former. 

On  the  subject  of  natural  history,  Kant  deals  with 
the  question  of  races.  Is  there  a  distinction  amongst 
the  human  races,  of  such  a  kind  that  one  of  them 


KANT  311 

should  have  the  right  to  claim  for  itself  alone  the 
dignity  of  manhood  and  reduce  the  rest  to  a  state  of 
slavery  ?  The  question  is  answered  by  a  consideration 
of  origins.  Fecundation  is  possible  between  human 
beings  of  all  races ;  consequently  they  have  one  identical 
origin  and  form  only  one  species.  Races  are  stable 
varieties  ;  unalterable  by  intermixture  and  transplanta- 
tion. They  have  become  differentiated  by  adapting 
themselves  to  climatic  conditions.  As  there  are  four 
climates,  so  there  are  four  races  :  the  white,  the  yellow, 
the  black  and  the  red.  In  the  formation  of  these  races, 
external  causes  have  played  an  indispensable  role,  but 
these  alone  could  not  have  brought  about  stable  changes ; 
they  merely  developed  the  internal  dispositions  of  the 
species.  The  real  cause  of  the  existence  of  races,  is 
man's  capacity  for  adapting  himself  to  external  con- 
ditions. 

In  answer  to  the  attacks  of  Forster,  who  would 
explain  life  by  none  but  geological  causes,  Kant,  from  the 
year  1788  onwards,  affirmed  the  necessity  of  a  special, 
immaterial  principle  as  alone  conforming  to  the  require- 
ments of  criticism.  To  attribute  to  matter  a  power  of 
organisation  which  observation  could  not  find  in  it, 
is  to  reject  the  guiding  clue  of  experience.  Doubtless 
Forster's  explanation  is  neither  absurd  nor  impossible, 
but  it  goes  beyond  our  means  of  knowing.  The  only 
finality  we  can  grasp  is  in  ourselves,  in  our  conscious 
activity  ;  nothing  authorises  us  to  admit  that  an  uncon- 
scious thing  has  the  power  of  acting  with  a  view  to  an 
end.  We  do  not  know  what  causes  life,  but  we  explain 
it  by  finality  :  this  is  the  point  of  view  taken  by 
criticism. 

Whereas  the  natural  history  of  man  goes  back  to 
his  origin,  moral  history  considers  his  end.  The  philo- 


3i2  STUDIES  IN  PHILOSOPHY 

sophy  of  history  finds  its  principle  in  the  idea  of  this 
end,  as  natural  philosophy  does  in  the  idea  of  attraction. 
Now,  the  development  of  reason,  the  essence  of  man, 
cannot  tend  to  anything  else  than  the  establishment  of 
a  regime  of  freedom,  i.e.  to  the  realisation  of  justice. 
Consequently  the  historian  ought  to  find  in  facts  the 
various  phases  of  the  realisation  of  justice. 

History  begins  when  man  becomes  a  moral  being, 
i.e.  when  he  acts  by  will  instead  of  by  instinct.  His 
primitive  state  was  one  of  innocence  ;  his  abode,  paradise. 
He  formed  one  with  nature,  wherein  his  will  was  buried. 
The  awakening  of  his  will  showed  itself  by  a  desire 
for  rule,  an  act  of  pride,  rebellion  against  the  nature 
to  which  he  was  united.  Original  sin  is  freedom's 
first  step.  From  that  time,  a  new  life  begins  for 
man.  In  order  to  dominate  nature,  he  must  work. 
From  work  there  arise  discord,  society,  property,  civil 
inequality  :  civilisation  has  succeeded  a  state  of  nature. 
What  does  this  new  condition  stand  for  ?  Had  human 
activity  no  other  end  than  individual  happiness,  then 
Rousseau  would  be  quite  right  in  longing  for  a  return 
to  the  paradise  of  innocence.  But  what  man  wills  is 
to  be  free,  and  effective  freedom  can  be  found  only  in 
the  disinterested  agreement  of  wills,  on  the  ground  of 
reason.  Now,  civilisation,  the  conflict  of  wills,  is  the 
necessary  antecedent  of  their  reconciliation.  The  reign 
of  justice,  the  source  of  moral  harmony,  is  the  third 
phase  of  universal  history. 

In  the  realisation  of  this  progress  of  freedom,  the 
will  is  not  left  to  itself.  It  is  aided  by  nature  ;  con- 
sequently, progress  is  constant  and  has  the  character  of 
a  natural  law.  A  law  beneficent  and  necessary,  for 
were  man  to  believe  that  his  works  perish  wholly  with 
himself  how  could  he  keep  alive  an  earnest  desire  to 


KANT  313 

work  for  the  good  of  mankind  ?  Nature  stirs  up  man 
to  quit  nature  ;  she  stimulates  his  freedom.  She  is  an 
artist,  a  providence,  capable  of  bringing  forth  good  out 
of  evil.  She  makes  men  selfish  and  violent,  and  violence 
engenders  war  ;  but  war  calls  a  judicial  regime  into 
existence.  She  separates  men  through  differences  in 
constitution,  language  and  religion  ;  but  these  differ- 
ences render  universal  domination  impossible.  Whilst 
evil  succumbs,  sooner  or  later,  to  the  contradiction 
within  itself;  good,  which  reason  substitutes  therefor, 
when  once  established,  continues  and  increases,  because 
it  is  in  harmony  with  itself.  For  logic  is  the  one 
supreme  force.  At  first,  man  wills  union,  and  believes 
himself  wise  ;  but  nature  knows  better  what  is  suitable 
for  him,  she  wills  a  state  of  war. 

The  first  object  of  this  collaboration  between  nature 
and  will,  is  the  establishment  of  the  rational  State,  a 
combination  of  freedom  and  legality.  The  second 
object  is  the  establishment  of  an  Amphictyonic  council 
of  nations,  ensuring  the  maintenance  of  peace.  Without 
such  an  institution,  mankind  cannot  advance  to  its  goal. 
War  is  a  return  to  a  state  of  nature.  In  the  ideal  of 
reason  is  implied  the  idea  of  eternal  peace.  If  this 
object  is  unreal isable,  then  Rousseau  is  right  in  advocat- 
ing a  return  to  a  savage  state.  Better  barbarism  than 
culture  without  morality. 

But  is  not  this  a  purely  theoretical  conception  ? 
Will  real  humanity  accept  such  views  ?  Has  not  Hobbes 
shown  that  the  real  man  is  influenced  only  by  interests, 
not  by  ideas  ?  Such  a  doctrine  must  be  utterly  rejected  ; 
the  belief  must  not  gain  ground  that  what  is  good 
in  theory  can  ever  be  impossible  or  evil  in  practice. 
What,  indeed,  is  not  practical  is  that  unlimited  power 
Hobbes  confers  on  sovereigns,  and  the  rebellion  he 


STUDIES  IN  PHILOSOPHY 

admits  of  in  subjects.  Interests,  certainly,  in  the  State, 
should  have  a  place  of  their  own,  but  does  it  follow 
that  principles  should  be  excluded  ?  Can  one  not  be 
both  as  wise  as  the  serpent  and  as  harmless  as  the 
dove  ?  To  the  man  who  guards  against  both  idealism 
and  empiricism,  the  real  and  the  ideal,  instead  of  ex- 
cluding, include  each  other,  and  politics  ceases  to  be 
incompatible  with  morals.  There  is  a  practical  means 
of  bringing  the  former  into  harmony  with  the  latter  : 
publicity.  Whosoever  thinks  he  can  be  useful  to  his 
country  ought  to  seek  publicity.  Now,  only  what  is 
in  conformity  with  justice  can  bear  publicity.  Here 
as  elsewhere,  universality  is  the  point  of  contact 
between  the  real  and  the  rational,  the  form  and  token 
of  truth. 

According  to  this  theory,  what  is  the  present  phase 
of  the  history  of  the  human  species  ?  It  is  the  phase 
of  enlightenment  {Aufkl&rung),  and  its  characteristic  is 
the  emancipation  of  the  intelligence.  Man,  reflecting 
upon  himself,  finds  that  there  is  a  contradiction  between 
his  reasonable  nature  and  his  position  as  a  minor  : '  he 
makes  an  effort  to  liberate  his  reason.  Sapere  aude  is 
his  motto. 

The  progress  of  enlightenment  cannot  be  realised  by 
overthrowing  political  institutions,  by  revolution,  which 
has  no  other  result  than  the  substitution  of  new  for  old 
prejudices.  Personal  reflection  alone  can  truly  enlighten 
a  man.  Consequently  freedom  to  think  and  make 
known  his  thoughts  is  the  condition  of  the  progress  of 
enlightenment. 

How  can  this  freedom  be  reconciled  with  the  rights 
of  the  State  ?  Here,  a  distinction  must  be  made  between 
man  as  a  citizen  of  a  limited  community,  and  man  as 
a  citizen  of  the  whole  world.  In  his  dealings  with  the 


KANT  315 

members  of  his  community,  man  is  bound  to  submit  to 
the  statutes  by  which  it  is  governed  ;  but  as  a  citizen 
of  the  world,  he  is  free.  As  such  a  person,  indeed,  he 
speaks  from  the  summit  of  reason,  for  the  generality 
of  reasonable  beings,  whereas  as  a  citizen  of  a  State, 
he  limits  his  action  to  some  particular  place  and  time. 
Only  by  identifying  itself  with  the  universal  does  the 
will  attain  to  freedom.  Therefore  each  citizen  will 
unresistingly  pay  taxes,  though  retaining  the  right  to 
dispute  such  payment.  The  teacher,  as  an  official,  will 
respect  such  symbols  as  are  recognised  in  his  own 
country  ;  but  as  a  scholar,  he  will  have  the  right  to 
criticise  all  doctrines.  In  accordance  with  these  prin- 
ciples, the  rights  both  of  legislators  and  of  citizens  are 
clearly  defined. 

And  so,  fully  maintaining  the  harmony  of  nature 
and  of  freedom  in  the  moral  history  of  man,  Kant 
guards  against  asserting  that  progress  is  a  mere  develop- 
ment of  natural  powers.  To  his  mind,  the  Leibnitzian 
theory  of  Herder  is  radically  erroneous.  In  nature 
dwells  the  means  ;  but  the  end,  which  is  the  spring  of 
progress,  can  come  only  from  moral  reason,  superior  to 
nature.  This  is  why  the  moral  ideal  can  never  be 
expressed  by  the  individual  as  such ;  it  cannot  be 
represented  except  in  the  whole  of  mankind.  True 
history  is,  of  necessity,  universal.  Certainly,  the  indi- 
vidual is  a  reality,  but  in  the  whole,  there  is  something 
that  goes  beyond  it,  and  only  by  union  with  the  whole 
can  it  attain  to  freedom. 

Not  content  with  expounding  his  general  views  as 
to  the  ends  of  human  activity,  Kant,  in  some  things, 
deals  with  practice  proper.  We  here  refer  to  his  ideas 
on  education  and  university  instruction. 


316  STUDIES  IN  PHILOSOPHY 

It  is  impossible  for  education,  in  its  present  state, 
to  satisfy  him.  It  neglects  the  will,  drills  and  over- 
burdens the  mind  instead  of  moulding  it  for  reflection. 
Here  a  radical  reform  is  necessary.  The  pedagogic 
theories  of  Rousseau,  the  practical  attempts  of  Basedow 
come  just  at  the  right  time  to  support  his  criticism. 
He  is  passionately  in  favour  of  these  innovators,  and 
demands  the  organization  of  elementary  schools  as  the 
indispensable  condition  of  reform.  But  he  remains 
himself,  even  on  this  ground,  subordinating  all  authori- 
tative direction  to  moral  ends. 

The  body,  he  tells  us,  ought  to  be  hardened  and 
exercised,  subjected  to  such  discipline  as  will  make  it 
the  powerful  and  obedient  auxiliary  of  the  mind.  Let 
the  child  grow  up  in  perfect  freedom,  but  at  the  same 
time  teach  him  to  moderate  his  movements  :  one  can- 
not accustom  oneself  too  early  to  live  according  to  rule. 

As  regards  the  intellect,  a  sane  education  awakens 
and  guides  the  mental  faculties  instead  of  loading  the 
memory  with  facts.  There  are  two  exercises  of  the 
faculties  :  the  one,  which  is  free,  is  play  ;  the  other, 
which  is  imposed  from  without,  is  work.  The  latter 
is  obligatory  in  itself,  and,  in  instruction,  it  could  not 
be  replaced  by  the  former.  The  faculty  of  intuition 
should  be  formed  before  the  understanding.  Thus,  all 
instruction  will  at  first  be  intuitive,  representative,  tech- 
nical. A  beginning  may  be  made  with  geography.  So 
far  as  it  has  the  cultivation  of  the  understanding  for  its 
object,  instruction  will  be  Socratic  and  catechetical.  It 
will  go  to  the  root  of  things,  and  make  the  pupil  really 
master  of  his  knowledge.  A  robust  intellect  is  the 
condition  of  a  will  that  is  free. 

Paedagogics  has  the  formation  of  the  moral  per- 
sonality as  its  end.  Here  education  is  needed,  for 


KANT  317 

virtue  is  not  innate.  This  education  comprises  moral 
instruction  and  its  corresponding  practice. 

Moral  instruction  is  catechetical.  Aiming  at  demon- 
strating obligatory  laws,  it  proceeds  by  principles,  not 
examples  :  if  examples  come  in,  that  is  only  in  order 
to  prove  the  principles  to  be  really  applicable.  Kant 
left  in  writing  a  fragment  of  moral  catechism,  wherein 
the  pupil,  prompted  by  questions,  discovers  for  himself 
moral  conceptions  of  life. 

Practice,  or  moral  ascetics,  cannot  create  morality, 
which  must  come  from  ourselves  ;  it  does,  however, 
produce  in  man  the  disposition  that  favours  morality. 
It  aims  at  a  hardening  process,  for  effeminacy  or  indol- 
ence is  opposed  to  virtue.  Instead  of  destroying  the 
will,  it  strengthens  it.  It  makes  us  masters  of  ourselves, 
contented  and  happy.  Moral,  education  tends  to 
develop  the  inner  aversion  to  evil,  self-esteem  and 
dignity,  the  domination  of  reason  over  the  senses. 
It  does  not  reward,  but  it  punishes.  It  never  humiliates, 
lest  it  make  the  child  despise  himself,  except  when  he 
has  been  guilty  of  that  one  fault  which  effectively  degrades 
mankind,  to  wit :  falsehood.  In  all  things  it  puts  for- 
ward the  moral  motive,  the  law  of  duty  itself,  certain 
that  this  motive,  when  set  forth  in  all  its  purity,  will 
be  more  powerful  than  any  material  stimulus,  any 
assurance  of  benefit  or  harm. 

With  paedagogics  we  may  compare  the  question  of 
university  instruction.  On  this  point,  too,  the  Critique 
throws  fresh  light.  A  University  consists  of  four 
Faculties :  Theology,  Law,  Medicine,  the  so-called 
superior  Faculties,  and  Philosophy,  the  so-called  inferior 
Faculty.  Between  the  first  three  and  the  fourth,  con- 
flict naturally  arises.  The  object  of  the  latter,  indeed, 


3i 8  STUDIES  IN  PHILOSOPHY 

does  not  differ  from  those  of  the  former,  but  the  one 
studies  from  a  universal  and  theoretical  point  of  view 
what  the  others  study  from  a  special  and  immediately 
practical  one.  This  gives  rise  to  jealousy  and  rivalry. 
Each  of  the  two  sides,  claiming  the  whole  realm  of 
knowledge,  repels  the  other  as  a  usurper.  The  title  of 
superior,  borne  by  the  first  three  Faculties,  is  nothing 
less  than  the  superiority  that  tradition  attributes  to  the 
positive  over  the  rational.  Is  this  hierarchy  justified  ? 

The  conflict  between  theologians  and  philosophers 
is  based  upon  the  use  to  be  made  of  the  holy  Scrip- 
tures. The  Critique  does  not  deny  the  legitimacy  and 
utility  of  the  sensible  vehicle  of  religious  truth  ;  but  it 
claims  for  reason  the  right  to  distinguish,  in  the  Scrip- 
tures, between  the  moral  and  eternal  substratum,  and 
the  sensible  outer  form,  made  up  of  narratives  and 
contingent  circumstances.  To  understand  the  Scrip- 
tures is  to  interpret  them  in  a  moral  sense.  Theology 
presupposes  this  mode  of  interpretation,  and  so  cannot 
condemn  it.  How,  indeed,  does  it  distinguish  true 
from  false  revelation,  except  by  the  rational  idea  of 
God.  How  can  it  maintain  the  divine  character  of 
consecrated  texts,  in  detail,  except  by  making  frequent 
use  of  an  allegorical,  moral  interpretation  ? 

The  conflict  between  philosophers  and  jurisconsults 
is  based  on  respect  for  law :  the  Critique  shows  that 
legality  has  a  good  foundation,  consequently  it  condemns 
the  revolutionary  spirit.  But,  in  addition,  it  claims  the 
right  to  examine  existing  laws.  And  who  can  refuse  it 
this  right?  Jurisconsults,  in  order  to  attain  to  their 
practical  ends,  need  to  know  whether  mankind  is  going 
backwards  or  forwards,  or  remaining  stationary.  Now, 
this  is  a  question  that  cannot  be  solved  empirically  :  it 
concerns  reason.  And  reason  answers  it  by  postulating 


KANT  319 

indefinite  progress  in  the  name  of  the  moral  law.  But 
what  if  the  commandment  is  only  an  idea  incapable  of 
realisation  !  Experience,  under  the  guidance  of  reason, 
removes  the  doubt.  Beneath  our  very  eyes,  we  can  see 
where  reason  and  history  coincide.  There  is  one  fact 
which  is  at  the  same  time  an  idea.  This  fact  is  the 
French  Revolution.  Whatever  comes  of  this  enter- 
prise, writes  Kant  in  1798,  whether  it  succeeds  or  fails, 
it  stirs  up  a  sympathy  that  is  akin  to  enthusiasm  in  all 
who  witness  it  by  reason  of  the  object  it  has  in  view  : 
now,  a  purely  moral  ideal  is  alone  capable  of  affecting 
the  soul  of  man  in  this  way.  The  Revolution  is  the 
effort  of  man  to  create  a  rational  State,  it  is  the 
eternal  entering  into  time.  Such  a  phenomenon,  once 
witnessed,  can  never  be  forgotten. 

The  problem  for  philosophers  and  doctors  to 
solve,  is  whether  the  art  of  healing  depends  on 
experience  alone,  or  whether  reason  has  any  share  in  it. 
Now,  the  Critique  demonstrates  that  reason  may  be 
will,  and  that  will  bears  some  relation  to  phenomena. 
Reason,  then,  must  also  possess  a  healing  virtue.  And, 
indeed,  man  can  do  a  great  deal  in  modifying  his 
physical  condition  by  the  sole  exercise  of  his  will. 
Here,  Kant  relates  his  personal  experience  :  by  moral 
force,  he  is  able  to  keep  himself  free  from  hypochondria, 
and  even  to  master  spasmodic  states.  Once  the  disease 
has  entered,  the  will  may  be  unequal  to  the  task  before 
it,  but  at  all  events  it  can  do  much  to  prevent  it,  and 
keep  the  body  in  a  state  of  health.  The  will  is  the 
first  condition  of  health.  Far  from  reason  ever  being 
the  servant  of  experience,  it  is  the  latter  that,  under  all 
circumstances,  borrows  from  the  former  its  truth  and 
possibility. 


320  STUDIES  IN  PHILOSOPHY 

VI. — KANT'S  INFLUENCE 

The  Kantian  philosophy  had  difficulty  in  making  a 
way  for  itself  in  the  field  of  thought  already  occupied 
by  the  Leibnitzo-Wolfian,  the  English,  French  and 
popular  philosophies,  without  counting  the  increasingly 
flourishing  positive  sciences.  Kant  did  not  deceive 
himself  as  to  the  strange  novelty  of  his  work,  which 
met  with  its  first  favourable  reception  at  lena,  thence 
spreading  by  degrees  all  over  Germany  and  finally 
throughout  the  world.  Not  only  was  metaphysical 
speculation  renewed,  as  it  were,  thereby  :  most  of 
the  departments  of  intellectual  activity  felt  its  in- 
fluence. 

In  Germany,  the  history  of  Kantianism  forms  an 
important  element  in  the  general  history  of  ideas  and 
sciences. 

Amongst  its  first  opponents  may  be  cited  :  Selle  and 
Weishaupt,  followers  of  Locke ;  Feder,  Garve  and 
Tiedemann,  electics ;  Platner,  Mendelssohn,  Nicolai 
and  Meiners,  representatives  of  popular  philosophy  ; 
Ernst  Schulze,  the  skeptic  ;  Jacobi,  the  philosopher  of 
belief,  and  along  with  him,  Hamann  ;  Herder,  who 
reconciled  nature  with  history.  The  main  reproach 
these  philosophers  bring  against  Kant  is  that  the 
affection  or  action  of  things  on  sensibility,  implied  by 
his  system,  is  made  impossible  by  the  abolition  of  all 
casual  connection  between  "  things-in-themselves  "  and 
the  feeling  subject.  Consequently,  the  system  was 
alleged  to  be  fundamentally  contradictory. 

Among  Kant's  immediate  disciples  may  be  mentioned 
Johannes  Scultz,  the  first  commentator  on  the  Critique 
of  Pure  Reason  ;  Karl-Leonhard  Reinhold  ;  Krug  ; 
Fries,  who  attempted  to  give  criticism  a  psychological 


KANT  321 

basis ;  Salomon  Maimon,  who  deduced  from  con- 
sciousness both  the  matter  and  the  form  of  our  re- 
presentations and  so  abolished  the  "  thing-in-itself "  ; 
Beck,  and  Bardili. 

Whether  in  the  way  of  development  or  by  combining 
with  foreign  elements,  Kantianism  gave  birth  to  a 
number  of  important  systems.  The  philosophies  of 
Fichte,  Schelling  and  Hegel  are  so  many  stages,  as  it 
were,  in  a  connected  line  of  thought  dealing  with  the 
problems  he  raises.  The  subjective  idealism  of  Fichte 
deduces  the  theoretical  from  the  practical  I,  regarded 
as  originally  unconscious,  and  so  makes  of  none  effect 
the  concept  of  the  "  thing-in-itself."  Schelling  objects 
to  call  I  this  first  principle  of  Fichte,  for,  in  reality, 
it  is  neither  subject  nor  object  :  to  his  mind,  the 
principle  is  absolute  identity,  no  less  superior  to  the 
I  than  to  the  not-I,  an  identity  that  is  first  realised 
as  nature  and  afterwards  as  spirit :  his  system  is  ob- 
jective idealism.  Hegel  establishes,  defines  and 
methodically  develops  the  principle  of  this  new 
idealism.  The  absolute  cannot  be  absolute  identity, 
otherwise  it  would  be  immovable  :  it  must  of  necessity 
be  spirit.  Its  movement  is  the  methodical  effort  it 
makes  to  remove  the  ever-recurring  contradictions 
developed  by  reflection  deep  in  its  own  nature.  The 
philosopher's  dialectic  gives  itself  up  to  the  objective 
movement  of  concept  and  thus  brings  forth  in  suc- 
cession :  logic,  the  philosophy  of  nature  and  the  philo- 
sophy of  spirit.  Idealism  has  become  absolute. 

Apart  from  this  somewhat  organic  development, 
several  German  systems  sprang  from  a  fusion  of  Kant- 
ianism with  other  doctrines. 

Schleiermacher,  placing  Spinoza,  Plato  and  Christianity 
alongside  of  Kant,  compares  being  and  thought,  and 


322  STUDIES  IN  PHILOSOPHY 

regards  space,  time  and  causality  as  forms  both  of 
things  and  of  knowledge.  God  becomes  the  unity  of 
the  universe.  Supreme  good,  the  unity  of  the  real  and 
the  ideal  is,  in  morals,  substituted  for  the  purely  formal 
principle  of  Kant. 

Herbart  draws  upon  Kant,  the  Eleatics,  Plato  and 
Leibnitz.  Like  Kant,  he  sees  in  philosophy  the 
criticism  of  experience.  According  to  him,  however, 
the  "  thing-in-itself "  is  not  inaccessible.  It  is  obtained 
in  its  true  form,  if  we  eliminate  from  the  data  of 
experience  all  the  self-contradictory  and  consequently 
subjective  elements  found  therein.  It  consists  of  a 
plurality  of  simple  beings  with  no  real  relation  to  one 
another :  it  is  we  ourselves  who  introduce  relations  and 
a  process  of  becoming. 

Like  Kant,  Schopenhauer  limits  space,  time  and 
causality,  to  phenomena  ;  but  instead  of  considering 
the  independent  reality  of  our  representation  as  in- 
capable of  being  known,  he  places  it  in  will,  as  given  by 
internal  perception. 

All  the  same,  the  difficulties  inherent  in  these 
divers  systems,  more  particularly  the  foolish  claim — 
set  up  by  absolute  idealism — to  build  up  in  detail  the 
laws  of  nature,  very  quickly  brought  these  developments 
of  Kantianism  into  disfavour.  It  was  considered  that 
Kant's  system  of  thought  had  been  perverted  by  his 
successors,  and  that  the  line  of  reflection  must  be 
picked  up  just  where  Kant  had  dropped  it.  Such  was 
the  idea  of  an  important  school  of  philosophers,  called 
the  Neo-Kantian,  especially  after  a  famous  lecture  by 
Zeller  on  the  theory  of  knowledge,  published  in  1862. 
They  proposed  either  to  defend  Kant's  own  principles, 
or  to  develop  them — without  considering  the  great 
metaphysical  systems  that  have  sprung  therefrom — in  a 


KANT  323 

manner  strictly  suited  to  the  spirit  of  our  times.  The 
principal  members  of  this  school  were :  Lange,  Cohen, 
Liebmann,  Bonna  Meyer,  Paulsen,  Krause,  Stadler, 
Riehl,  Windelband,  Schultze.  Most  of  them,  along  with 
Lange,  insisted  especially  on  the  distinction  between 
knowledge  and  belief,  corresponding  to  that  between 
phenomena  and  "  things-in-themselves,"  so  far  as  this 
distinction  guaranteed  the  possibility  of  science,  whilst 
at  the  same  time  limiting  it.  Philosophy  ought  to  be 
a  theory  of  knowledge,  not  a  conception  of  the  world. 
Moral  things  may  be  a  matter  of  faith,  not  of  science. 
With  few  exceptions,  including  Paulsen,  these  philo- 
sophers put  in  the  background,  or  even  neglected 
the  moral  and  religious  part  of  Kant's  work,  and 
emphasised  the  critical  and  antimetaphysical  part. 

Apart  from  philosophy,  Kantianism  in  Germany 
has  long  since  left  its  mark  on  the  majority  of  in- 
tellectual disciplines. 

Following  on  Kant,  Schiller  entered  into  philosophic 
speculations  on  esthetics,  endeavouring  to  define  the 
connection  of  beauty  with  nature  and  morality. 

In  theology,  Kant  initiated  a  moral  rationalism  that 
long  held  sway.  Even  of  recent  years,  Ristchl,  the 
theologian,  has  returned  to  Kant,  protesting  against 
the  metaphysical  fancy  which  claims  to  know  the 
suprasensible. 

In  jurisprudence,  the  Kantian  theories  of  natural 
right  are  found  as  leading  ideas  in  the  works  of 
Hufeland,  Schmalz,  Gros,  Feuerbach,  Rehberg  and 
Zachariae. 

In  science,  Kantianism  has  exercised  a  varied  in- 
fluence according  to  the  way  in  which  it  has  been 
understood.  Radically  idealistic  in  interpretation — 
though,  truth  to  tell,  this  interpretation  was  repudiated 


324  STUDIES  IN  PHILOSOPHY 

by  Kant — there  came  into  being  the  famous  philosophy 
of  nature,  which,  bringing  matter  entirely  within  the 
compass  of  unconscious  thought,  boldly  deduces  the 
phases  of  its  development  from  the  laws  of  the  for- 
mation of  consciousness  itself.  On  the  other  hand, 
the  Kantian  theory  of  experience,  as  the  sole  origin  of 
knowledge,  is  accepted  by  many  modern  scholars  in 
quest  of  a  rational  justification  of  their  own  methods. 

In  mathematics,  the  Kantian  point  of  view  is  char- 
acterised by  the  admission  of  synthetical  a  'priori 
principles,  or  extralogical  rational  principles,  and  in 
particular  by  the  negation  of  the  metageometrical 
space  of  the  Leibnitzians,  as  an  object  of  possible 
intuition. 

In  the  psycho-physiology  of  the  senses,  the  negation 
of  Johannes  Muller,  who  maintains,  in  opposition  to 
empiricism,  the  primitive  character  of  the  representation 
of  space,  is  based  on  transcendental  esthetics. 

Finally,  Kantianism  exercises  considerable  influence 
over  the  political  life  of  Germany.  It  represents  the 
idea  that  reason,  even  in  politics,  is  the  true  norm,  and 
commands  man  to  act  in  accordance  with  the  universal 
idea  of  duty  and  humanity  :  a  highly  philosophical 
doctrine,  which  has  certainly  not  altogether  given  way 
to  that  of  historic  right  and  an  exclusively  national  ideal. 

In  other  countries  besides  Germany,  the  influence 
of  Kant's  philosophy  is  still  great,  though  more  tardy 
in  making  itself  felt,  and  less  profound  in  the  im- 
pression it  has  made. 

In  1773,  Kant  began  to  be  appreciated  in  Stras- 
bourg. In  1796,  the  translation  of  his  works  into  French 
was  begun.  In  1799,  Degerando  sets  forth  his  system. 
Mme.  de  Stae"!  speaks  enthusiastically  of  the  man  she 
looks  upon  as  an  apostle  of  sentimental  spiritualism. 


KANT  325 

In  1818,  Victor  Cousin  lectures  on  Kant's  morals; 
in  1820  he  expounds  on  the  Critique  of  Pure  Reason. 
In  his  own  theory  of  reason,  he  is  indebted  to  Kant 
for  several  of  his  ideas.  After  being  thus  utilised  in 
doctrines  based  on  other  principles,  such  as  electicism, 
positivism,  and  independent  morals,  Kantianism  was 
studied  and  developed  for  its  own  sake,  especially  by 
Renouvier,  Janet,  Lachelier,  and  Pillon.  Renouvier, 
Pillon,  and  Dauriac  advocate,  under  the  name  of 
criticisme,  a  doctrine  which,  in  contradistinction  to 
German  Neo- Kantianism,  emphasises  the  excellence 
of  Kantian  morals.  They  directly  subordinate  theor- 
etical to  practical  reason,  looking  upon  the  will 
as  the  first  principle  of  all  certainty  ;  and  not  only 
that,  but,  doing  away  with  the  noumenon,  they  set 
up  natural  laws  as  ultimate  reality,  and,  following  on 
phenomena,  they  prepare  a  place  for  the  initiative  of 
freedom.  Under  Kant's  inspiration,  also,  M.  Secretan, 
of  Lausanne,  limits  the  rights  of  science,  and  places 
above  it  belief  in  freedom.  In  divers  forms  and 
degrees,  Kantianism  even  now-a-days  is  to  be  found  in 
most  of  the  doctrines  whose  aim  it  is  to  reconcile  science 
with  morals,  without  injuring  either. 

In  England,  Kant's  influence  was  mainly  felt  by 
Hamilton  and  the  agnostics.  Combining  Kant's 
doctrine  with  that  of  Reid,  Hamilton  maintained  that 
the  representation  of  the  absolute  was  impossible  for  a 
mind  limited  to  human  knowledge.  Spencer's  agno- 
sticism, also,  though  dependent  on  positivism,  owes 
much  to  the  Kantian  antinomies.  In  the  realm  of 
psychology  the  revolutionist  school  claims  to  be  the 
reconciler  of  Kantian  apriorism  with  Locke's  empiricism. 
At  the  present  time,  Kant  is  scrupulously  studied  for 
his  own  sake.  In  the  translation  of  the  Critique  of 


326  STUDIES  IN  PHILOSOPHY 

Pure  Reason  which  Max  Miiller  published  in  1 8  8  i ,  he 
declares  the  work  to  be  an  Aryan  monument  as  precious 
as  the  Vedas,  and  says  that  throughout  all  time  it  may 
be  criticized  but  never  ignored. 

In  Italy,  the  Critique  of  Pure  Reason  was  translated 
in  1821-1822,  and  Jose  del  Perojo  translated  it  into 
Spanish  in  1883. 

Looking  at  the  matter  from  a  general  point  of  view, 
what  was  the  historical  role  of  Kant  ?  What  relation 
has  his  philosophy  with  present-day  speculation  ? 

Kant's  main  purpose  was  analogous  to  those  of 
Socrates  and  Descartes.  Socrates  undertook  to  show 
that  practice,  even  regarded  as  the  end  of  human 
activity,  cannot  exclude  science,  because  in  reality  it 
implies  this  latter.  Descartes  grants  that  a  commence- 
ment be  made  with  universal  doubt  :  this  doubt  does 
not  abolish  certainty,  but  rather  creates  a  foundation 
for  it.  Kant,  in  turn,  declares  that  experience  is  the 
starting  point  of  all  our  knowledge.  Are  we  to  conclude 
that  reason  is  a  mere  word  ?  By  no  means,  for 
experience  is  based  on  reason.  And  in  the  very 
development  of  the  doctrine,  analogy  follows  its  course. 
Deduced  from  practice,  the  science  of  Socrates  is 
limited  to  morals  and  the  objects  connected  therewith. 
Cartesian  certainty  at  first  extends  only  to  thought,  the 
condition  of  doubt ;  it  restores  the  objects  that  doubt 
had  overthrown,  only  in  so  far  as  they  are  capable  of 
being  connected  with  thought.  Kantian  criticism,  like- 
wise, allows  to  persist  only  that  in  a  priori  notions 
which  is  required  for  experience ;  it  makes  the  possi- 
bility of  this  latter  the  norm  of  the  entire  use  of  pure 
reason. 

Like  Socrates  and  Descartes,  Kant  contends  that  his 


KANT  327 

method  is  constructive  rather  than  destructive.  Science, 
limited  as  regards  "  things-in-themselves,"  is  at  all  events 
the  abode  of  certainty.  Idealism  melts  away  before 
empirical  realism.  Nor  is  this  all  :  criticism  is  to  give 
even  better  results.  The  very  deduction  that  establishes 
science  allows  morals  to  stand  by  its  side,  without  risk 
of  offending  it.  True,  morals  also  must  put  up  with 
limitation.  It  must  be  based  on  an  exclusively  formal 
principle,  the  simple  notion  of  duty.  But  here  again, 
criticism  restrains  only  in  order  to  secure.  Morality 
may  be  absolute  and  remain  practical,  if  it  has  no  other 
object  than  the  determinations  of  the  will  that  is  free. 
The  insoluble  antinomy  of  mysticism  and  eudemonism 
vanishes  in  the  system  of  rational  autonomy. 

Indeed,  throughout  Kant's  philosophy,  it  is  reason 
that  creates  as  well  as  destroys,  that  supplies  principles 
to  replace  those  it  has  abolished.  In  Descartes,  it  had 
already  discovered  within  itself,  in  its  faculty  of  intuition, 
that  principle  of  certainty  which  it  found  neither  in  the 
senses  nor  even  in  demonstrations.  Kant  shows  us 
reason  making  an  inventory  of  its  content,  and  finding, 
in  its  very  constitution,  all  the  principles  necessary  for 
science  and  morals.  Naturally,  it  does  not  suffice  unto 
itself,  the  absolute  goes  beyond  it.  Its  science,  con- 
sequently, is  relative  ;  its  morals,  in  application,  limited 
to  endless  progress.  None  the  less  does  reason  offer 
man  all  the  resources  he  needs  to  realise  the  ideal  of 
mankind,  for  it  is  freedom  and,  at  the  same  time,  law. 

Such  being  the  essential  elements  of  Kantianism,  this 
philosophy  stands  at  the  term  of  the  rationalistic 
development  which  began  with  Descartes.  Reason, 
according  to  Kant,  drives  to  the  utmost  limits  both  its 
renunciation  of  the  comprehension  of  absolute  being 
and  its  efforts  to  provide,  by  the  principles  it  finds 


328  STUDIES  IN  PHILOSOPHY 

within  itself,  for  the  intuition  in  which  it  is  lacking. 
One  more  step  in  either  direction,  and  rationalism  will 
lose  itself  either  in  skepticism  or  idealism.  Kant,  whilst 
shutting  himself  up  in  the  world  of  time,  claimed  that 
he  found  in  the  heart  of  reason,  which  forms  part 
thereof,  a  means  of  converting  this  world  into  a  symbol 
of  eternal  being. 

Such  is  the  historical  signification  of  his  work. 
Regarded  theoretically,  it  is  of  supreme  interest,  even 
in  our  days. 

The  human  mind,  influenced  by  the  positive  sciences 
and  by  philosophy  alike,  asks  itself  more  than  ever 
what  is  our  relation  to  the  reality  of  things,  and  whether 
or  not  it  is  possible  to  know  that  reality.  Now,  tran- 
scendental idealism  has  an  answer  to  give  to  this 
question.  Beyond  phenomena,  according  to  Kantianism, 
we  can  yet  grasp  the  laws  of  thought  by  which 
phenomena  are  conditioned,  and  constitute  philosophy 
as  a  theory  of  knowledge  ;  but  as  for  forming  an 
ontological  theory  of  the  universe,  as  the  ancients  did, 
we  must  give  up  all  ambition  in  this  direction  :  a  plain 
solution,  and  one  of  grave  consequence,  finding  much 
support  in  present-day  science. 

On  the  other  hand,  the  progress  of  the  positive 
sciences,  in  extent  and  in  certainty,  makes  us  wonder , 
if  whatever  interests  man  cannot  at  least  be  dealt 
with  according  to  the  methods  of  these  sciences,  and  if 
morals  itself  cannot  be  assimilated  thereto.  Kant 
answers  this  question  with  his  stern  dualism,  limiting 
science  in  order  to  give  it  a  basis,  and  establishing  morals 
in  the  domain  opened  up  by  this  very  limitation.  Now, 
neither  the  sovereignty  of  science  in  the  practical  order 
of  things,  nor  the  theoretical  impossibility  of  freedom, 
are,  even  in  these  days,  sufficiently  clearly  demonstrated 


KANT  329 

for  it  to  be  possible  to  relegate  to  the  past  the  Kantian 
solution. 

As  regards  the  philosophy  of  science,  Kantianism 
deals  just  with  those  problems  that  increasingly  occupy 
the  modern  mind.  How  can  experience  alone  afford 
certainty,  how  can  the  knowledge  of  a  law,  in  the  exact 
meaning  of  the  word,  be  purely  experimental  in  its 
origin  ?  Aristotle  taught  that  the  general,  so  far  as  it 
is  known  by  experience  alone,  necessarily  includes  excep- 
tions, and  that  only  intellectual  knowledge  can  have 
universal  value.  And  this  has  been  the  classic  doctrine 
up  to  the  present.  Descartes,  however,  had  already 
declared  that  there  is  a  true  science  of  phenomena,  that 
what  is  transitory  in  its  nature  may  be  reduced  to 
immutable  essence  ;  and  science,  in  its  onward  march, 
has  been  increasingly  unconscious  of  Aristotle's  objec- 
tion. And  yet,  what  right  have  we  to  reject  a  doctrine 
which  seemed  to  be  evidence  itself  ?  How,  and  in  what 
sense,  can  a  fact  be  a  law  ?  Kant  accepted  this  question 
as  modern  science  states  it  ;  it  is  the  object  of  his 
doctrine  of  forms  and  categories  to  answer  it.  The 
solution  is  a  profound  one  ;  it  cannot  be  avoided  by  any 
who  persistently  determine,  without  fearing  contradic- 
tion, to  unite  experience  with  certainty. 

Kant's  system  of  morals,  too,  is  far  from  having 
become  foreign  to  us.  We  are  at  present,  as  regards 
action,  in  a  position  similar  to  that  in  which  science 
places  us  as  regards  being.  We  accept  only  facts,  and 
yet  we  cannot  renounce  certainty,  law,  belief  in  duty. 
We  are  determined  to  reject  every  motive  of  action 
adopted  from  the  idea  of  a  suprasensible  world,  and  yet 
we  claim  to  maintain  a  system  of  absolute  morals,  a 
doctrine  of  obligation.  Are  we  not,  then,  almost 
prepared  to  appreciate  a  philosophy  which  actually 


330  STUDIES  IN  PHILOSOPHY 

brings  duty  out  of  the  very  heart  of  experience,  and 
holds  aloof  from  mysticism  and  utilitarianism  alike  ? 

And  if,  in  social,  religious  and  political  questions, 
we  are  troubled  by  the  conflict  between  history  and 
reason,  between  what  is  and  what  ought  to  be,  between 
form  and  idea,  between  fact  and  right,  between  the 
national  ideal  and  the  human  one,  do  we  not  thereby 
find  ourselves  somewhat  in  the  same  position  as  Kant 
when  he  was  investigating  the  relations  between  theory 
and  practice  and  reconciling  the  necessity  of  nature 
with  the  sovereignty  of  reason  in  his  doctrine  of  moral 
progress  ? 

Not  in  vain,  then,  was  it  that  Kant  endeavoured, 
both  in  the  sphere  of  action  and  in  that  of  knowledge,  to 
adopt  that  point  of  view  of  the  universal,  at  once 
real  and  ideal,  which  is  also  the  point  of  view  of  reason  : 
his  doctrine  thereby  receives  a  lofty,  positive  character, 
such  as  could  not  be  met  with  either  in  the  pure 
generalisations  of  experience  or  the  dreams  of  imagina- 
tion. It  is  not  the  mirror  of  a  single  epoch,  nor  even 
the  expression  of  a  nation's  thought  :  it  belongs  to 
the  whole  of  mankind. 


INDEX 


Acroamatic,  instruction,  77,  80,  8 1 

Adam,  217-221 

Adam,  Charles,  241-244 

Aesculapius,  75 

Alexander,  75-78,  120,  151,  152 

Alexander  of  Aphrodisias,  157 

Alexandria,  157 

Amaury  of  Chartres,  1 60 

Amphictyonic  council,  313 

Amyntas  II.,  75 

Anaxagoras,  n,  84 

Andronicus,  79,  156 

Antipater,  152 

Antisthenes,  40 

Apellicon,  79 

Apodeictic,  98 

Apollo,  69 

Arabs,  158 

Aristides,  43 

Ariston,  152 

Aristophanes,  14,  19 

Aristotle,  6,  175,  203,  204,  205,  241, 
252,  280,  329;  biography,  74-79; 
writings,  79-83  ;  ensemble  of  his  work, 
84,  85  ;  classification  of  sciences, 
85-87 ;  method  and  point  of  view, 
87-89;  historian,  89-91;  logic,  92- 
103;  metaphysics,  103-112;  general 
physics,  112-116;  mathematics,  116, 
117;  cosmology,  117-120;  astro- 
nomy, 120-122  ;  meteorology,  122  ; 
mineralogy,  122,  123  ;  general  bio- 
logy, 123-125  ;  botany,  125  ;  animal 
anatomy  and  physiology,  125-130; 
zoBlogy,  130-132;  psychology,  132- 
135;  morals,  135-140;  economics, 
140,  141 ;  politics,  142-146  ;  rhetoric, 
146-148  ;  esthetics,  148,  149  ;  poetics, 
149,  150;  grammar,  151  ;  speeches 
and  poems,  151;  letters,  152;  lan- 
guage, 153-156  ;  influence,  156-162; 
Aristotle's  system  compared  with 
Kantian  idealism  and  with  evolution- 
ism, 162-164;  the  essential  features 


of  Aristotelianism  retained  in  modern 
science,  164-166  ;  the  three  principles 
of  Aristotelian  finality,  166-168 

Aristoxenus,  156 

Asclepiads,  75 

Aseity,  182 

Atarnea,  76,  152 

Athenians,  69 

Athens,  75-78 

Aufkl'drungsphilosophie,  256,  314 

Aulus  Gellius,  77 

Aurora,  229 

Averroes,  159 

Baader,  231 

Babylon,  120 

Bacon,  35,  70,  88,  102,  161,  244,  245 

Baillet,  241,  242,  243,  246,  247,  248,251 

Barbarians,  141 

Bardili,  321 

Basedow,  316 

Batanaea,  157 

Baumgarten,  273,  294 

Beccaria,  306 

Beck,  321 

Berkeley,  234 

Bible,  173,  226 

Boehme,  Philosophus  Teutonicus,  169-170. 
His  first  book,  Aurora,  171  ;  Pro- 
found influence  of  the  Bible,  173  ; 
Meditation  on  the  phenomena  of 
nature,  174;  his  life-aim,  the  enter- 
ing into  the  Kingdom  of  God,  175- 
1 76  ;  the  nature  of  evil,  1 77  ;  his 
task :  to  enquire  into  the  genesis  of 
things,  178  ;  threefold  division  of  his 
system,  178  ;  methods  of  ordinary 
philosophy  useless,  179,  180;  man 
must  adopt  the  standpoint  of  God, 
181  ;  problem  of  aseity,  182;  exami- 
nation of  Eckhart's  doctrine,  184 ; 
revelation  requires  opposition,  185  ; 
expose  of  Boehme's  teaching  on  the 
birth  of  God,  186-202;  comparison 


331 


332 


STUDIES  IN  PHILOSOPHY 


of  his  philosophy  with  that  of  the 
ancients,  202-206  ;  his  theology 
neither  monism  nor  dualism  ;  per 
crucem  ad  lucem :  the  human  and 
divine  law,  206-208  ;  opposition  to 
pantheism  and  to  theism  ;  solution 
of  problem  of  creation,  definition 
of  creation,  208-215  ;  origin  of  sin 
and  hell,  215-217  ;  man's  three  parts, 
soul,  mind,  and  body,  correspond  to 
the  principles  of  fire,  light,  and 
essence  or  reality,  217-218  ;  spiritual 
interpretation  of  the  Fall  of  Man, 
218-222  ;  reconciliation  of  good  and 
evil  ;  realisation  of  man's  salva- 
tion by  Christ,  election  by  grace  ; 
faith,  the  sign  of  election  and  the 
beginning  of  regeneration ;  contrast 
between  faith  and  works,  prayer  and 
sacraments,  222-226  ;  life  of  the  re- 
generate Christian,  226-228;  Aurora: 
the  spiritual  interpretation  of  Chris- 
tianity, 229-231  ;  spirit,  i.e.  infinite 
freedom,  alone  is  true  being  ;  evil  is 
not  a  lesser  good  but  rather  a  living 
force  that  tends  to  destroy  good,  231, 
232  ;  characteristics  of  German  philo- 
sophy, 232,  233 

Boethus,  157 

Boetius,  158 

bonk  metis,  254 

Br£al,  Michel,  quoted,  vii 

Buhle,  75 

Burke,  294 

Cagliostro,  268 

Ca  ili  pus,  121 

Callisthenes,  77,  120 

Cartesianism,  5,  97,  326 

Chalcis,  75-78 

Chaldaeans,  120 

Christ,   159,   171,   188,  217,  222,  225, 

308,  309 
Christian,    224,    225,    226,    235,     270, 

308,  309 
Christianity,    72,    160,    169,    175,    177, 

201,  205,  229,  264,  321 
Church,  305,  308,  309,  310 
Cicero,  91,  154 
Cilicia,  157 
Clerselier,  243 
Cogito,  234,  235,  237 
Cohen,  323 

Collegium  Fredericanum,  257,  258 
Copernicus,  284 
Cordova,  159 
Corneille,  5,  242 
Cousin,  Victor,  325 
Criticisme,  325 


Critique  of  Pure  Reason,  261,  262,  302, 

3X7>  3l8»  3J9»  325»  326 
Crito,  40 
Critolaiis,  156 

Damascus,  157 

Dauriac,  325 

David  of  Dinant,  160 

Davis,  Prof.  W.  M.,  quoted,  vi 

Decearchus,  156 

de  Coulanges,  Fustel,  I,  vi 

de  Cusa,  Nicolas,  172 

Degerando,  324 

d'Eichthal,  Gustave,  10 

del  Perojo,  Jose,  326 

Demetrius,  152 

Democritus,  84,  116,  119,  123,  152 

Demosthenes,  78 

Denys  of  Halicarnassus,  74 

Descartes,  5,  6,  161,  167,  174,  265, 
326,  327,  329  ;  The  Cogito,  the  living 
germ  from  which  have  sprung  all 
great  modern  systems  of  philosophy  ; 
problem  of  transition  from  thought 
to  existence,  234-236 ;  in  thought 
alone  lies  human  dignity  and  the 
principle  of  certainty,  236  ;  Cartesian- 
ism  was  the  soul  of  contemporary 
philosophy  and  science,  affirmed  Hux- 
ley, 237  ;  Descartes'  style,  a  perfect 
model  of  prose,  238-240;  relation 
between  morals  and  science,  241-244  ; 
everything  in  nature  is  brought  about 
mathematically  ;  life  itself  is  a  mech- 
anism and  is  subject  to  our  control, 
245,  246  ;  purpose  and  aim  of  the 
Discours  de  la  Methode,  the  Lettres,  the 
Principes,  and  the  Traite  des  Passions, 
248-252  ;  no  limit  to  the  conquests 
of  science  over  nature  ;  the  sovereignty 
of  reason  ;  Cartesian  morals  closely 
united  with,  though  not  derived  from, 
the  science  of  natural  things ;  the 
full  realisation  of  reason  is  the  goal 
of  life  ;  the  bona  mens  is  the  supreme 
aim,  253,  254 

de  Stagl,  Madame,  324 

Dialectic,  100 

Diogenes  Laertius,  74,  75,  79,  152 

Discours  de  la  Methode,  236,  239,  241, 
242,  247,  249,  250,  254 

Dynamics,  301 

Eckhart,  160,  170,  172,  184 

Egyptians,  120,  160 

Eleatics,  113,  322 

eX^7%o^res,  15 

Elizabeth,  Princess,  241    251,  253,  254 

Empedocles,  113 


INDEX 


333 


Encyclopedism,  259 

Epicureanism,  157 

Epicurus,  241 

Eristic,  101 

Eucharist,  159 

Euclid,  40 

Eudemus,  152,  156 

Eudoxus,  121 

Exoteric  instruction,  77,  80,  8 1 

Fall,  230 

Faye,  271 

Feder,  320 

Feuerbach,  323 

Fichte,  72,  169,  232,  255,  262,  321 

Finsterniss,  191 

Forster,  311 

Fouillee,  9,  II,  14,  50,  54,  57,  58 

Franck,  10,  50,  172,  173 

Frederick  II.,  256,  261,  265,  306 

French     Revolution,    236,     256,     263, 

3'9 

Fries,  320 

Garve,  320 

Gazette  <f  Anvert,,  243 

Gennadius,  161 

George  of  Trebizond,  161 

God,  no,  in,  112,  116,  118,  159,  171, 
173,  176-193,  195,  198-231,  246, 
249,  251,  254,  259,  272,  286,  288, 
292,  293,  297,  300,  307,  308,  310, 
318,  322 

GSrlitz,  170 

Goethe,  262  ;  quoted,  ix 

Gorgias,  22 

Greeks,  141 

Gros,  323 

Grote,  i,  9,  n 

Hamann,  169,  262,  320 

Hamilton,  325 

Harvard  University,  vii 

Hegel,  i,  31,  72,   162,   169,   170,  232, 

233.  3ZI 
Helmholtz,  271 
Herbart,  322 
Herder,  268,  315,  320 
Hermias,  76,  151,  152 
Herpyllis,  76,  78 
Herz,  Marcus,  267,  270 
Hhtoria  animalium,  77,  130 
Hobbes,  313 
Homer,  35 
Hufeland,  323 
Hume,  235,  260,  262,  275,  276,  277, 

278,  288 
Hutcheson,  260 
Huxley,  236 


Huyghens,  235 
Hypomnematic  writings,  81 

lena,  320 
Isocrates,  76 

Jacob!,  169,  320 
Janet,  10,  325 
Jesus,  222,  224 
Jews,  158,  159 
John  the  Baptist,  159 
Jupiter,  112 

Kabbah,  193 

Kant,  162,  163,  169,  170,  231,  232,  234, 
235  ;  the  influence  of  Kantianism, 
255;  biography,  256-264;  his  poli- 
tics, religion,  morals,  264 ;  chrono- 
logical list  of  works,  265  -  270  ; 
theory  of  actual  history  of  the  forma- 
tion of  the  world,  271  ;  relations 
between  possibility  and  existence, 
philosophy  and  mathematics,  272, 

273  ;    how  far   can   one   become   ac- 
quainted with  supra-sensible  existences  ? 

274  ;  nature  of  space  and  time,  275  ; 
principle  of  causality,  276-278  ;   ex- 
planation   of   the    conditions    of  ex- 
perience, 278-288  ;  what  is  morality 
and   on   what   does    it    rest  ?    Kant's 
categorical  imperative,  289-291  ;  his 
postulates,  292.,  293  ;    the   notions   of 
taste    and    finality,    294-300 ;    meta- 
physics of  the  science  of  nature,  301, 
302  ;  the  metaphysics  of  morals,  the 
various  kinds  of  rights  and  possessions, 
303-306  ;  duties,  307  ;  maxims,  308  ; 
the  four  essential  ideas  of  the  Chris- 
tian   religion,   308-310;    applications 
of  the  metaphysical   doctrine  to  the 
natural  and  the  moral  history  of  man  ; 
Aufktirung,    the     present    phase     of 
human    history,    310-314;    freedom 
to  express  thought,  the  condition  of 
the  progress   of   enlightenment ;    re- 
conciliation   of     freedom    with     the 
rights   of  the   State;   taxation,  315; 
reform  needed  in  methods  of  educating 
will,  intellect,  and  body ;  the  goal  of 
education  is  the  formation  of  virtue, 
316,  317;  university  instruction,  the 
conflict    between     philosophers     and 
(i)  theologians,  (2)   lawyers,  (3)   doc- 
tors, 318,  319;  influence  of  Kantian 
philosophy,  systems  born  of  it ;  various 
developments    in    esthetics,    theology, 
jurisprudence,    science,    mathematics, 
sense  psycho-physiology,  and  German 
politics,    320-324;    appreciation    in 


334 


STUDIES  IN  PHILOSOPHY 


France,  England,  Italy,  and  Spain, 
325,  326  ;  the  historical  role  of 
Kant's  philosophy  ;  experience,  based 
on  reason,  the  starting  point  of  all 
knowledge  ;  reason  creates  as  well  as 
destroys,  it  is  both  freedom  and 
law,  326-328  ;  world-wide  scope  and 
import  of  Kant's  doctrines,  329, 

33° 

KdBapffis,  149 
Knorr,  264 
Knutzen,  258,  259 

KUl/JLTj,    142 

K8nigsberg,  257,  258,  261,  264 

Krause,  323 

Krug,  320 

Kuno,  Fischer,  234,  255,  267 

Kuntzschmann,  Catherina,  170 

Lachelier,  325 
La  Mettrie,  234 
Lamian  war,  78 
Lampsacus,  156 
Lange,  323 
Laplace,  259,  265 

Leibnitz,  6,  99,  162,  169,231,232,234, 
235,  258,  260,   261,  265,   266,   275, 

3i5»  3" 
Leibnitzians,  324 
Leibnitzo-Wolfian,  271,  288,  320 
Lesbos,  156 
Lettres,  249 
L6v6que,  9 
Liebmann,  323 
Lilienthal,  258 
Lisbon,  266 

Locke,  92,  234,  235,  260,  320,  325 
Lucifer,  215,  216,  217,  218,  221,  222 
Luther,  161,  172,  176,  177,  178 
Lycaeum,  77 

Machaon,  75 
Macedonia,  75,  76 
Maieutics,  38] 
Maimon,  Salomon,  321 
Malebranche,  234,  235 
Martin,  Saint,  231 
Mechanics,  301 
Meiners,  320 
Mendelssohn,  266,  320 
Mentor,  152 
Mersenne,  247,  251 
Messena,  156 
Meyer,  Bonna,  323 
Milky  Way,  122 
Moliere,  5 
Moses,  216 

Moses  Maimonides,  159 
Miiller,  Johannes,  324 


Miiller,  Max,  326 
Mytilene,  76 

Neleus,  79 

Neo-Kantianism,  322,  325 

Neo-Platonism,  205 

Neo-Platonist,  157,  206 

Newton,  235,  247,  258,  260,  265,  271, 

^73>  *78>  303 

Newtonianism,  258,  272,  274,  286,  301 
Nicolai,  320 
Nicolas,  157 

Nicomachean  Ethics,  71,  79,  155 
Nicomachus,  75,  79 
Nisard,  D£sir6,  238 
vovs,  96,  97,  132,   133,    134,   138,  139, 

157,  160,  175 

Olympic  games,  91 
Orestes,  41 
Organon,  158 

Paphlagonia,  157 

Paracelsus,  172,  173 

Pascal,  6,  8,  17,  18,  175,  236 

Pauline  origin,  172 

Paulsen,  323 

Peripatetic  school,  77,  79,  82,  156,  157, 

160 

Phenomenology,  302 
Philip,  76,  152 
Philopon,  157 
Philoxenes,  152 
Phoronomics,  301 
$&ris,  112,  157,  167 
Physiognomonica,  129 
Pietism,  256,  259,  270 
Pietist,  257,  258 
Piety,  Socrates'  definition  of,  62 
Pillon,  325 
Pirithous,  41 
Plainer,  320 
Plato,  11-16,  27,   30-34,  40,  70,  71,  75, 

78,  84,  89,  103,  104,  116,  123,  140, 

143,  H4,  146,   148,  IS1.  152.  !57» 

158,  160,  175,  321,  322 
Plenum,  115,  273 
Plotinus,  157 
Plutarch,  79 

7r6Xts,  142 
Polycrates,  13 
Pomponatius,  161 
Porphyry,  120,  157,  158 
Postulates,  Kant's,  292 
Principes,  242,  243,  248 
Protagoras,  22,  68,  89 
Protestantism,  177,  256 
Pseudo-Ammonius,  75 
Pseudo-Hesychius,  75 


INDEX 


335 


Pylades,  41 
Pythagoreans,  206 
Pythian  games,  91 
Pythias,  76,  78 

Quintessence,  118 
Quito,  266 

Rauch,  264 
Redemption,  230 
Regulae,  241,  242,  250,  254 
Rehberg,  323 
Reid,  235,  325 
Reinhold,  320 
Renaissance,  161,  238 
Renan,  I 
Renouvier,  325 
Reuter,  Anna  Regina,  257 
Rhetoric,  101 
Rhodes,  79,  156 
Richter,  262 
Riehl,  255,  323 
Riga,  261 
Ristchl,  323 

Roman  Catholic  Church,  225 
Rome,  79,  156 
Rosenkranz,  267 

Rousseau,  259,  260,  273,  300,  312,  313, 
316 

Satan,  221 

Scaliger,  161 

Schelling, 72, 169,231,  232,  255,  286,321 

Schiller,  262,  323 

Schleiermacher,  12,  13,  20,  30,  321 

Schmalz,  323 

Schoolmen,  70,  71,  171 

Schopenhauer,  267,  322 

Schultze,  323 

Schulz,  257 

Schulze,  Ernst,  320 

Schwenckfeld,  172,  173 

Scolion,  152 

Scotland,  257 

Scriptures,  318 

Selle,  320 

Selymbrians,  152 

Sensor mm  commune^  132 

Shaftesbury,  260 

Sidon,  157 

Simmias,  40 

Simplicius,  120,  152,  157 

Skepsis,  79 

Socrates,  245,  255,  316,  326;  modern 
writers'  accounts  of,  9-12  ;  testimony 
of  Xenophon,  Plato  and  Aristotle,  13- 
14  ;  judgment  of  physics  and  sophistic, 
16-26;  "Know  thyself,"  26-29; 
science  and  morals,  30-34  ;  method  : 


maieutics  and  irony ;  definition  and 
induction,  self-  possession  and  love  ; 
refutation,  35-46  ;  teleology  and  moral 
theology ;  virtue  and  the  science  of 
good  ;  <ro(pta  the  virtue  par  excellence, 
49-65  ;  "Virtue  is  one  and  can  be 
taught,"  fj.d6r]ffit  and  fj.e\frri,  66-67  5 
"  moral  science  ":  the  central  fact  of 
his  teaching  ;  science  does  not  attain  to 
things  divine  ;  belief  in  an  Apollonian 
mission ;  philosopher  and  man  of 
action  alike,  68  -  70  5  "  The  only 
science  is  that  of  the  general  " ;  the 
Nicomachean  Ethics  ;  influence  on 
modern  philosophy,  71-73 

Socratics,  70 

ffcxpla.,  ix 

Sophists,  21,  22,  23,  25,  29,  51-55,  61, 
66,  245 

Sorbonne,  160 

Spencer,  325 

Spinoza,  231,  234,  235,  321 

Stadler,  323 

Stageira,  75 

State,  305,  306,  307,  313,  315,  319 

Stoics,  157,  252,  253 

Strabo,  79 

Strasbourg,  324 

Strato,  156 

Sucht,  1 86 

Suidas,  75 

Sweden,  241,  244 

Swedenborg,  267,  274 

Sylla,  79 

Syracuse,  75 

Tarentum,  156 

Tauler,  170,  173 

Thales,  89,  122 

Themistius,  157 

Theodoric  of  Freiburg,  160 

Theodorus,  146 

Theodota,  65 

Theophrastus,  79,  156 

Theosophist,  169,  184,  231 

Theosophy,  172,  173,  182 

Theseus,  41 

Thomas,  Saint,  241 

Thomists,  170 

Thrasymachus,  146 

Tiedemann,  320 

rl  tffrl,  10 

Tisias,  146 

Traite  des  Passions,  248,  251,  252 

Trinity,  189,  230 

Ulysses,  35 
Ungrund,  183 
University  of  France,  160 


336 


STUDIES  IN  PHILOSOPHY 


Vacuum,  115 

Vanini,  161 

Vedas,  326 

Victor  Hugo,  quoted,  vi 

Virgin,  the  Eternal,  174,  175,  189,  219, 

222 

Voltaire,  3 

Von  Baader,  Franz,  169 

Von  Nettesheim,  Agrippa,  172 

Walther,  Dr.,  170 
Weigel,  Valentin,  172,  173 
Weishaupt,  320 


Wille,  1 86 

Windelband,  255,  323 
Wolf,  258,  272,  300 
Wundt,  255 

Xanthippe,  65 
Xenocrates,  76 
Xenophon,  12,  13,  33,  40 

Zachariae,  323 

Zeller,  9,  II,  20,  22,  30,  36,  54,  80,  83, 

322 
Zeno,  241 


THE    END 


Printed  by  R.  &  R.  CLARK,  LIMITED,  Edinburgh 


By   Dr.    F.   C.   S.   SCHILLER 

HUMANISM.     Philosophical  Essays.      8vo.     8s.  6d.  net. 
STUDIES  IN   HUMANISM.      8vo.      los.  net. 

RIDDLES  OF  THE  SPHINX.  A  Study  in  the 
Philosophy  of  Humanism.  New  and  Revised  Edition. 
8vo.  i  os.  net. 

FORMAL  LOGIC.  A  Scientific  and  Social  Problem. 
8vo.  i  os.  net. 


IDOLA  THEATRI  :  A  Criticism  of  Oxford  Thought 
and  Thinkers  from  the  Standpoint  of  Personal  Idealism. 
By  HENRY  STURT.  8vo.  xos.  net. 

PERSONAL  IDEALISM.  Philosophical  Essays.  Edited 
by  HENRY  STURT.  8vo.  los.  net. 

THE  PRINCIPLE  OF  INDIVIDUALITY  AND 
VALUE.  Being  the  Gifford  Lectures  for  1911.  By 
BERNARD  BOSANQUET,  LL.D.,  D.C.L.  8vo.  los.  net. 

THE  MEDIEVAL  MIND.  A  History  of  the  Develop- 
ment of  Thought  and  Emotion  in  the  Middle  Ages.  By 
HENRY  OSBORN  TAYLOR.  Two  vols.  8vo.  2is.net. 

THE  ORIGIN  AND  DEVELOPMENT  OF  THE 
MORAL  IDEAS.  By  EDWARD  WESTERMARCK,  Ph.D. 
Two  vols.  8vo.  145.  net  each. 

A  SHORT  HISTORY  OF  ETHICS— GREEK  AND 
MODERN.  By  REGINALD  A.  P.  ROGERS,  Fellow  and 
Tutor  of  Trinity  College,  Dublin.  Crown  8vo.  35.  6d.  net. 

MACMILLAN  AND  CO.,  LTD.,  LONDON. 


THE  IDEALISTIC  CONSTRUCTION  OF  EXPERI- 
ENCE. By  Prof.  J.  B.  BAILLIE,  M.A.,  D.Phil.  8vo. 
8s.  6d.  net. 

THE  ORIGIN  AND  SIGNIFICANCE  OF  HEGEL'S 
LOGIC.  By  Prof.  J.  B.  BAILLIE,  M.A.,  D.Phil.  8vo. 
8s.  6d.  net. 

THE  PHILOSOPHY  OF  GASSENDI.  By  G.  S. 
BRETT.  8vo.  xos.  net. 

HISTORY  OF  PHILOSOPHY.  With  Especial  Refer- 
ence to  the  Formation  and  Development  of  its  Problems 
and  Conceptions.  By  Dr.  W.  WINDELBAND.  Translated 
by  J.  H.  TUFTS.  8vo.  1 75.  net. 

HISTORY  OF  THE  PROBLEMS  OF  PHILO- 
SOPHY. By  PAUL  JANET  and  GABRIEL  SEAILLES.  Trans- 
lated by  ADA  MONAHAN  and  Edited  by  HENRY  JONES,  LL.D. 
Two  vols.  8vo.  i  os.  net.  each. 


THE   SCHOOLS    OF    PHILOSOPHY 

A  HISTORY  OF  THE  EVOLUTION  OF 

PHILOSOPHICAL  THOUGHT 

BY  VARIOUS  WRITERS 

EDITED    BY 

SIR  HENRY  JONES 

PROFESSOR   OF   MORAL    PHILOSOPHY   IN   THE   UNIVERSITY  OF   GLASGOW 

Ready. 

THE  EVOLUTION  OF  EDUCATIONAL  THEORY. 
By  Professor  JOHN  ADAMS,  M.A.,  B.Sc.,  LL.D.  8vo. 
i  os.  net. 

In  Preparation. 

THE  HISTORY  OF  GREEK  PHILOSOPHY  FROM 
THALES  TO  ARISTOTLE.  By  Professor  JOHN 
BURNET,  LL.D. 

THE     HISTORY    OF     MODERN     PHILOSOPHY 
FROM  HOBBES  TO  REID.     By  Professor  G.  F.  STOUT. 
Other  -volumes  to  follow. 

MACMILLAN  AND  CO.,  LTD.,  LONDON. 


y