Psychological review

i!

in MM

HIMW

7 0J

THE

Psychological Review

EDITED BY

J. MARK BALDWIN HOWARD C. WARREN

JOHNS HOPKINS UNIVERSITY PRINCETON UNIVERSITY

CHARLES H. JUDD, Yale University (Editor of the Monograph Series).
WITH THE CO-OPES A TION FOR THIS SECTION OP

A. C. ARMSTRONG, WESLEYAN UNIVERSITY ; ALFRED BINET, ECOLE DES HAUTIS-
ETUDES, PARIS ; W. L. BRYAN, INDIANA UNIVERSITY ; WILLIAM CALDWELL, Mc-
GILL UNIVERSITY; MARY W. CALKINS, WELLESLEY COLLEGE; JOHN DEWEY,
COLUMBIA UNIVERSITY ; J. R. ANGELL, UNIVERSITY or CHICAGO ; C. LADD FRANKLIN,
BALTIMORE; H. N. GARDINER, SMITH COLLEGE; G. H. HOWISON, UNIVERSITY or
CALIFORNIA ; P. JANET, COLLEGE DE FRANCE ; JOSEPH JASTROW, UNIVERSITY or WIS-
CONSIN; ADOLF MEYER, N. Y. PATHOL. INSTITUTE; C. LLOYD MORGAN, UNIVERSITY
COLLEGE, BRISTOL; HUGO MUNSTERBERG, HARVARD UNIVERSITY; E. A. PACE,
CATHOLIC UNIVERSITY, WASHINGTON ; G. T. W. PATRICK, UNIVERSITY OF IOWA ; CARL
STUMPF, UNIVERSITY, BERLIN ; R. W. WENLEY, UNIVERSITY OF MICHIGAN.

Volume XL, 1904.

THE MACMILLAN COMPANY,

41 NORTH QUEEN ST., LANCASTER, PA.
66 FIFTH AVENUE, NEW YORK

AGEWT.- G. E. STKCHERT, LONDON (a Star Yard, Carey St., W. C.);
LEIPZIG (Hoipiul St., 10); PARIS (76 rue de Rcnne*).

P7
MI

PRESS or

THE NEW ERA PRINTING COMPANY
LANCASTER, PA.

CONTENTS OF VOLUME XI.

January.

The Participation of the Eye Movements in the Visual Perception of Motion : RAYMOND

DODGE, i.

An Inquiry into the Nature of Hallucination, I. : BORIS Sims, 15.
The Limits of Pragmatism : J. MARK BALDWIN, 30.
Discussion : The Sexual Element in Sensibility : W. I. THOMAS, 61 ; Or. Morton Prince

and Panpsychism, C. A. STRONG, 67.

March.

Theory and Practice : President's Address, WILLIAM LOWE BRYAN, 71.

On the Attributes of the Sensations : MAX MEYER, 83.

An Inquiry into the Nature of Hallucination, II. : BORIS SIDIS, 104.

Discussion : The Mechanism of Imitation : F. C. FRENCH, 138.

May.

The Law of Attraction in Relation to some Visual and Tactual Illusions : HAYWOOD J.

PEARCE, 143.
The Relation between the Vaso-Motor Waves and Reaction-Times : WILLIAM R.

WRIGHT, 179.

On the Horopter : GEORGE T. STEVENS, 186.
Shorter Contributions : The Logical and Psychological Distinction between the True and

the Real: C. L- HERRICK, 204. The Period of Conversion: G. A. TAWNBY, 210.

The Genetic Progression of Psychic Objects : J. MARK BALDWIN, 216.
Notes : On the Attributes of Sensation : M. W. CALKINS, 221.
Editors' Note : 222.

July-September.

An Experimental Study of the Physiological Accompaniments of Feeling : L. PEARL
BOGGS, 223.

The Psychology of ^Esthetic Reaction to Rectangular Forms : THOMAS H. HAINKS and
ARTHUR ERNEST DA VIES, 249.

Conceptions and Misconceptions of Consciousness : RALPH BARTON PERRY, 282.

Retinal Local Signs : WALTER F. DEARBORN, 297.

Studies from the California Psychological Laboratory. VI. Some Peculiarities of Fluc-
tuating and of Inaudible Sounds : KNIGHT DUNLAP, 308.

Some Observations on Visual Imagery : H. B. ALEXANDER, 319.

Incipient Pseudopia : CHARLES CAVERNO, 338.

November.

The Classification of Psycho-Physic Methods : EDWIN B. HOLT, 343.

Studies on the Influence of Abnormal Position Upon the Motor Impulse: CHARLES

THEODORE BARNETT, 370.
Discussion : Mind and Body — The Dynamic View: C. L> HBRIUCK, 395.

The Psychological Review,

MONOGRAPH SERIES.

The following Monographs have already appeared :

Vol. X.

1. * On Sensations from Pressure and Impact: HAROLD GRIPPING. Pp.88.

2. Association : MARY WHITON CALKINS. Pp. vii+56.

3. * Mental Development of a Child : KATHLEEN MOORE. Pp. iv+i5o.

4. A Study of Kant's Psychology: EDWARD FRANKLIN BUCHNER. Pp. viii+2o8.

TTol. XX.

5. Problems in the Psychology of Reading : J. O. QUANTZ; Pp. iv+5i.

6. The Fluctuation of Attention : JOHN PERHAM H-SLAN. Pp. 11+78.

7. *Mental Imagery : WILFRID LAY. Pp. 11+59.

8. Animal Intelligence : EDWARD L. TrfORNDiKit. Pp. ii+io9.

9. The Emotion of Joy : GECRGE VAN NESS DEARBORN. Pp. ii+yo.
10. Conduct and the Weather : EDWIN G. DEXTER. Pp. viii+io5.

TTol. XXX.

1 1. On Inhibition : B. B. BREESE. Pp. iv+65.

12. On After-images : SHEPHERD IVORY FRANZ. Pp. iv+6i.

1 3. * The Accuracy of Voluntary Movement : R. S. WOODWORTH. Pp. vi+ 1 14.

14. A Study of Lapses : H. HEATH BAWDEN. Pp. iv+i22.

15. The Mental Life of the Monkeys : F,. L. THORNDIKE, Pp. iv+57.

16. The Correlation of Mental and Physical Tests : C. WISSLER. Pp. iv+62.

17. Harvard Psychological Studies, Vol. I. ; containing sixteen experimental investi-

gations from the Harvard Psychological Laboratory : Edited by HUGO MUNSTERBERG.
Pp. viii -}- 654. $4.00.

Vol. -V.

18. Sociality and Sympathy : J. W. L. JONES. Pp. iv+gi. 75 cents.
19- The Practice Curve : J. H. BAIR. Pp. 70. 75 cents.

20; The Psychology of Expectation : CLARA M. HITCHCOCK. Pp. iv -f- 78. 75 cents.

21. Motor, Visual and Applied Rhythms: J. B. MINER. Pp. iv+ 106. #1.00.

22. The Perception of Number : J- F. MESSENGER. Pp. iv -f- 44. 50 cents.

23. A Study of Memory for Connected Trains of Thought : E. N. HENDERSON. Pp.

iv -}- 94. 75 cents.

(To contain about 500 pages.)

24. A Study in Reaction Time and Movement : T. V. MOORE. Pp. iv+86. 75 cents.

25. The Individual and his Relation to Society : J. H. TUFTS. Pp. iv+58. 50 cents.

26. Time and Reality: J. E. BOODIN. Pp. ¥+119. $1.00.

* Monographs so marked are not sold separately. Vols. I-III are #7.50 each.

THE PSYCHOLOGICAL REVIEW, XI., 1904.
10 20 30 40 50

H.
Hor.

PLATE VI.
60 70 80 90 100

H.

Vert.

Wa.
Hor.

Wa.
Vert.

CHART 2.
NOTE. The uppermost row of figures should be on the base line, i. Hor.

N. S. VOL. XI. No. i January, 1904

THE PSYCHOLOGICAL REVIEW.

THE PARTICIPATION OF THE EYE MOVEMENTS
IN THE VISUAL PERCEPTION OF MOTION.1

BY PROFESSOR RAYMOND DODGE,
Wesleyan University.

Psychological tradition is practically unanimous in distin-
guishing two fundamentally different conditions under which
the perception of motion may occur. The one presupposes a
relatively motionless eye, and offers sensory data composed ex-
clusively of the displacement and modification of the retinal
image. The other condition is characterized by certain forms
of eye movement which are supposed to furnish perceptual data
quite independent of all modification of the retinal image.

Concerning the exact form of the sensory data which oper-
ate under the former circumstances, there is less unanimity.
For the purposes of our discussion it is sufficient to recall four
main varieties. The most important of them all is generally
held to be the simple displacement of the retinal image, when
the consequent successive excitation of different local signs is
supposed to condition an immediate perception both of the direc-
tion and of the extent of motion. This main angular displace-
ment of the retinal image must be distinguished from a second-
ary angular displacement which was especially emphasized by
Hoppe. The latter consists of irregular movements of the ret-
inal image following the accidental variations which occur in
the direction or velocity of most perceivable forms of objective
motion. This might seem to be merely a subgroup of the first

1 Read in part before the New York Branch of the American Psychological
Association in session with the Philosophical Club of Yale University.

2 R. DODGE.

variety, but it depends for its peculiar force, not directly on the
successive stimulation of different local signs, but rather on cer-
tain temporal peculiarities of that succession, which experience
has created one of the surest criteria of objective motion. A
third variety was especially emphasized by Exner who para-
doxically called it the 'visual sensation of motion,' and who
regarded it as entirely independent of all spatial synthesis. It
consists of a peculiar sensory datum, produced by every change
of retinal stimulation and immediately apprehended as move-
ment, even when it cannot be referred to any specific object or
direction. Notwithstanding the psychological anomaly involved
in this group, I believe it is destined to play a role of consider-
able importance in the theory of the visual perception of motion.
The fourth variety also involves a consciousness of change, but
it is characterized by a more or less definite comparison of the
spatial relations within a given field of view with the immedi-
ately preceding spatial relations in the same field.

Obviously, any movement of the eye in pursuit of a moving
object must more or less distort all four varieties of purely sen-
sory data. The main angular displacement of the retinal image
will be practically annihilated, while Exner's ' sensation of
motion,' and even the comparison of successive conditions
within the total field, would seem to be ambiguous unless sup-
plemented and corrected by some factor concurrent with the
eye movements. This factor in the visual perception of motion
has been variously characterized as a feeling of innervation,
sensations from the orbital muscles, either of activity or strain,
and finally sensations of motion arising from contact between
the sclerotic and various parts of the orbit, notably, in the most
recent theory, between the sclerotic and the eyelid.

Extreme emphasis was given to this motor factor in Strieker's
monument to the vagaries of pure introspection, while almost all
recent discussions of the visual perception of motion hold it in
one form or another to be a datum of fundamental importance.
That its importance has been overestimated has been indicated
by Fleischl, Aubert, James and Wundt, and most emphatically
announced by Stern.

Recent experimental study of the eye movements has dis-

VISUAL PERCEPTION OF MOTION. 3

covered a new and serious ground of suspicion against the tra-
ditional importance of the motor data by exposing the poverty
and inaccuracy of our direct apprehension of the eye move-
iiu-nts. The continuous involuntary eye movements, when the
eye seems to the subject to maintain a constant fixation ; the fixa-
tion pauses which interrupt every natural sweep of the eye across
a complex field of view ; the discreet corrective movements at
the end of every considerable eye sweep ; in fact, most of the
known characteristics of the eye movements yield no introspec-
tive data at all, or only such as are ambiguous or absurdly in-
adequate. It scarcely seems probable that eye movements
which we cannot even count, of whose amplitude we have no
immediate subjective clue, of whose very existence we are often
not aware even under the most rigid self-observation, could be
very important factors in the perception of minute spatial
changes. Not only, however, is there no independent con-
sciousness of the eye movements, adequate to the refinement of
the visual perception of motion, but the character of the eye
movements which occur when we view a moving object fur-
nishes evidence that, if our consciousness of them were complete
and exact, it would be either useless or misleading as a datum
in the visual perception of motion.

CHARACTERISTICS OF THE PURSUIT MOVEMENTS OF

THE EYE.

Photographic registration has disclosed two distinct types of
eye movements, which are directly involved in the pursuit of a
moving object. The one is the primary, reactive displacement
of the line of regard towards an eccentric point of interest.
The other is the true pursuit movement. The most important
characteristics of movements of the first type are the relative
constancy of their duration, under similar conditions of fatigue,
of original orientation, and of the direction and angle of eye
movement ; and secondly, the fact that, under ordinary circum-
stances of illumination and complexity of the field of view, they:
are never moments of new effective retinal stimulation. The
true pursuit movements of the eye, by which the constant fixa-
tion of a moving point of regard is maintained, differ in every

4 R. DODGE.

respect from eye movements of the first type. Their chief char-
acteristics may be summed up as follows : (i) The velocity of
the pursuit movements has no fixed value, but varies with the
apparent velocity of the object of regard as it moves across the
field of view. (2) Unlike movements of the first type, the pur-
suit movements are moments of clear vision. Indeed, they are
the essential condition for the clear perception of a moving
object of regard. (3) While movements of the first type are
fundamentally reactions to specific eccentric stimuli, pursuit
movements sometimes assume the character of habitual move-
ments, and may persist after the occasion for them has ceased.
(4) Finally, whereas movements of the first type are always
separated by relatively long intervals of rest, movements of the
second type are separated chiefly if not entirely by movements
of the first type.

The true pursuit movement, however, neither begins nor con-
tinues through any considerable angle of displacement, uncom-
plicated by movements of the first type. In the first place the
line of regard naturally wanders over a moving object just as it
does over a motionless object, but all displacements of the line
of regard in response to a change in the point of interest are
rapid movements of the first type. Moreover, the first phase of
pursuit is never a pure pursuit movement. Photographic records
show a well-marked group of irregular movements of the first
type, separated by moments of complete rest, before there is any
trace of the true pursuit movement. Finally, even after the
pursuit movement proper has begun, it is always interrupted,
more or less frequently, by eye movements of the first type.
These interruptions occur when there is no conscious change of
the point of regard within the moving object, and even in spite
of the best endeavors of the subject to maintain the constant
fixation of a single point. Every photograph that we have ob-
tained of the pursuit movements shows these minute but char-
acteristic irregularities, though they are less frequent in suc-
cessive pursuit sweeps of the same rhythm than they are in the
first sweep of a series ; and even in any given sweep of con-
siderable amplitude there is a marked tendency for the inter-
ruptions to decrease towards the end of the sweep. It appears

VISUAL PERCEPTION OF MOTION. 5

that the true pursuit movement consistently lags behind the
object of regard, and that the fixation is automatically renewed
from time to time by the slight corrective movements of the first
type.

A comparison of these two types of eye movements discloses
a functional difference between the fovea and the periphery of
the retina, with respect to the motor response to moving stimuli,
that has hitherto been entirely neglected. The simple reactive
displacements of the line of regard are apparently identical in
general character, whether the eccentric object of interest is at
rest or in motion. Under both conditions they function to bring
the point of regard to a new object of interest. The only
noticeable difference between the two cases is found not in the
resulting eye movements, but in the stimuli which bring about
the reaction. There is a certain liveliness in the moving stimu-
lus which serves both to detach it from its background, even
when the color differences would otherwise be imperceptible,
and to compel our attention in a notable manner. The basis of
this peculiar effectiveness of the moving stimulus would be an
interesting problem, but it lies outside the field of our present
discussion. The fact of especial significance seems to me to
be that notwithstanding the real and apparent difference in the
character of the stimuli, the immediate motor response is the
same in both cases. One does not need a clearer indication of
the relative unimportance of the motor reaction in the visual
perception of motion. The initiation of the true pursuit move-
ments, on the other hand, seems to depend in some way on
foveal stimulation ; since try as one may, they never begin until
the initial phase of the pursuit has successively brought the
retinal image of the moving object to the fovea.

This functional difference between the fovea and the pe-
riphery at once assumes considerable theoretical importance
when we remember that not only is the relative efficiency of
the periphery much higher in the case of moving than in the
case of stationary stimuli, but that, in the overwhelming ma-
jority of cases, the initiative in the perception of motion must be
taken by the periphery, even when the relatively constant fixa-
tion of the moving object is afterwards subserved by the true

6 R. DODGE.

pursuit movements. Furthermore, the primary reaction of the
eye to peripheral stimulation, if it ever furnished a factor in the
general perception of motion, would be an utterly misleading
measure of the objective motion, since it varies, not according
to the apparent angle velocity of the moving object, but accord-
ing to the angular distance of the peripheral stimulation from
the fovea. Thus the movement of the peripheral stimulus
through an arc of 2° at a distance of 40° from the fovea would
be accompanied by a motor impulse corresponding to twice the
angle that would accompany a similar movement at 20° from
the fovea. It may be objected that the motor factor, under
these circumstances, might not correspond to the actual eye
movement, but to the difference between the impulses needed
to fixate the peripheral stimulus at the beginning and at the end
of a given arc of movement. This hypothesis is untenable.
Exact registration shows that the average error in the eye move-
ments by which we seek to fixate a peripheral stimulus at 40°
from the fovea is from 20 to 30 times as great as the total arc
through which a point of light must move in order to be appre-
hended as moving in a given direction. Obviously the gradua-
tion of the motor impulses in response to peripheral stimulation
is altogether too inaccurate to account for the delicate sensitive-
ness of the periphery to moving stimuli.

We must, I think, conclude that, however much the tradi-
tional motor factor may enter into the foveal perception of mo-
tion, it cannot enter into the peripheral perception of motion as
an immediate datum.

Pursuit movements of the eye seem to be regarded by those
who maintain the existence of a kingesthetic factor in the visual
perception of motion, as though they were analogous to the pas-
sive movements of the hand, as it rests on a moving object.
Naturally, such an analogy is altogether false. There are no
passive movements of the eyeball, except the purely mechanical
displacements resulting from pressure. All phases of the pursuit
movement involve definite motor reactions to retinal stimulation.

Even if it were possible for the pursuit movement to begin
in response to some central cue, as Holt seems to maintain, it
is obvious that, until corrected by subsequent visual data, the

VISUAL PERCEPTION OF MOTION. 7

eye movements could be no better clue either to the direction or
to the velocity of the actual movement than the centrally condi-
tioned expectation which occasioned them. While, unless the
antecedent expectation were altogether correct, a kinaesthetic
factor must be a source of error and confusion.

If, on the other hand, pursuit movements occur only in re-
sponse to definite peripheral stimuli, as I believe, there are cer-
tain general characteristics of all reactions that render the value
of kinaesthetic data in the visual perception of motion extremely
problematic. Every reactive pursuit movement of the eyes
must be conditioned both in direction and in velocity by certain
definite characteristics of the sensory stimuli which occasion it.
Not only can its accuracy never transcend the accuracy and
completeness of the data on which the reaction occurs, but the
two would be equal only in a perfect organism. It follows that
no kinaesthetic factor from a reactive pursuit movement of the
eyes could ever correct or materially augment the data furnished
by the stimulus to reaction. That the antecedent data should
be ignored in favor of a less accurate and delayed kinaesthetic
factor seems to me a highly improbable hypothesis.

Not only, however, would the kinaesthetic data from a reac-
tive pursuit be useless if it existed, but it would be a positive
source of error and confusion, since, as a reaction, the pursuit
sweep can follow the stimulus only after the elapse of a definite
reaction interval. It might be supposed that, by reason of its
hypothetical importance in the visual process, if not on purely
anatomical grounds, the reaction time of the eye would be
unusually short. The fact that it is in reality unusually long,
160-170°, indicates at once the relative unimportance of imme-
diate pursuit and a considerable elaboration of the stimulus in
what seems to introspection like a simple reaction. But any
reaction interval at all means that, at the beginning of a pursuit
sweep, neither the velocity nor the extent of the eye movements
parallels the movement of the object of interest. Before the
two could even approximate each other, the line of regard must
first overtake the moving object. It is evident that kinaesthetic
data from these pre-pursuit movements would not only be mis-
leading in themselves, but that if they ever came into operation,

8 R. DODGE.

general psychological law would tend ultimately to effect their
elimination. Naturally this objection to the interjection of a
kinassthetic factor in the visual perception of motion holds only
for the first phase of the pursuit sweep. But the recurring
positive corrective movements, together with the lagging of the
true pursuit movements in the second phase of the pursuit sweep,
are also incapable of furnishing reliable kinaesthetic data, either
for the perception of motion, or for the successive corrections of
the pursuit. Moreover it is evident that, before the second phase
of the pursuit sweep begins, the objective movement must have
been already apprehended both as to its direction and its velocity.

EXPERIMENTAL VERIFICATION.

Any attempt to verify the theoretical deductions from the
nature of the pursuit movements will be embarrassed by the
practical impossibility of isolating the hypothetical kinaesthetic
data. No natural pursuit movement, as we have already seen,
can yield the motor data in pure form. It must always be con-
taminated by some displacement of the retinal image. And,
unless the background be thoroughly homogeneous, all the
usual forms of modification of the retinal image may occur.
If all modification of the retinal image is to be avoided, an iso-
lated stimulus must be produced at the exact time of a homo-
geneous eye-movement; it must have the same velocity as the
eye-movement, and cease when the latter ceases. Since these
conditions can never be satisfied by a reactive movement of the
eye in response to a moving stimulus, the required homoge-
neous eye movement must be induced in some way independ-
ently of the moving stimulus, while they are, nevertheless,
exactly concurrent. These requirements appeared at first sight
utterly unrealizable. They are, however, fulfilled with striking
fidelity in an experiment which was originally arranged as a
test of the possibility of retinal stimulation during eye move-
ment.1

A disk of black cardboard, perforated near the periphery by

a concentric circle of small round holes made by a leather

punch, was rotated by suitable clockwork between the eye and

the clear sky. By purely empirical means, a critical velocity

1 PSYCHOLOGICAL REVIEW, Vol. VII., p. 458.

VISUAL PERCEPTION OF MOTION. 9

was found, such that, when a point just behind the perforated
disk was fixated, the intermittent stimulation through the per-
forations fused to a circle ; but when the line of regard was
allowed to wander in the direction of the disk's rotation, from
the primary fixation point to another about 4.7° distant, the
fused circle broke up into bright, clean-cut perforations. The
significant part of the experiment, in view of the present dis-
cussion, is the fact that, whenever the perforations were seen at
all, they appeared to be standing still. They flashed out from
the fused circle of light and disappeared again, apparently at
the same point in space. The experiment is singularly exact.
There was no opportunity for secondary corrective movements
of the eyes, since the whole duration of the eye movement was
less than 30*7, and corrective movements do not occur in move-
ments of five degrees from the primary point of regard. The
stimulus must have remained at approximately the same point
of the retina from the time it differentiated itself from the fused
circle of light until it disappeared. Finally, since, under ordi-
nary circumstances, as the line of regard passes from one fixa-
tion point to another in a motionless complex field of view,
there is no new effective stimulation of the retina ; at exactly
the same moment when the discreet stimuli appeared the entire
background must have disappeared. Only one of the tradi-
tional data for the perception of motion is present, namely, the
persistent stimulation of the same point of the retina through-
out homogeneous eye movement, and that signally fails to effect
a perception of motion.

One obvious objection detracts from the conclusiveness of
the experiment. If we accept the differentiation of the two
types of eye movements which I have been at some pains to
establish, we must acknowledge that our experiment proves
nothing for the true pursuit movements, but only for the eye
movements of the first type. The change of the line of regard
from the primary to the secondary fixation point was in no
sense a pursuit movement. It occurred as a simple reaction to
an eccentric stimulus, entirely independent of the moving per-
forations on the periphery of the disk. The question whether
the slower normal pursuit movements yield sensory data for

10 R. DODGE.

the perception of motion is consequently not answered by the
experiment. One important step, however, is taken. Since
the first phase of every pursuit sweep involves only movements
of the first type, it is clear that the apprehension of movement
must proceed entirely independently of data from the eye move-
ments until the second or true pursuit phase begins. This
means that kinaesthetic data are not available, if indeed they are
available at all, until so late in the process that they would be
useless if they ever existed.

There is no corresponding experiment for the true pursuit
movements. The one prohibitive circumstance is the presence
of the rapid corrective movements of the first type, and the
persistent lagging of the line of regard which occasions them.
There is, however, a form of eye movement which approximates
the velocity of the true pursuit movements, which is conspicu-
ously free from the minute corrective movements and the conse-
quent displacement of the retinal image. This form I have
elsewhere called the coordinate compensatory movements.1 The
type may be defined as those movements of the eyes by which
the constant fixation of an unmoved object of regard is main-
tained during rotation of the head. Photographic registration
of the coordinate compensatory eye movements proves conclu-
sively that they are not preceded by any reaction interval after
the head begins to move. They show no intercurrent corrective
movements and no lagging of the pursuit unless the head move-
ments are extremely rapid. The movements of the third type
are, consequently, true pursuit movements in everything except
in origin. Both approximate the apparent angle velocity of the
object of interest ; only in the third type the fixation is without
interruption and without measurable error, so that there is no
discoverable displacement of the retinal image.

Disregarding the matter of origin, which theoretically ought
to have no influence in the matter, these characteristics conform
admirably with the experimental requirements for isolating the
kinaesthetic factor, if it exists. If an intermittently luminous
point of light is fixated in an otherwise darkened room, while
the head is rotated slowly from side to side, on a vertical axis,

1 American Journal of Physiology, Vol. VIII., p. 322.

VISUAL PERCEPTION OF MOTION. II

through from io°-2O°, it will be found that there is no apparent
motion of the point of light so long as the intermittent flashes
fuse completely to one undistorted point. If, subsequently, the
velocity of the head movements is increased to the maximum,
the coordinate compensatory movements will no longer be
exact. The point of light will appear distorted or multiplied,
and coincidently there will be a marked illusion of motion of the
luminous point. This seems to me an almost perfect verifica-
tion of our theoretical conclusions. We have produced an eye
movement of the general characteristics of the pursuit type, in
which an undisturbed fixation of the object of regard is main-
tained without any of the other cues of motion. As long as
these conditions persist there is no appearance of motion, not-
withstanding almost continuous eye movement. The moment
a slight displacement of the retinal image occurs, however, there
is a vivid illusion of motion, which there appear to be no
kinsesthetic factors to correct.

A simpler though in some respects less satisfactory variation
of the above experiment serves the double purpose of an easy
test of the main point and an answer to a possible criticism. If
a grating of fine wire is suspended about half way between the
subject's eye and a smooth wall with a single conspicuous
figure, it will be found that, when the grating is fixated during
moderately rapid movements of the head on a vertical axis, the
wall figure will appear to move back and forth behind the
grating. Whenever, on the other hand, the wall figure is
fixated during similar movements of the head the grating will
appear to move. The principle is the same as in the dark-room
experiment : coordinate compensatory movements of the eyes
maintain the fixation, once established, whether of the grating
or the wall. In either case the image of the fixated object
remains practically motionless at the fovea. The eccentric
position of the eyes with relation to the axis of the head, and
the consequent lateral displacement of the eyes during each
head movement causes an apparent displacement of the objects
lying along the line of regard, which varies directly with the
lateral displacement of the eye and inversely with the distance
of the object. There is of course no real motion of the objects,

12 H. DODGE.

and no reason outside the habitual interpretation of the sensory
cues why the illusion of motion should attach itself to one
object in the immediate foreground and not to another. That
those objects appear motionless whose images remain motion-
less on the retina during head and eye movement, while the
slightest displacement of the retinal image causes the illusion of
motion, clearly indicates not only the utter irrelevance of the
hypothetical kinassthetic data, but also the real source of the
relevent data. In view of the fact that the amplitude of the
coordinate compensatory eye movements varies indirectly with
the distance of the point of regard as well as directly with the
amplitude of the head movement, it would be absurd to object
that the kinsesthetic data from the eye movements were in some
way counterbalanced by the kinassthetic data from the head
movements in the opposite direction. Finally, it would be inde-
fensible to contend that, in this particular form of eye move-
ment, experience had eliminated the false kinassthetic data,
since the same experience ought also to have eliminated the data
which occasion the persistent illusion.

It is not improbable that eye movements of the third type
compensate for other bodily movements besides those of the
head ; but the difficulties of registration have thus far prevented
a demonstration of the hypothesis. There is, however, some
indirect evidence in the fact that phenomena, similar to those
described above, may be observed by myself when I walk and
when I sway the trunk at the hips. The point fixated always
seems to remain fixed ; while adjacent objects which lie consi-
derably nearer than the object fixated, or which lie considerably
further away, seem to move up and down at every step. The
fact that some of my students have not obtained self-consistent
results from this last form of the experiment may be due to the
general difficulty of maintaining a constant fixation for one
point while steadily observing another; or it maybe due, on the
other hand, to faulty compensatory movements, such as are
demonstrable in my own case, whenever I try to maintain a
constant fixation while rotating the trunk on a vertical axis at
the hips, when the neck is held stiff. Whenever the latter
explanation is the true one, the eye movements will be of the

VISUAL PERCEPTION OF MOT/ON.

»3

second type, and will give evidence of the minute corrective
movements which belong to that type. In my own case, just
described, this results in illusions of motion covering the entire
field of view, irrespective of the distance of the object fixated.
If the bodily movements are continued long enough, dizziness de-
velops. In the case of one of my pupils it was possible to demon-
strate a faulty fixation, and this is my excuse for what might seem
like an impertinent caution to those who repeat the experiment.
The familiar attempt to measure the importance of the motor
factor in the visual perception of motion by the least perceptible
motion of a point of light in a dark room needs no detailed cri-
tique in addition to what we have already said concerning the
character of the pursuit movements. But besides the displace-
ment of the retinal image, incident to the initial reaction interval,
and to the more minute corrective movements of the pursuit,
there are new complications introduced by the involuntary lapses
of fixation, and the consequent persistent illusion of motion
with which everyone is familiar who has worked with isolated
visual stimuli. While the ordinary form of the dark-room ex-
periment is thus rendered altogether equivocal and meaningless,
a modification of it was accidentally hit upon, which constitutes
a faultless experimental test of our conclusions. We have al-
ready called attention to the fact that the end of every pursuit
sweep is freer from corrective movements than its beginning.
This is conspicuously true of the pursuit sweeps by which the
line of regard follows a swinging pendulum. Photographs of
such sweeps give no indication of corrective movements either
negative or positive within the last quarter of the swings studied.
This ought perhaps in itself to have suggested the experiment.
That the observation actually occurred without premeditation
only made it the more striking. We were studying Exner's
comparison of the apparent velocity of a moving object when
pursued and when not pursued, and as a variant of his experi-
ment we used the long counterbalanced pendulum which was
previously used to furnish the stimulus for the above-mentioned
photographs. Movable points of light were attached to the
pendulum rod, one above the axis and one below. If the dis-
tance of both from the axis was equal, both would move through

14 R. DODGE.

equal distances in the same time. The one fixated however
always appeared to move much less than the one seen periph-
erally. It was found that if the two were to appear to move
through equal arcs, the pursued must actually move through
about three times the arc of the unpursued. This of course
could be accurately measured by the relative distances of the
two points from the axis.

This alone is good evidence that the hypothetical kinaesthetic
factors in the perception of motion must be of less relative
importance than the displacement of the retinal image. The
objection might still be raised, however, that, if the fixated
moving point be seen to move at all, some kinaesthetic data
must be postulated. The force of such an objection has been
already weakened by the preceding demonstration of the pres-
ence of corrective eye movements in all true pursuit movements.
But while these corrective movements always involve some dis-
placement of the retinal image, it is obviously difficult to dem-
onstrate that in any one case the fortuitous displacement of the
retinal image entirely accounts for the perception of motion.
The most interesting and conclusive phase of the experiment I
have never seen reported but it may be easily verified with the
simplest kind of apparatus. When the point fixated approaches
its extreme position in each oscillation, it seems to rest for an
appreciable interval, while the other point seems to continue
moving as though the two were connected by an elastic rod,
which regularly gave the unfixated point a considerable addi-
tional oscillation after the fixated point had been arrested at the
end of each swing. The illusion is persistent and striking, and
is capable of only one explanation. It occurs at that part of
the pursuit movement which photographic registration shows to
be practically free from corrective movements. The fact that
the point whose image remains motionless on the retina during
an unbroken pursuit movement seems to stand still, while the
other point, which is in reality moving no faster than its fixated
companion, seems to make a little gratuitous whip-lash excur-
sion, serves at once to show the utter inability of the pursuit
movement either to subserve the perception of motion of the
fixated point or to correct the exaggerated data from the dis-
placement of the retinal image of the non-fixated point.

AN INQUIRY INTO THE NATURE OF
HALLUCINATIONS. I.

BY BORIS SIDIS,

Director of the Psychopathic Hospital and Laboratory of the New York Infir-
mary for Women and Children.

The subject of hallucinations forms the stumbling block both
of the psychologists and psychopathologists. The deeper one
penetrates into the subject the greater confusion he encounters.
Some regard hallucinations as being of peripheral origin, others
regard them as central in character, while still others go to the
extent of claiming that the most central hallucinations are of
supernatural origin, being communications and messages from
a transcendent world. In view of the great importance of the
subject it may be well to make an attempt to throw some ad-
ditional ray of light on this obscure matter from the standpoint
of psychological and psychopathological analysis.

The usual definition of illusion and hallucination is with re-
gard to the external object. Illusion is defined as fallacious
perception of some actually existing object, while hallucination
is perception of a non-existing object. This definition is good
for practical purposes of the clinician, but it is not psychological.
From a strictly psychological standpoint illusions and halluci-
nations cannot possibly be differentiated from other psychic
states by the presence or absence of external objects. External
objects can hardly be regarded as constituents or necessary in-
gredients of psychic states. Illusions and hallucinations should
be defined in terms of psychic processes. As far as process is
concerned it is quite possible that the same processes underlie
both normal and fallacious or abnormal perception. In order to
get a clearer insight into the nature of illusions and hallucinations
it may be well to begin with a brief analysis of the process of
perception.

'5

1 6 BORIS SID IS.

PART I.

If we take a cross-section of a moment of consciousness
and try to fixate it with our mental eye, we find a central psy-
chic experience, or psychic element round which other psychic
experiences or psychic elements are crystallized and organ-
ized. This psychic experience, or central element, is promi-
nent, vivid and constitutes the vital point of all the other
organized states, giving the tone to the rest, to a whole, to one
organized experience. The psychic matter that surrounds the
luminous central point does not stand in a free more or less dis-
connected relation to the latter, it is intimately related to the
center and cannot be separated without destroying the moment
as a whole and even the life existence of each particular con-
stituent. The whole moment seems to form an organic network
in which the other elements take their place according to a plan.
The structure of the moment may in this respect be compared
with that of the cell. In the cell we discriminate a nucleus
round which cytoplasm is grouped. The protoplasm is con-
nected with the nucleus by a network imbedded in the cyto-
plasm by a cytoreticulum. The destruction of the nucleus
affects the cytoplasm and the destruction of the cytoplasm
affects the nucleus. The two are intimately, organically inter-
related by the common network, the general plan of their or-
ganization. If we closely examine the percept, we find in it a
central sensory element surrounded by other elements. This
central element stands out prominently in the given psychic
state, while the other elements are subordinate. Not that those
elements are unimportant for the percept, on the contrary they
are of the highest consequence and moment, they only lie out-
side the focus of the mental state. Along with the focus those
elements form one organized whole. All the elements of the
percept form one texture having the central sensory element as
its nucleus.

Integrated, however, as all these elements are they are not
of equal value and importance to the life existence of the whole.
The central sensory element is of the utmost consequence, it is
the vital point of the total experience. While the change, or
destruction of one or of some of the subordinate elements may

THE NATURE OF HALLUCINATIONS. l^

still leave the total percept unchanged, or but slightly modified,
a change of the central sensory element or of the nucleus will
profoundly modify all the other elements and their interrelation.

The elements of the percept may be regarded as bound up
in a * chemical ' compound, so to say, together giving rise to the
qualitative aspect of the total combination, the difference being
that in the psychic compound there is a central element that gives
the keynote to the combination ; in the chemical compound the
elements are all equal in value and importance. To form water
for instance, an atom of hydrogen to two of oxygen is required ;
the oxygen and the hydrogen are both equally requisite to the
formation of the compound ; one is not more important than the
other. Not so is it in the psychic compound ; there the ele-
ments are of unequal value. The most important of them is
the nucleus ; it determines the interrelation of the psychic ele-
ments and also the outcome of the whole combination.

From a biological standpoint we can well see why this
should be so. A psychic compound is biological, not purely
physical or chemical. In the physical components there is no
higher and no lower, all are of equal value ; in the psychic, as
in all life existence, there is a higher and a lower in structure.
In other words, the biological compound is peculiar and dif-
ferent from the chemical, inasmuch as the former is really not
a compound, but an organization. The characteristic of organi-
zation is just this systemic combination of parts related in dif-
ferent grades and orders of importance for the total life exis-
tence of the whole.

Looked at from another standpoint we can further see the
necessity of such a central element. We have pointed out in
another place that one aspect of the biological process is that
of purpose, and if that be granted, then psychic processes re-
garded as highly developed biological processes should present
this general characteristic of purposiveness in its fully developed
form. Now, where purpose is involved the end alone is the
important thing, all the other elements are for that end, sub-
ordinate and determined by it. Not that the other elements
are unimportant ; they may be intimately related, but they, after
all, are only means to accomplish the end. The elements that

1 8 BORIS SID IS.

have for the time being the organizing power to aggregate
round themselves the proper elements and lead towards the re-
quired end are predominant. For every psychic state is essen-
tially for some reaction and that sensory element which gives
the cue for the formation of the psychomotor elements leading
to some given reaction is for the time being the center, the
nucleus of the total state.

The flower before me attracts my attention. I see its color
of a light violet tint, its rounded bell shape and its velvet-like
softness ; I stretch out my hand carrying the flower to my nose,
to enjoy its fragrance. What I really see is the light violet tint ;
the rest of the elements are not given directly, they are largely
inferred. The shape is largely an inference from previous
muscular eye-movements and its softness is derived from pre-
vious tactual experiences. They are all, however, perceived
by the eye, the cue being given by the prominent sight sen-
sation. Although the flower as percept appears as an organ-
ized whole, still the sensation of sight forms the nucleus round
which the others cluster ; the perceptual tone is given and de-
termined by the particular sensations of sight. The softness,
though perceived, is still altogether different from the sensation
softness as directly experienced through the organ of touch ; it
is a sensory sight softness. The same is true of all the other
sensory elements ; they are all essentially determined in their
quality by the central sensation. In seeing a lump of ice we
perceive its whiteness, its transparency, its hardness, its smooth-
ness, etc. The hardness and smoothness are essential elements
of the percept ice. These elements seem to be given directly
in sensory experience. They seem to be directly perceived
and still are qualitatively different from the hardness, smooth-
ness and coldness as given directly by the experience, when
the hand gets hold of a smooth lump of ice. The sensory ele-
ments are determined and colored by the central visual sensory
elements. The sensory elements of coldness, hardness, smooth-
ness are of a visual character.

It is usually claimed that such additional elements that are
not given directly by the stimulated organ are elements repre-
sentative in character, derived from memory. This statement

THE NATURE OF HALLUCINATIONS. 1$

is not quite correct. In seeing the piece of ice the hardness,
smoothness are not represented, they are presented to the eye ;
we really seem to see, to experience these sensations going to
make up the percept ice. It is not true that on catching sight
of a whitish, transparent, glittering lump we remember that it
is also hard, smooth and cold. The whole percept with all its
sensory elements appears at once in the synthesis of the percept
ice — we see, we perceive the hard, smooth, cold ice. The
hardness simply remembered is altogether different from the
hardness perceived in the seeing of the lump of ice. The hard-
ness, smoothness as they appear in the ice are different to the
eye and as different psychologically from the corresponding
representations as the latter differ from the corresponding sen-
sations directly experienced by the appropriate sense organs.

Pathological cases seem to confirm the same point of view.
There are certain mental diseases, when the patient can per-
ceive the object correctly, though he cannot represent it to
himself. The patient can perceive all the elements on being
confronted with the object, but by no means can he remember
them. On the other hand, there are cases when the patient can
easily represent to himself objects, but cannot recognize the
object when directly confronted with it.

The subordinate elements in their turn play an important
role in the total unity of psychic experience, in the percept, in-
asmuch as they give the content of the total moment, fermented,
so to say, by the predominant nuclear element. The visual
elements, the perception of play of light and shade would
have been nothing but mere play of light and shade, if not for
the subordinate tactual and motor elements that give rise to
the perception of distance, dimension, size, body. The leading
element gives the character to the content by having it appear
under its own special sensory aspect, while the other elements
give the content to the whole moment. Thus where space is
perceived through the organ of sight mainly, the space is visual in
character, though the content that gives rise to the perception
of space itself is filled in by other psychic elements.

The central element with its content may be regarded as the
nucleus of the cell surrounded by its cytoplasm and the total

20 BORIS SID2S.

organic whole may be termed psychic compound. Now in the
psychic compound the constitutent elements of the content can
no longer be directly discriminated. In the lower forms of
mental life the elements are firmly bound as we find it to be in
the psychic compound — the percept. In the forms where the
elements are more complex the synthesis results in fusion in
which the elements can be more or less easily discriminated.
Thus if one listens to the beats of a metronome and to the rhyth-
mical vibrations of a pendulum the sensations blend and fuse, the
sounds seem to proceed from the vibrating pendulum. The
same case is well illustrated in the well-known amusement of
having one hidden behind a screen and making a speech, while
another one is watched who is gesticulating in accordance with
the modulations of the speaker's voice. The two series of sen-
sations blend and the voice seems to proceed from the gesticu-
lating person. The "synthetized elements here are fused or
more or less 'mechanically' joined instead of being firmly
combined in a sort of ' mental chemistry.'

In the higher and more complex mental states the constitu-
ents of the synthesis are neither ' chemically,' nor ' mechani-
cally ' fused. The constituents stand out free and distinct.
While I am writing this page I see on my table at one glance the
light, the lamp, the paper, the pen writing words and lines and
hear and feel the pen move on the surface of the paper ; at the
same time I see the table, the chair in the room and hear the
ticking of the clock. The multiplicity of all these experiences
is simultaneously synthetized and at the same time discriminated
in the unity of the total experience.

In the processes of succession of complexes of psychic
elements, in the trains of ideas, the constituent elements attain
their greatest freedom and independence. In the states of per-
ceptual synthesis, on the contrary, the directly experienced
sensory elements constituting the nucleus of the percept calls
out immediately its appropriate associated perceptual elements
and the compound, the percept, appears, as a whole that can
be analyzed only under highly artificial conditions. Fixity is
the essential characteristic of lower mental stages as well as of
the percept.

THE NATURE OF HALLUCINATIONS. 31

The constituent psychic elements are so intimately united in
the percept that they resist efforts at decomposition. If a per-
cept A is composed of elements a, Z>, c, d, and if a be the
nucleus, the stimulation of the nucleus brings out the rest — £,
c, d. The central, or nuclear element is purely sensory, but
the rest of the psychic elements £, c, d, are not sensory in the
same sense as a is, since they are not derived from direct stimu-
lations of the appropriated sense organs. Their character is
not primarily, but only secondarily sensory. The retina gives
only light sensations. The most differentiated and highly
organized retinal structure of the highest vertebrates such as
that of the higher mammals can only give rise to local signs, to
highly differentiated light sensations varying with each retinal
point or element, or cellular termination. The stimulated
retinal elements with the neuron terminations of the optic nerve
can give nothing else but light sensations and the image formed
on the retina is in fact nothing else but a series of light sensa-
tions. If this be so, how then do we come to see that tangible,
heavy, solid body yonder at a more or less definite distance?
Solidity, bulk are not given in light sensations as such, how
then are those spatial and physical characters perceived so dis-
tinctly as to assume a direct sensory character? It cannot be
ascribed to the principle of association of ideas. For the object
and its distance appear at once in one single glance before any
idea comes to the mind. Furthermore, an idea from its very
nature stands out distinct and definite ; it is essentially free, but
the psychic elements of object and distance are not discrimi-
nated. Again phylogenetically and ontogenetically sensation
and perception precede ideation. The infant, the animal per-
ceives objects and distance and certainly with little or no idea-
tion present. In the visual perception of distance the subordi-
nate psychic elements derived from other senses are not of an
ideational character, they are of a sensory character. The eye
sees the distance. The eye sees distance or volume directly,
because of other elements involved in the process of perception,
such as the kinaesthetic sensations coming from the movements
of the eyes in their adjustment to the stimulations from the
external environment, also tactual, muscular and kinaesthetic

22 BORIS SID IS.

sensations derived from skin, muscles, joints and articular sur-
faces, all synthetized in the given percept. The subordinate
psychic elements are neither of the character of pure sensations
nor are they of the nature of pure ideas. What are they?
They seem to be intermediary in character, intermediary be-
tween the nature of sensation and that of idea. Perception
appears to be an intermediary process.

We may regard the same process from a hypothetical
physiological standpoint which may possibly help us in pictur-
ing the mechanism. A specific physical stimulus produces in
the peripheral sense organ a definite physiological process
which is transmitted to groups of neuron systems stimulating
them to activity and giving rise to specific physiological proc-
esses. Whenever these specific physiological processes are
peripherally induced, the special sensory elements arise. If
groups and systems of such psycho-physiological elements
become associated and organized round a central nucleus, the
result of the functioning activity of the total organic complex is
a psychic compound, a percept. Whenever one of the groups
is peripherally stimulated and is awakened to activity, the other
elements become stimulated and the result is the organized
activity of function of all the elements, thus giving rise to the
synthesis of all the psychic elements, namely the percept.

Now we should postulate some difference in the pyschic
state as to whether psychophysiological elements are stimu-
lated directly through their own appropriate sense organ or
whether they are awakened to activity indirectly through other
sense organs. The direct peripheral stimulation gives rise
to psychic elements characteristic of the particular sense organ
and its nervous tracts and central systems of neurons, whilst
the indirect peripheral stimulation gives rise to psychic ele-
ments whose pure and real sensory character is not clearly
revealed in the total psychic state or moment. These indi-
rectly induced sensory elements are so much colored and
infused with the sensory qualities of the nuclear sensory
elements that their character and origin are transformed and
they appear not to differ in their nature from the nuclear
elements. A closer inspection however fully reveals their

THE NATUKR OF HALLUCINATIONS. 23

real nature as sensory elements extraneous to the nuclear
elements and derived from different sensory sources. The
nuclear elements are primarily derived, in so far as they are
directly initiated by the incoming peripheral stimulation, while
the extra-nuclear sensory elements may be regarded as sec-
ondarily initiated by peripheral stimuli.

Let V be the sensory visual system, T and M tactual and
sensory motor systems, A sensory auditory systems. Let Vl
be the visual sensations peripherally stimulated, 7\, M^ A{, the
tactual motor and auditory sensations of the corresponding
sensory systems. Let, further, F2, T2, M2, A2 be the psychic
elements indirectly or secondarily initiated ; then the percept
when Fj is the nucleus may be represented by Vl T2M2AZ.

Psychic elements primarily or secondarily peripherally ini-
tiated are not identical with ideational states. An idea differs
qualitatively from a percept and its elements — an idea lacks
sensory character. An idea is more generic, while a percept is
more specific. T see that lamp-post yonder ; it is a particular
object rigidly limited in a particular space ; not so is the idea,
the idea of the lamp-post refers to lamp-posts in general.
When I perceive an object and then try to represent it to myself,
the object is not presented to consciousness in its sensory per-
ceptual form — it is present to consciousness rather as a symbol
ideally representing perceptual experience peripherally initiated.

From an anatomical and physiological standpoint it is quite
probable that ideo-motor systems are different neuron organiza-
tions from those of the sensory-motor systems. Psychopath-
ology with its rich store of facts seems to favor this view. As
we have already pointed out there are pathological cases
when the patient does not know the object on perceiving it, al-
though he can represent it to himself and again there are other
cases where the patient cannot represent to himself the object,
but he knows the object on perceiving it. Flechsig's embryo-
logical studies go further to show that the sensory centers are
different from the associative centers which do not stand in di-
rect relation with the external environment and appear rather
late in the course of ontogenetic development. The view often
maintained that the same sensory structures underlie both sen-

24 BORIS SID IS.

sory and ideational processes does not seem to be probable in
the light of recent research. The activity of the sensory-motor
neuron systems does not give rise to ideas, but to psychic states
essentially sensory in character. In the case of the percept the
subordinate psychic elements entering into the synthesis of per-
ceptual psychic compounds are of a sensory nature ; they only
differ from pure sensations in so far as they are not directly
periperally initiated, but centrally, or truer to say, indirectly peri-
pherally initiated and as such occupy an intermediary state be-
tween sensation and ideation. In other words, the subordinate
perceptual elements may be regarded as reflex in character, as
being of the nature of secondary sensations.

The nature of illusions and hallucinations is more or less
cleared up from this standpoint and the latter in its turn may be
still further illustrated and confirmed by the facts coming from
the domain of abnormal mental life. Let us take a series of
cases of abnormal or fallacious perception. In looking through
the stereoscope the two plane dissimilar views are combined and
give the illusion of a solid object. Here the illusion is due to
imitation of external conditions ; the external stimulations that
give rise to the perception of a solid object are here closely re-
produced. The visual sensory elements are stimulated and the
rest of the groups are reproduced, the rest of the sensory ele-
ments or secondary sensations emerge and the perceptual syn-
thesis arises. The illusions to which in my student days I
attracted Professor Miinsterberg's attention are of similar char-
acter. If each eye looks through a separate tube and if the
other ends of the tubes are brought together, the openings of
the tubes coincide, appearing as one, and the eye appears to
look through one tube only. If now only one tube is looked
through and the other eye glides along the surface of the tube
the opening of the tube appears outside, removed and raised
higher than the real opening ; the opening appearing to be
directly seen not by the eye which looks through the tube, but
by the other eye that does not look through. The illusion can
be emphasized by putting the hand where the illusory opening
appears and the hand appears to be pierced by a round hole.
Here the conditions are such that the convergence of the eyes

THE NATURE OF HALLUCINATIONS. »5

displaces the lighted-up opening towards the field of vision of
the open eye not inclosed in the dark tube. Similarly when
closing one eye and having the other wide open we press the
closed eye sideways towards the nasal side the round phos-
phene seems to be projected into the field of vision of the other
eye and the phosphene really appearing in the field of the
closed eye as one can convince himself by closing the open
eye, appears to be directly seen by the open eye. In all these
experiments the arrangement is such as to imitate conditions
under which other percepts normally arise and the result is the
reproduction of those specific states of perception. To take
another example, in a fog or in the darkness we may take a
tree for a man or mistake a rope for a snake. Similarly, in
the shape of clouds and blots we can often see different figures.
The illusion here is rather due to the vagueness of the cue or of
the sensory nucleus, the character of which may vary with dis-
tance or with the intensity of light.

In mental derangements such as in the different forms of
insanity or of psychopathic functional diseases, in hypnotic,
posthypnotic and hypnoidic states the object is perceived as dif-
ferent, independent of external conditions, such, for instance, as
convergence, divergence, light, distance. A chair may be
perceived as a tiger no matter how the visual axis is placed or
what the distance be, or how intense the light is. Certain defi-
nite visual sensations may be correctly perceived, but on account
of central dissociation in psychopathic states quite different than
the customary associated secondary sensations are aroused which
in turn arouse different secondary sensations in other sensory
motor systems of neurons and the result is a different psychic
compound, an illusion or a hallucination.

In the preceding cases the nuclear elements obscured in
different ways by the subordinate elements are nevertheless
present in consciousness and still form the nucleus of the per-
cept into which other subordinate elements enter as organic con-
stituents, and give rise to fallacious perception. Should now
the nuclear elements themselves on account of inattention or of
their minimal sensory intensity, or what is still more often the
case, on account of states of dissociation, should such nuclear

26 BORIS SIDIS.

elements be left out of consciousness or remain in the subcon-
sciousness as in dissociative states, then the fallacious percept
stands out clear and distinct in the light of consciousness and a
fully developed hallucination results. Sensory elements which
themselves may remain unperceived stimulate other sensory ele-
ments that give rise to a perceptual compound which is entirely
of a secondary sensory character. The hallucinatory percept
does not contain the primary sensations aroused by the stimulus ;
it consists of secondary sensory elements and as such a hallu-
cination may be regarded as a secondary percept. Hallucina-
tions are of the nature of secondary sensations.

The simplest state of hallucination is possibly found in the
phenomena of synaesthesia or in the phenomena of secondary
sensations, such as light-phonisms, sound-photisms, etc., when
one sensation, instead of giving rise to a subsequent idea,
awakens instead a qualitatively different sensation derived from
another sense organ — a color or letter arousing a certain
sound, definite sounds arousing certain colors and so on. When
a certain stimulus makes an impression on a peripheral sense
organ and gives rise to secondary sensations, we really have a
hallucination, but in its simplest form. He who on seeing the
letter A, for instance, also hears a sound or feels a prick, or a
touch may be regarded as having a hallucination. In this
simple form we can possibly more clearly discriminate the char-
acter of hallucination. When on seeing letter A, we hear a
sound, the indirectly aroused auditory sensory elements do not
contain the primary sensory visual elements. In the secondary
sensation or in the more complex state of perception of sec-
ondary character the primary elements are left out. A stimulus
may arouse sensory elements in one sensory center, which in its
turn may stimulate systems of sensory elements in other sensory
centers, thus giving rise to a group of secondary sensations syn-
thetized into a percept, while the original sensation with its
nuclear sensory elements may remain in the background. Such
a physiological stimulus may often be not an external physical
stimulus, but a pathological process going on either in the pe-
ripheral sense organ from which the nuclear sensory elements
arise or in the sense organs from which the secondary sensory
elements originate.

THE NATURE OF HALLUCINATIONS. 27

Let 6" be the stimulus and V^ the visual sensory elements
and At T^M.^ the secondary sensory elements, then f, may be
dissociated while the secondary elements A2TtMt stand out
alone in consciousness as a secondary percept or hallucination.

It may again be that not only the primary but also the ap-
propriate system of secondary elements may be left out of con-
sciousness, while associated systems of secondary elements may
be awakened and stand out fully in the light of consciousness
and thus give rise to a hallucination removed in its character
from the original primary elements with their organized sec-
ondary elements.

The preparedness of remotely aroused secondary groups may
often be determined by the type of mental structure. Halluci-
nations of visions, or of voices, or of movements will predomi-
nate, according as the type of mental structure is visual, audile,
or motile. The mental type plays, no doubt, a very important
part in the formation of illusions and hallucinations. In the in-
sane auditory illusions and hallucinations predominate in the
audiles ; and while, on the one hand, paranoiacs are often
audiles, on the other hand, audiles are inclined to paranoia. In
hypnosis hallucinations become more easily realized, if they are
adapted to the mental type of the subject.

Preparedness and subexcitement of ideo-motor groups with
which the secondary sensory groups are associated also form
an important factor in the final determination of the character of
the illusion or a hallucination. This is rather of an indirect
character. It is not that the ideo-motor groups themselves directly
enter into the structure of fallacious perception, but they often
may determine which of remote secondary sensory groups should
be stimulated to activity. Groups of elements are more easily
brought into active functioning the greater the activity of the
elements with which they are associated, the course of group
excitation being, so to say, in the direction of least resistance.

Pathological processes going on in one sense organ may
sometimes give rise to secondary sensory elements belonging to
other sense organs especially when favored by general states
of dissociation ; in fact we may say that from our point of view
a state of dissociation is an indispensable condition to the for-

28 BORIS SIDIS.

mation of hallucination. The following cases may be taken as
clear typical instances. Thus in one of the cases reported to
me by one of my associates, Dr. Wm. A. White, the patient
saw spirits and regarded them as ghosts of her deceased
daughter. On examination her eyes were found to be normal
in all respects. The patient saw the spirits even when her eyes
were shut, and furthermore the hallucinations were not in the
least affected even when her eyes were injected with atropine.
When, however, the ears were examined a pathological
process of old standing was discovered. Now when audi-
tory stimuli were applied to the ear, the hallucinations were at
once strongly affected, the spirits multiplied in number. This
increase of spirits ceased as soon as the auditory stimuli were
removed. A closer examination revealed the fact that the
patient was greatly affected by the loss of a daughter. The
pathological process served as the stimulus, while the excit-
ability of the ideo-motor systems along with the general state of
dissociation determined the nervous processes initiated in the
ear in the direction of the sensory visual systems and gave rise
to secondary sensory elements formed in the hallucinatory per-
cept of ghosts and spirits resembling the patient's daughter.
The aural pathological process itself remained in the back-
ground of consciousness and was unknown to the patient.

A similar case came under my notice in a paranoiac who
had visual hallucinations of spirits, hobgoblins and saints. The
organs of sight and hearing were found normal, but a patho-
logical state was found in the skin of his scalp and especially
in the muscular sensibility of the muscles of the neck. An in-
clination of his head in any direction caused him to see the
spirits and hear their voices. In another case of mine definite
auditory stimuli such as the singing of birds brought about
hypnoidic states which are really complex states of hallucina-
tions. In another case, in a female paranoiac with clearly defined
auditory hallucinations, a similar state was revealed. The pa-
tient heard voices not through the ear, but through a spot located
just over the region of the Fallopian tubes. Examination of
the spot revealed tenderness and painfulness to pressure. The
hallucinations, which were of a sexual character, became ex-

THE NATURE OF HALLUCINATIONS. 29

acerbated during the menstrual period. Similarly in another
case under investigation the auditory hallucinations were shown
to be intimately connected with phenomena of unconscious
phonation and with frequent earaches, with a limitation of the
field of vision due to an error of refraction which, when cor-
rected by eye glasses, modified the auditory hallucinations, the
latter finally becoming dissolved. More cases of similar nature
could be adduced, but the ones referred to are sufficient, and ex-
treme as they are, they bring out clearly the secondary reflex
character of hallucinations. Hallucinations are essentially sec-
ondary percepts.

Hallucinations are frequently due to peripheral processes,
pathological or otherwise, occurring under conditions of disso-
ciation, within the same sense organ, but the reflex halluci-
nations originating in other sense organs bring more clearly to
light the secondary nature of hallucinations. The contention
generally maintained that there are hallucinations independent
of peripheral sources, or of ' purely central origin ' which some
even regard as supernormal experiences is highly dubious. As
far as directly observed facts go, whether they be normal or
abnormal, there is little to justify the central point of view.
Like percepts hallucinations are peripheral in character, and
are only in so far central as peripherally initiated secondary
sensations are concerned. Hallucinations are of peripheral
origin and may be regarded as complex cases of secondary
sensations with the original primary sensation dissociated from
or left in the background of consciousness. If, however, hallu-
cination is abnornal perception, perception, on the other hand, is
normal hallucination. If a hallucination is a secondary com-
pound with the primary sensations ABSENT, a percept, in so far
as it consists of secondary sensory elements, is a hallucination
with the primary sensations PRESENT. Normal perception, illu-
sion and hallucination have the same underlying process and
as such may be arranged in a continuous series, according to
presence or absence of the primary sensory elements.

( To be concluded. )

THE LIMITS OF PRAGMATISM.1

BY PROFESSOR J. MARK BALDWIN,
Johns Hopkins University.

Pragmatism — variously understood, variously approached,
variously turned to account — is in the air. I shall not venture
to define pragmatism, much less to ascribe any definite form of
it to any particular writer ; on the contrary, I shall merely state
broadly the general idea involved in it, as a way of looking at
things, and from the outcome try to reach an understanding as to
what in fact it is.

Broadly speaking, enquiries are pragmatic which, with more
or less thoroughness, make such conceptions as thought, exis-
tence, truth, reality, etc., relative to other terms in a movement,
development, or evolution ; relative to antecedents, consequences,
modes of function, ends. All such determinations are not
only ends reached in a movment, but also means to ends yet to
be reached; and all of them, considered thus functionally, as
terms of genetic organization, in so far forbid definition in a
static, absolute, once-f or- all-fixed system. Now whether or
not all those who call themselves pragmatic — not to say prag-
matists — admit that this fairly characterizes that feature of their
thought, still that is what I now mean, and that is what this
paper is about.

In the theory of knowledge, it is one of the main claims —
and one of the prime advantages — of pragmatic theory, that
it avoids and denies any dualism between reality and thought,
in the sense that thought, or knowledge, somehow represents or
reveals a system of realities which are already fixed, definite,
and absolute, apart from the processes of cognition. It claims
that reality is determined as truth, and truth is a mode of mental
organization. It claims to be able to point out the adequate
processes of knowledge and action, of selection and systemati-
zation, which determine truth ; and thus to yield a theory of

1 Paper read before the joint seminaries, Department of Philosophy,
Princeton University, December i, 1903.

30

THE LIMITS OF PRAGMATISM. 31

reality which admits dualism only in the mechanism of the
psychological processes themselves.

Now fully admitting both this claim and this advantage up
to a certain point — I have myself worked out in recent publica-
tions such a view of truth, and of reality as cognized — I wish
now to suggest and discuss certain limitations of this standpoint.1

The discussion may be brought on under suggestions made
in reply to the general question whether there is anything in the
conception and implications of reality considered reflectively
and for the purposes of philosophical theory, which is not ex-
plained both as to its origin and mechanism, and also as to its
validity, by this theory of pragmatically determined cognitions
or truths. And this question may be resolved into certain more
restricted ones :

1. Are there any realities apprehended apart from the cog-
nitive function, or at least not adequately apprehended through
it? If so, what is their relation to cognitive reality or truth?

2. Are there any realities not, or not yet, discovered at all;
and if so what meaning do they have for us?

3. Are there any types of thought, or modes of treating
reality generally whose meaning is not exhausted in the state-
ment of their pragmatic origin?

I

These are large questions, and it is the outcome of centuries
of effort that rationalism or rational idealism — the philosophy
with which pragmatism comes most evidently into opposition —
has worked out intelligible answers to them all. Its answer to

1 This suggestion of the need of limitation, made in articles in the writer's
Diet, of Philosophy (arts. ' Pragmatism,' and ' Truth in Psychology ') is cited
by Mr. Schiller in his book Humanism, p. 8, as involving an inconsistency. I
wish to work it out here, fully maintaining this theory of the relation of truth to
reality, however, as suggested there (art. 'Truth') and stated out more fully
earlier in the address 'Selective Thinking' (now Chap. VII. in the work De-
velopment and Evolution). Mr. Schiller asks how on pragmatic principles of
origin we can ' get at reality without knowing it ' ; I reply (see the end of this
paper), by some other experience better qualified to report it exhaustively.
His other question — how our estimation of what truth is can disregard and be-
come independent of our modes of establishing it ? — is also answered here : ».
e., they can not, but they (our estimations of truth) can include and revise the
results of the theory of the establishment of it.

32 J. M, BALDWIN.

the first question is : ' No ' — but with hesitation ; for it has great
difficulty in mediating the first contact of thought and reality.
To the second it answers : " Yes ; there is a system of real, but
cognizable, relationships which are yet to be developed in the
system of thought." To the third question its answer is an
emphatic affirmative, for here is the region of its strength ; it
claims that there are two, at least two, marks of thought which
must lift it in our estimation out of the empirical and pragmatic
movement — its teleological character, on the one hand, and
its normative character on the other hand. In these respects,
say the logicist thinkers, thought not only reveals reality but it
is a unique and most distinguished mode of reality in its very
self.1

How then, we may ask, may the pragmatic thinker answer
these three questions? Before taking that up, however, I
think it would be profitable to inquire as to the presuppositions
of a pragmatic theory as such : that is considered as an account
not merely of knowledge but also of reality.

In the restricted sphere of knowledge any genetic or develop-
mental account of thinking necessarily makes the thought
function in some manner utilitarian, instrumental, adaptive.
Thought proceeds by a series of constructions, discoveries,
serviceable adjustments, etc. But unless we admit that the
system is a self-developing one which advances under imminent
rules of its own, which is just the assumption of the ' rational-
ists,' there must be a series of points of origin, stimulation —
and to the onlooker, points of observation — with reference to
which, and by the mediation of which, all the readaptations
and new discoveries are accomplished. Of course this is what

1 It is this claim with reference to thought which has had nothing like
enough attention on the part of the pragmatic logicians, such as Dewey and
his colleagues, in pressing home the refractory and persistent dualism of the
idealistic theory of knowledge (Dewey, Miss Thompson, in Studies in Logic}.
The idealist, e. g., Bosanquet, may reply that there is a fundamental identity of
thought and reality, whereby the development of thought is a mode of reality
which is fully realized only in the system of which thought, the individual's
thought, is a part. For a criticism of the selective theory of thinking, as
an exhaustive account of truth, see Bosanquet's strictures on the present
writer's views in the PSYCHOLOGICAL REVIEW, July, 1903 (following up a dis-
cussion in earlier numbers).

THE LIMITS OF PRAGMATISM. 33

in the biological and sociological sciences is called the environ-
ment. Even from the strictly psychic point of view, even for
reflective thought itself, there is not entire autonomy within the
movement of thinking. So true is this that the determinations
of reality, not alone in the sphere of the external world, but
also in that of the most abstract truth, now most current among
psychologists, differ as to the place of such coefficients as * re-
sistance,' 'stubbornness,' 'limitation of activity,' « experience of
control ' ; but they do not ask whether they are there at all. The
real, the fact, the truth, is in some sort or other, that which is ac-
commodated to, that which must be recognized whatever else is
refused recognition.1 Putting this purely in psychic terms, sensa-
tional or other, we have to say that there are modes of conscious
experience, entering essentially into the determination of truth,
which are not determined entirely by earlier modes of experi-
ence ; and these have their place and value just in virtue of their
character as essentially determining future reality. Mr. Roose-
velt's recognition of the Republic of Panama determines to me
the truth or reality of that State. No one can view the patho-
logical thought systems, with their pseudo-truths and realities,
and refuse to admit that thinking is thus essentially conditioned
upon what is both to the individual and to the onlooker extra-
psychic.

Now what is this something, this environment, for pragma-
tism? Something itself constructed, selected, postulated by
the thought processes? — something itself a part of that sys-
tem of discovered truths and facts which it is the merit of
this view to identify with reality? So the answer may read,

1 It is an interesting instance of Werthurtheile, that pragmatic thinkers in-
stinctively emphasize the manageable, selected, ' workable ' aspects of external
(including logical) reality. But the very question why some thoughts ' work,'
while others do not, throws us back upon the environmental tests.

Professor Dewey seems genetically quite right (loc. cit., p. 76) in making
psychic objectivity an aspect of experience of control, rather than control a
result of objectivity (externality); but before the rise of the dualism of which
objectivity, in this sense, is a term, there is no subjectivity or ' inner ' experi-
ence. If we are to have any explanation of the reason of the determination of
both terms it must be one which does not assume one term — the sphere of
subjectivity, the purely psychic — to account for the other (see the next point
in the text).

34 /• M. BALDWIN.

when it is written : I know of no writer so far who has seriously
proposed this question or who has seemed aware of the vital char-
acters of the presupposition.1 But such an answer would be
open to certain grave objections. First, to maintain that all
there is in the reality called the environment is what has already
been through processes of thought and discovery and established
as true, would mean that future thought processes should be
entirely autonomous — entirely undisturbed by intrusions or
stimulations which could disturb them and require new adjust-
ments. Earlier truth would be in the thought system. Again,
on this supposition, we might trace backwards the path of
knowledge, and when we reached the first glimmer of cognition,
ask what stimulated it, what were the necessary conditions of
this sort of function in consciousness. Surely it could not be
said that its environment is its truth-system. Further we might
ask a similar question of the movement of thought as a whole
in society or in history ; if each step has been a tentative one,
secured by the struggle and adaptation to which the thought
function is ministrant and instrumental, with reference to what
system, to what larger whole, are these adaptations and dis-
coveries made? To deny that there is such a whole — such an
environment progressively adjusted to — would be, it would seem,
to give up the pragmatic method ; for then there would be no
recourse except to the idealistic position that thought is a self-
sufficient and self-developing teleological system, not an instru-
ment to anything.2

Another presupposition appears when we ask from what
point of view pragmatism cites evidence of its truth : it must be

1Mr. Schiller dabbles in it {Personal Idealism, II.), I think to little profit
Professor Moore intimates it in avowing a biological point of view (as does also
Professor Angell, Relations of Psychol. to Philos., p. 12 f.), but just then he
stops ! (Studies in Logic, p. 374.) Possibly the 'fragmentary ' character of the
pragmatic discussions is what we should expect from the truth of this theory of
thought ! Certain of these writers do seem to be urged on by what Royce
calls a ' certain indefinite restlessness.'

2 An alternative would be a form of 'radical empiricism' which actually
accepts the postulate of various sorts of reality external to the individual as
pragmatic development requires them. I suppose such a genetic realism, if it
stopped there, might still by courtesy be called a philosophy ! But on this
see below.

THE LIMITS OF PRAGMATISM. 35

replied — from an objective point of view. The adjustments,
active adaptations, instrumental achievements, consequential
sequences, etc., attendant upon an act of reasoning, for ex-
ample, whereby the pragmatist defines it and validates it, are
such to him, a spectator, not to the thinker himself. The psy-
chic claim of thought is to be simply and only objective, to
terminate in the object which it constructs. Granted the phi-
losopher's * will-to-believe,' even with it the believer claims to
think exclusively on evidence, claims to free his mind from
prepossession, prejudice, and voluntary bias. To be sure, in
many reflective thought-processes, thinking is a conscious in-
strument, a means to a practical end ; but to say that is far from
saying that such pragmatic reference is a mark of thought, or con-
stitutes its validity to the thinker. Such a general criterion can
be claimed only from the psychological, in distinction from the
psychic, point of view. I hold indeed that the instrumental
character of thinking is marked, and that it is a function of
utility in development and evolution : but that is my theory ; I
can not say that I am conscious of such features in my thought.
It is just its own claim that thought is held to standards quite
livorced from the individual's private volitions.

But once admitted that pragmatism takes an objective point
of view of the thought function as a whole, and certain em-
barrassments at once follow. To assume this point of view is
to accept the objective criteria of the whole process of thought :
and that, when we find it in a mc'lange of phenomena of other
sorts and classes — physical, social, inferential — each claim-
ing objective value also by reason of its own set of objective
coefficients. Once admit the validity of such an objective
claim, and all sorts of truths follow, with the sorts of existence
to which they belong : mental existence, physical existence,
ideal existence, etc. In short, we have, when we assume that
we can rely upon objective phenomena which pose as em-
bodying a type of reality and claim to constitute it — we have by
the same act to validate all such types as being what they claim.
It suffices to make this point here ; it comes up again below
more forcefully where the dualistic implications of pragmatism
are brought out. Here it may suffice to have made it clear that
pragmatic philosophy involves this presupposition.

3 6 /. M. BALDWIN.

II.

Coming back now to the definite questions which we pro-
posed to put to the pragmatist we may proceed to consider his
answers to them one by one.

First, the question of acognilme, and what may be called
mixed modes of reality .

In the literature, the development of pragmatic views has
been largely in connection with the determination of cognitive
reality as a system of accepted truths. Apart from the pro-
cedure of inferring from the results of the criticism of opposing
views, the method characteristic of the writers of the Chicago
publications, this has proceeded upon the psychological view
of cognition which makes it a phase in a process of which
action is another phase, the whole being the process of the
treatment of experience for practical purposes? If this be a
valid way of looking at those modes of reality which are con-
stituted by the exercise of the cognitive function, it would seem
possible to pursue the same method in reference to those other
aspects of mental function which also in some manner lay claim
to real reference. If it be true, that is, that the criteria of truth
upon which cognitive reality reposes and which serve as its
coefficients, have their significance as being points of advantage
in the life of active adaptation, the same may be true of the

1 Such a view is explicitly carried out in ray own development of the re-
lation of thought and action in Social and Ethical Interpretation, Chap. III.,
VII. and Mental Development, Chap. XI. My treatment is to the last degree
pragmatic (jcf. the remarks of Caldwell, Amer.Jour. of Sociology, Sept., 1899),
but being stated in terms of reaction it is objective in its point of view. Pro-
fessor Dewey seems to prefer to speak in psychic or experiential terms, but
gives us no account of the actual psychophysical factors entering into the con-
crete determinations of thought (such as kinsesthetic sensations, images, etc.)
and thus, to my mind, loses the advantage of such psychophysical explanations
as those given by the 'action,' 'synergy,' and other theories. Another result is
his failure to work out the selective tests and checks upon thinking which are
fully discovered only from an explicitly objective point of view. I have de-
cribed the function as a self-repeating ' circular ' (imitative) reaction, illustrat-
ing ' habit ' and issuing in ' accommodation ' ; but I am quite ready to say, with
the psychologist Dewey, that there is a conscious organization resolved into
strains and tensions and reorganized in a new thought under ' stimulation '
(but can not this term 'stimulation ' be avoided? — it is shockingly objective !),
or with the philosopher Bosanquet, ' that the thoughts which precede and fol-
low, taken together, really illustrate ' identity in difference ' !

THE LIMITS OF PRAGMATISM. 37

coefficients of reality of other sorts — external, aesthetic, ethical,
ideal, etc. It would be our task, as genetic psychologists, to
work out the processes by which, in the active life, such and
such marks of experience serve the life of adaptation in the
sphere of reality respectively which this or that class of ex-
perience postulates. The objects of the external world, thus
construed, are the experiences which mediate organic adjust-
ments ; aesthetic realities, those which mediate emotional ad-
justments ; ethical, personal adjustments, etc. This would
seem to be a legitimate and fruitful task — a larger problem of
the genetic logic of reality — and one in whose accomplishment
the pragmatist has a distinct advantage over the rationalist, by
virtue of his doctrine of the concrete experiential determination
of reality of whatever sort. So far, I think, the pragmatic
method has great value, inasmuch as it recognizes the protests
of heart and will — and anything else that can make good its
claim — against an exclusively intellectualist theory of reality.1
But this is so far psychology, and, in a sense, logic. Can
we stop there, and simply recognize the pluralistic real postu-
lates of practical life? One of the fathers of pragmatism, Pro-
fessor James, seems content to do so. Others again seem to
have each his favorite among these different modes of reality.
Some abhor pluralism, but rest content in dualism. Some say
there are a lot of reals, but they are all ethical — thus deserting
the pragmatic point of view. The question for the pragmatist
here would seem to be : Is there any practical need for further
adjustment of these realities to one another? — any issue of life
which is subserved by the dominance of one mode of reality
over the others? — or is there a further real construction which
is realized by the subsumption of these varied realities ? In con-
nection with such an attempt at reduction the question as to the
motive to it would arise. Is it a practical one — that is, is the
demand for philosophy pragmatic and not purely theoretical ?
If so then are not the scales in any such weighing of claims
loaded from the start with various personal, utilitarian and
other practical preferences of individuals? — all of which are

1 Cf. Dewey, Studies in Logic, p. 432, and the papers of Stuart in the same
volume.

38 /, M. BALDWIN.

to this type of thought in themselves legitimate — or with social
and conventional, types of belief, judgments of value, etc.,
which would hopelessly prevent any general unanimity or per-
manence of world-view? Possibly, then, we should have to ap-
peal to the statistical logician, who deals with purely theoretical
matter, after all, and asks him to establish a •pragmatic equation
of error whereby we might correct up the conclusions of those
philosophers who ventured to think beyond the rankest pluralism !

But there is more than that in this question of the modes of
reality. The objective standpoint itself issues in a dualism
more sweeping than the various dualisms of the objective reality
coefficients themselves. The prime and fundamental dualism
is the old subjective-objective distinction. It is true it is now
becoming, to present-day thought, a dualism for cognition, that
is, a dualism between two objective constructions — the me and
the not-me. Genetic and social psychology are making the
' me ' the same as the ' other-me ' of the social fellow, and both
a part of the system of constructions determined as objective.
But over and above this essentially objective dualism of ' me
and not-me ' there is, together with it and necessary to it, the
dualism of the subjective and the external, of the * inner ' sphere
of experience itself and, over against it, the ' trans-subjective.'
This latter dualism would seem to be genetically earlier, at
least earlier than the full form of that which embodies the
distinction of mind and body ; yet the self-not-self distinc-
tion, in the form in which it precedes the inner-outer distinc-
tion is largely one of organic or strictly presentative data. The
dualism — to return to the last positive statement — which re-
mains for reflection is that of ' inner ' and ' outer ' ; it opposes
experience-as -such, with its constructive principles of unity, con-
tinuity, identity, etc., to outer-reality-as-such. Here are two
spheres, one of cognized realities including selves as objects, the
other of mental reality which functions as agent of the construc-
tions of cognized reality. Now what is pragmatism able to do
with these great dualisms?

As to the me-not-me distinction there is no difficulty in secur-
ing full pragmatic justification for it. It has been a labor of
the present writer in recent j^ears to attempt to work out just

THE LIMITS OF PRAGMATISM. 39

that problem ; the functional and active adjustments, principally
social, under stress of which the me and the not-me are first
distinguished in thought and furthered to maturity. If a person
is to be a person, he must know that he is a person, and that
others are persons ; also that there are things which are not
persons. So we may say : so much to the credit of a prag-
matic method in psychology, so far as its attempt bears fruit.

But there is the other, the inner-outer dualism. This puts to
pragmatism the different question : what is the practical origin
and value of this distinction? And upon this the further ques-
tion : is it possible to give a pragmatic account of this distinc-
tion without assuming it? This second question is put as it is
in order to suggest the point of criticism now to be made — a
criticism which, I think, lays bare one of the essential limita-
tions of the pragmatic method.

The question raised in attempting to account for the inner-
outer distinction is really that of a genetic account of reflection :
for this dualism cuts through consciousness just where the spon-
taneous dualism of me-not-me passes over into a way of treat-
ing data of both these classes which recognizes the possibility
that that which is objective may also be and perhaps only be
'inner' or peculiar to the individual's constructive processes,
and not trans-subjective at all. The * me ' is objective but not
external. Here the representative theory of knowledge at
once springs up to interpret this na'ive or common-sense reflec-
tion. The treatment of illusion, from the dictum that « the
senses deceive ' of the Heraclitians and the Greek Sceptics, down
to the elaborate theories of ideas and * imagination ' in modern
times, witnesses to the real problem raised here : a world of
' inner' events and objects, ideas, images, etc., which may or
may not be external or * outer ' also.

The ordinary idealistic solution is simply to take conscious-
ness at its word, and concede that the representative series does
not actually embody reality, but only copies or duplicates it.
Then the problem is not so much to account for error in some
representations as to account for truth in others : how can we be
sure that reality really is in any case and anywhere what we in
our * truths ' conceive. On the other hand, the pragmatist has

40 J. M. BALDWIN.

on his hands the problem of error : if reality is that which the
cognitive function normally reports, with no further reference
to an outside system or series, how then can such a distinction
ever arise as that between what is somehow referred to an ex-
ternal system, and what, though fully cognitive, is nevertheless
only the product of the thinker's mind, while possibly to the
thinker still making an erroneous claim to external value.1 I
see no great difficulty here, to the extent, at least, of finding
pragmatic justification for the distinction of inner and outer,
and with it of truth and error — provided we assume a thorough-
going objective point of view. The pragmatist indeed must
admit that such a point of view is, as we have found it to be,
his necessary presupposition. The play of images necessitates
mistakes and error ; certainly so, if the imaging function is the
vehicle of tentative, trial-and-error processes working in the
interest of practical accommodation to an environment, physi-
cal, social, and other. So far so good : we have a genetic sci-
ence, genetic psychology, depicting a development or evolution
process in which truth and error are correlative aspects.

But in recognizing this necessity of the objective standpoint,
we bring up again, and with increasing embarrassment, the
problem of the environment. It is possible, as we saw above,
to look upon the environment itself as a mode of pragmatically
determined reality, but only on the assumption of the reality of
mental function and development. But now we find that such
recognition of mental function as a mode of real process is
possible only in a dualism with the external — the dualism of
inner and outer — of mind and that external reality or environ-
ment which we are to explain as a construction of the inner
reality or mind. There is here a vicious circle. The only
point of view (that is, the inner) from which the pragmatist can
possibly explain the external environment is one from which
the very principle of his explanation, a mental development by
which the external is constituted as real, is quite destroyed.

1 The difficulties of the representative theories are brought out with especial
force by certain of the Chicago pragmatist writers : notably by Moore in Exis-
tence, Meaning, and Reality and by Dewey in Studies in Logic. But their
treatment of the problem of error and illusion is correspondingly halting and
inadequate, except so far as error is made to follow from a psychological
process which is experimental (see, however, below).

THE LIMITS OF PRAGMATISM. 41

It is a case of what I would fain call the genetic fallacy . In
a recent paper l there are worked out the stages, called there
* progressions,' successively reached by consciousness in the
development of the me-not-me dualism. And reason is there
found for saying that a theory which would be consistent must
not read into connection with a term on one side of this dualism
at a given progression, a term from another stage or progres-
sion ; the great case of it being, as there pointed out, the treat-
ment of mind as real and body as subjective, when the very
progression in which mind is found as real guarantees mind
only in a dualism with real body. This is the ' genetic fallacy'
— confusing the terms of different genetic progressions.2

We have here a case of it. The pragmatist must be a monist
if he would explain the environment ; but he can not account
for error — nay, he can not be a pragmatist — without being a
dualist. He takes advantage of dualism to refute dualism.

It is easy to see the necessity of insisting upon this require-
ment in comparative and child psychology, where it forbids us
to construe the individual's mental constructions apart from the
sort of function then and there possible ; as in expecting any
sort of moral insight in an individual which is not yet self-con-
scious. But it is harder to keep clear of the fallacy when we
are theorizing about the mind. We fall to using the outcome
of a later stage of reflection, which has its validity only as pro-
ceeding from an earlier, to criticise and invalidate the postu-
lates or requirements of the earlier. Reflective idealism is pos-
sible only by refining one term of a crude dualism ; it is then a
logical circle to show that this refined self can not be a term in
such a dualism. The proper method is to ask what becomes of
both terms of the dualism in the process of further mental devel-
opment. Pragmatism itself insists upon this in its protest against
the extreme logicism which reads its developed ' universals '
back into the crudest alogical states of mind.

The difficulty arising from this dualism is undoubtedly em-
phasized when we consider the question of time — a question

1 PSYCHOI,. REV., May, 1903.

2 This general requirement of genetic consistencies is recognized in his com-
ments on my paper by Professor Mead ( The Definition of the Psychical}. I
could not wish a stronger enforcement of it than that of Dewey, loc. cit.t pp. 16 f.

42 /. M. BALDWIN.

which the idealist is not slow in urging upon the pragmatist.
Time considered as a mode of pragmatically derived reality
must be considered real in the sense that other abstract or con-
ceptual modes are. The pragmatist has, from such a theory of
time, a weapon of advantage in one respect, inasmuch as he
can thus deny the reality of future, undiscovered, possibilities
of existence, considering them as projections from realities
already discovered and justified. But now the critic may say :
you admit the reality of mind as a principle having a develop-
ment in a time series, and recognize it as having constituted for
itself a series of corresponding realities at its several stages of
development ; the reality of a time series considered as a mode
of development of mental reality must be acknowledged, and
with it the reality of time as a mode of other realities so con-
stituted. The time mode is, in short, a mode common to the
self postulated and to the realities in dualism with which the
self all along knows itself. The independent time develop-
ment of reality of whatever sort, therefore, must be admitted if
the time development of the self be assumed. It is difficult to
understand, indeed, why the reality of the future time-mode
should not also be granted if we grant that of the past time-mode.

The advocate of some form of logical construction of reality
avoids this dilemma by making time a mere cognitive mode —
not real. But that has all the difficulties of a logical deduction
of time, and also labors under the acute embarrassment of
denying the progressive or dynamic character of reality. Here
again the only solution possible to the genetic and pragmatic
method would be one which actually depicted consciousness
passing out of the dualism under which it is compelled to think
both mental and external reality as in time.1

Writers of a pragmatic bent delight to press home the
dilemma of dualism upon the representative theory of knowl-
edge ; and their indictment is, in psychology, unanswerable.
And it is amusing how they instinctively fight shy of taking up

1 An analogous question arises regarding space, except that it is a mode of
external reality only ; but pragmatic dualism would require us to recognize real
space in the sense that physical existence in general is real. Only a mode of
experience which not only satisfies but transcends spacial properties would be
sufficient for a general theory of reality.

THE LIMITS OF PJtAGMATISAf. 43

their own dualism. Yet withal they give signs of a certain ap-
prehension of it. Notably is this the case in the able papers of
Professor Moore. He concludes his examination of Professor
Royce (Studies in Logic, XI.) by asking the question as to the
tests or checks upon the action-knowledge process which the
environment brings to bear ; and makes the entire external and
organic world as such environmental to the psychic function.1
But such a dualism is just the last refinement of reflection in the
distinction of self and not-self : it is the inner-outer antithesis
which consciousness reaches only in its most extreme personal
determinations. And it is this dualism of the reflective con-
sciousness which is one with that of the philosophical theorist.
Whether we consider the terms of the dualism as present to the
actor or only to the philosopher, still the 'inner* and the
' outer' are in both cases alike cognitive constructions, and each
is valid by the same right that the other is, and exactly to the
same degree. If, however, we put the question to Professor
Moore with reference to a stage of development at which the
psychic dualism is not yet come, we must then take the objec-
tive point of view and treat that consciousness as to us — not to
it — having such and such an environment. It has no ends,

1 Yet he calls it all 'habit,' thus somewhat surreptitiously bringing the ex-
ternal world into the sphere of that which is already adapted to, neglecting that
aspect of it which makes it that which is still to be adapted to. This is, of course,
an easy way of 'begging ' the pragmatist's dualism. In another place he ex-
plicitly speaks of ' habit somehow developing its own interruptions ' in order to
give thought a chance (Existence, Meaning, and Reality, p. 16). Remarking
upon Professor Moore's treatment of my ' extra-organic ' test (Development and
Evolution, pp. 250-1 ) under that of the ' effectiveness of an idea to organize habit'
(ibid.), I should say that it is just the intrusion of something that breaks up
habit, and hence cannot itself be described as habit, that makes the reorgani-
zation by the idea necessary and possible. For example, I am in the habit of
walking comfortably in a shoe with low heel ; and when a new shoe with a
higher heel strains my tendo achillis (as it just has !) my new idea of walking
(on my toe !) arises and is effective as an accommodation to this new, in no
sense habitual, experience of the high heel. In other words my old truthful
idea of walking based on habit has broken down under the test of working in
the extra-organic environment which includes the new shoe. This I have de-
scribed (loc. cit.) as the 'test of fact,' supplementary to the 'test of habit.' So
when Professor Moore says, quoting me, that a reorganization under stress of
experience is entirely in " terms of 'activities already going on' " (Existence,
Meaning, and Reality, p. 18) he omits the new sensation experiences whose
urgency compels the reconstruction.

44 «/• M. BALDWIN.

no self-determinations ; it makes no postulates ; it simply acts
straight-away as its stimulations bring it out. And any fruitful
constructions of its 'purpose,' 'plan of action,' etc., are read
into it by us. Much of the plausibility of the pragmatic
philosophy comes from this playing between the two points of
view.

In short — to sum up — the pragmatist must either frankly
swallow the camel of a real environment which the knowledge
function may then both truthfully and also erroneously reflect —
a step which would involve him in all the epistemological litiga-
tion of the representative theories of knowledge — or he must
find some guarantee for the reality of the mental principle which
is not rein pragmatisch. This latter is his better course ; the
present writer adopts it as a limitation on his pragmatism. As
psychological and logical method, the instrumental point of
view is true and inevitable in present day evolution thinking ;
but there must be found a way to preserve it without expanding
it into a one-sided metaphysics which then eats it up !

III.

The second general question proposed to advocates of prag-
matism in philosophy is this: Are they undiscovered realities?
What meaning can such a conception have for the pragmatist?

In order to give an adequate answer to this question, a detailed
theory of the different modes of cognitive and logical process
would have to be worked out. So far as I know, no one has
yet attempted it from the pragmatic point of view. Especially
would it be necessary to have an interpretation of the general,
and universal, and normative aspects of experience, with
accounts of the claims made by these modes, respectively, to
report or embody reality. It would then be seen how success-
fully their claim to exhaustiveness was in truth made out. So
far as these thinkers have gone they seem to be aware that if
reality is to be consistently interpreted as a construction of
experience, then there could be no realities which are not so
made up in experience, in the way we call discovery. We are
explicitly told that reality is actually made when it is discovered ;
that its development is, or proceeds -pari ^>assu with, the system

THE LIMITS OF PRAGMATISAf. 45

of truths which is in process of progressive establishment. Of
course in this matter the pragmatist is to be allowed all the pos-
sible reaches of cognition, and of other modes of experience,
such as that found in social relationships, whereby all possible
forms of apprehension are brought into play.

It seems to me, speaking tentatively, and subject to confirma-
tion from the detailed study of the demands made by logical
process generally,1 that pragmatism is able to give a fairly con-
sistent account of itself in this matter up to a certain point ; a
point which again marks the transition from psychology to
metaphysics. The aspects of reality very clearly distinguish-
able, and with them the habits of mind which yield these as-
pects of reality, in complementary relation to each other and
seeming when taken together to be exhaustive, are those called
variously * fact and value,' « appreciation and description,' 'sci-
ence and real life,' ' habit and accommodation,' ' prospective
and retrospective reference,' and in the writings of pragmatic
thinkers, though not so clearly expressing the same distinction,
« fact (or image) and meaning.' If we look at the distinction
from the point of view of the psychic processes and attitudes
involved, we may place on one side all that which is cognitively
or actively apprehended, as the given, the established, the-now-
and-here-existing ; and, on the other side, those aspects of our
mental determinations of all sorts which qualify the given or
aim to establish it in any indefinite, hypothetical, or prospective
way. Thus having divided the entire real predicate into that
which to consciousness already real, and that which merely may
be, seems desirable is not yet realized, we have to ask as to
the pragmatic meaning and value of the latter member of this
antithesis.

So far as pragmatic derivation and justification is concerned,
they are not far to seek ; the utility of hypothetical and normative
motives to action is admitted on all sides. The reference to reality
is, from the point of view of the pragmatic onlooker, that of a
demand for progressive organization of the content already

1 An attempt made in the volume on ' Genetic Logic ' which is to be pub-
lished in the near future, and of which certain of the developments are antici-
pated in this paper.

46 /. M. BALDWIN.

given and still subject to organization. Two questions arise,
however, so soon as we press the problem of objective status of
such hypothetical predications.

First, we have to ask : does the psychic movement postulate
more than the reality already given in the datum? In reply to
this it seems to be true that it does not. The real subject in any
judgment of value is only that which just by being constituted
as possibly real, already exists for action. So far as deter-
mined for action — that is, pragmatically — such judgments are
true. The real predicates involved in the ideal and normative,
no less than in the logically hypothetical states of mind, are
constructed from the data of existence then and there at hand.
So far we may go with a thoroughly pragmatic psychology.

The will to believe, for example, is effective, or enters into
the determination of reality, only in so far as the belief postulates
the result as already actually accomplished : the existence of
the thing believed enters into the psychosis and determines the
act of will. From the psychic point of view the will to believe
is as much determined by real existence as is any other sort of
intentional act (e. g.> that of moving my head to escape a blow) ;
for psychically the reality or existence is taken for true in the
act of belief. It is only to the onlooker that some new mode of
existence is determined by the action in accordance with the
belief. And even that is, to the onlooker, a mode of real deter-
mination only after it has happened. From neither point of
view does it involve the postulation of unrealized reality.

How is it then that we, in real life, assume such a sphere
of the undiscovered? Is not all ignorance failure to know
what there is to know? True: but so far as this is reality
at all, it is known reality in its warp and woof. The mental
attitude here is essentially the same as that by which we depend
upon the continued existence of what we have once known.
Both are re-readings of the established order under familiar
categories, and both are — when all is said in individual and in
social psychology — through and through pragmatic; that is,
both, determined by necessities and utilities of real life, are to
the same extent that the originally acquired items of reality
were.

THE LIMITS OF PRAGMATISM.

47

But, all is not yet said. If we ask, second, for the relation be-
tween action and thought in the determination of the joint func-
tion which issues in reality, we find a consistent and attractive
doctrine, as has already been said: thought — cognitive product
in general — is a reflection of habits of action, an organization
for future safe action and practical handling of experience. It
is the static term in an active process. So far as static, its
reference is trans-subjective ; it is reality. So far as a mental
organization of habits, it is dynamic, a ' meaning,' a * plan of
action,' a * purpose,' an ' instrument,' an ' end ' — to use current
designations of different writers, all of whom aim broadly at ex-
pressing about the same thing.1 The thought term is the term
which is hypostatized as real for the self and for others. It
stands as valid in experience until more effective handling of
experience issues in a modified thought.

On this view the psychological and biological utility of
thought or cognition in general resides precisely in its static
quality. It is a definition, a specification, a determination.
If it is a ' purpose,' it is a defined, usable, purpose ; if it is an
end, it is a presented, communicable, end ; if it is a meaning,
it is because it fits into a context of available meanings ; if it
is a ' plan of action,' it is as much a ' plan' as it is an « action.'
In minimizing the static aspect and reference of thought, prag-
matic writer are depriving it of just those features by which its
usable and effective character is to be established in mental
development. Where would language be, and with it all the
socially derived determinations of action, without this assump-
tion. I am sure there will be a reaction on this point from the
extreme view whose only justification is novelty.2

If this be allowed, we find it necessary to ask for each

1 My own formulation, arrived at from a detailed exploration of the factors,
individual, social, and other and stated from the objective point of view is :
' what we do is a function of what we think, and what we shall think is a func-
tion of what we have done ' {Social Interpretations, pp. 106, 301).

1 An extreme view which, in its emphasis on action, practically issues in a
dualism of thought and action, is that of Mr. Schiller (Humanism, and
'Axioms as Postulates' in Personal Idealism). The 'Instrumental Logic ' of
the Chicago writers avoids this extreme (who for this reason, as I am informed
by one of them, refuse to use the term pragmatism for their point of view); but
it is still open to the limitation developed in the text.

48 J. M. BALDWIN.

thought determination not only what sort of action is served,
what dynamic pragmatic meaning it has, but also what static,
theoretical system of realities it finds its place in. Every true
thought is true not only because it has active determinants but
also because it reflects all the meaning for life which those active
determinants have. So, to reverse the proposition, we may say
that every successful active adaptation or line of conduct must,
in its development, reflect itself in cognitions or thoughts which
are the reflection of all its meaning. It thus becomes definite,
socially available, and more than concrete. The categories of
general, social, and communicable thought are the normal
vehicle and embodiment of the utilities which are turned to
account in development.

From this we see that a pragmatism which denies or dis-
counts the validity of the logical point of view truncates its own
assumed psychological process and becomes helpless. Thought
becomes a by-product, a second-hand way of reaching reality,
which is * suspect ' to those other and more valid intuitions
given in feeling and action. And this becomes glaringly evident
when such a view blows itself up into the dimensions of a phi-
losophy. A philosophy is an attempt to think things : to reach
a general and communicable theory of reality. Its characteristic
feature is just its generality as opposed to concrete practical
enterprises which, no matter how valuable, are still inarticulate.
So the mere assertion that in its origin thought is a mode of
action, and its revelations are possible because of its origin as
serving the utilities of real life — even so much of a general
statement as this is itself a logical reinterpretation of the bare
reactions which it claims to interpret and generalize. To prove
its own truth, indeed, pragmatism can not be content with its
own formulation : for such a principle in action must itself issue
in a theoretical or logical system.

We might indeed stop here ; having a dualism of pragmatic
and logical explanations, the logical being the naive system of
thought reflecting the adaptations of which pragmatism takes
note. But another alternative would be to reach an interpre-
tation which should reconcile the two essential phases of the
action-thought process, and itself issue in the solution of the
contrasts in our experience.

THE LIMITS OF PRAGMATISM. 49

There is still lacking, I think, an attempt of the last-named
sort ; but in Dr. Peirce's and Professor James* formula for prag-
matism, we find something of an attempt at the first named —
an attempt to state logical meaning consistently with pragmatic
origin. James' formula is: " The whole meaning of a con-
ception expresses itself in practical consequences, consequences
either in the shape of conduct to be recommended or in that of
experience to be expected if the conception be true." l This
would seem to be — disclaiming, however, the attribution of
what follows to Professor James — a formula of the sort of log-
ical systematization of meanings in which pragmatically deter-
mined thinking would be reflected. The meanings intended and
accepted would all the while be subject to the selective, correc-
tive, substitutive and other revisory processes of practical life,
and the realities reflected in such a system of truths would be
the stable system of meanings thus produced. The limit of the
meaning to be preserved in any case would be, I suppose, its
lack of inhibitory or ' interfering ' effects, its negative fitness as
not producing confusion, * either in action to be recommended
or in experience to be expected ' ; and the tests, final and con-
clusive, of logical meaning, would be such concrete prac-
tical fitness. This seems to me to be quite consistent, and, for

1 In the writer's Dictionary of Philosophy, ' Pragmatism' (there also Peirce's
formulation and criticism). I know of no other attempt to formulate a con-
structive principle of logical meaning on the pragmatic basis. Professor Dcwey
aims, it would seem, at justifying the logical processes from the point of view of
genesis, rather than at treating of the structure or morphology of thought. And
yet by defining thought explicitly in instrumental terms (Studies in Logic,
I ff, 40, 76 ff, etc.) he really excludes the so-called analytic and deductive op-
erations. James' formula would also seem to be limited to those ' conceptions '
which no have some sort of practical consequences — a matter returned to be-
low. Cf. James' address, Philosophical Conceptions and Practical Results.

A concise statement of the requirements of deductive and ratiocinative
thinking, from the pragmatic point of view, is to be found in Professor Angell's
Relations of Psychol. to Philos. (Univ. of Chicago Publications), p. n. I
do not deny his contention that experience is a " universe or system, in which
truth is ultimately synonymous with the effective "; but this can not be used
to deny the competency of the logical point of view within the system and the
need of imminent logical criteria. The whole tendency of this way of thinking
is to deny the validity of 'cross-section ' or morphological principles, in favor of
4 longitudinal section ' or genetic principles — to use a figure whose meaning is
enforced by the writers now under discussion. (Cf. Dewey, loc. cit., p. 17.)

50 /. M. BALDWIN.

the sort of truths it really explains, to explain them ! It is
' radical empiricism' in both its members, pragmatism of origin
and pragmatism of meaning. The tests of utility are simply
converted into criteria of logical meaning and value.

The questions, however, which it excites are those involved
in certain of the idealist's most pregnant positions : those which
assert the essentially teleological and universal character of
thought. The criticisms 1 brought from such a point of view
deserve more adequate notice and refutation than that which
calls all universal and analytic judgments * dead,' and < no judg-
ment at all.'2 Not to account for such judgments is to fail to
account for all deductive, mathematical, and subsumptive rea-
soning— or, indeed, to call it logically abortive and tautolog-
ical ! Instrumental or pragmatic logic must take up this problem
with all its resources : the problem of the structural principles
of thought, which are not in any evident way in their origin
connected with experience at all. The theory of variations,
with natural selection, bravely stated in James' « back-door
process ' chapter, and repeated more feebly in Schiller's ' axioms
as postulates,' goes a long way from the objective biological
point of view ; but that completely deserts the processes of
knowledge, throws the epistemologist back upon native princi-
ples implicit in concrete thinking, and so sets a direct limit upon

'See Bosanquet's concise statement of points in his criticism of the present
writer's theory of selective thinking (a theory directly in the line of the position
required by a pragmatic logic), in the PSYCHOLOGICAL REVIEW, July, 1903,
already referred to.

2 Miss Thompson, in Dewey's Studies in Logic, p. 108. It is only on
such a view that inference can be made a wider term than judgment (ibid., p.
117). 'A judgment is an inference which is conscious of its ground ' to this
view ; but if inference be a process of analysis or composition in higher re-
flection, involving a setting together of the elements of a thought-situation
under certain rules of logical grounding, then it must be separately accounted
for and not smuggled in as unconscious and prelogical. Inference is, or may
be, preliminary to judgment Justin so far as a logical thought-system is presup-
posed. Many judgments are thus determined. But to deny that they are
judgments (as these writers do) makes it necessary to deny also that there is
anything in inference not already in the preliminaries of the act of judgment.
The real question is : can a genetic process of determination that of psycholog-
ical conditioning found in the unreflective stage of mental development, be
substituted for the logical determination, that of inference, found in the reflec-
tive stage ?

THE LIMITS OF PRAGMATISM. 51

pragmatism as a working theory even in the individual. We
have — that is, he has — to work out individual pragmatism and
then, to explain its limitations, assert pragmatism in a racial
sense from an objective point of view ; to explain the environ-
ment he must, as we saw above, take a subjective monistic
point of view, which again subverts the pragmatic theory of
reality with which he started out, by involving him in the ge-
netic fallacy.

It is quite possible that a more or less successful defense
against these criticisms is to be found in the line of a theory
maintaining the social constitution of knowledge, with social
embodiment and social transmission. I myself have found it
necessary to hold that a strain of universality and generality is
imparted to knowledge in the aspect which constitutes it ' public'
to a social group. The utility upon which pragmatism may
insist, therefore, as socially practical and concrete, may be,
from the individual point of view, general and universal. A
social practical value might be reflected in an individual's theo-
retical value. The psychological point of view might then be
conserved in the continuity of social thinking, although lost in
the biological constitution of the individual. Thus, as I have
intimated elsewhere,1 there might be a social derivation of the
categories. I have great faith in future work along this line.
But withal the limitation remains that such a theory would give
a logic of a stage of cognitive process — that at which pragmatic
tests are transferred to the social group — rather than a philos-
ophy of the entire movement of reality. We should then have
the formula that the individual's and society's common system

1 Social Interpretations, Chap. III., where it is shown that the judgment of
the individual, though privately competent, is the reflex, through organic and
natural selection, of social life. Such a theory would repay working out in all
the departments where the individual's norms of value seem independent of
all experience — in ethics and aesthetics, as well as in logic. It promises a
theory of the origin of the categories which would go far to reconcile the claims
of a priorism (making it individual) and empiricism (making it social). It is,
indeed, in this field that the battle has been joined by utilitarianism and
hedonism in their substitution of 'general ' happiness and utility for individual.
It is not sufficiently understood that these schools have already worked the
pragmatic hypothesis in their fields in ways which should instruct the novitiate
in pragmatism in logic. Mr. Spencer's attempt at a physiological theory of the
a priori should also be remembered with respect.

52 /. M. BALDWIN.

of logical meanings would be tested by private consent and
social consequences jointly.1

It would remain, therefore, to take up the other alternative
mentioned above and try to realize an actual reconciliation of
the pragmatic and logical points of view in a synthesis in which
they are equally essential members. This can not be entered
upon here and now ; it is the same need that we found under
the earlier head where it was a question of finally escaping the
dualism of inner and outer, with its necessary implication of an
extra-mental order of realities. I think, however, that it is
possible, as intimated at the end of this paper, and that it will
take full account of the social aspects of logical determination.

IV.

It remains to take up the third of our general questions : that
of the exhaustiveness, as measure of reality, of the modes of
apprehension based upon empirical marks and coefficients. Are
there modes or types of reality reached in experience for which
there practical criteria are not sufficient?

It has been intimated, under the last heading, that a genetic
account of the rise of all the modes of thinking — general, uni-
versal, normative, no less than particular and concrete — may
be worked out successfully from the pragmatic standpoint ; it
must be done if evolution is to be a general theory.2 But the

1 A state of things analogous to the establishing of truth as to the external
world by joint tests of the individual's habit and the external environment, as
we saw above.

2 The line of least resistance to the writer, worked out in earlier publica-
tions, leads to the view that the general aspects of our apprehension are sup-
plied by our general habits of treating things and hence are motor in their seat ;
motor habit is thus a means of reducing and grouping the embarrassing details
of sensory stimulation. M. Havard (Revue de Metaph. et de Morale, 1896, pp.
670 ff. ) discusses this view as le nouveau nominalisme ; and if no further logical
account than this can be given of ' generals ' his description and criticism would
seem to hold. This is one of the points on which I do not find myself able to
follow Professor Angell in his identification of logic with functional psychology
— though disposed, indeed, to claim as much of 'the earth' as possible for
psychology (Angell, The Relations of Psychology to Philosophy, University of
Chicago publications). The authors of Studies in Logic seem to adopt the
view (see pp. 113, 176, 198 ; and Moore, Existence, Meaning, and Reality, p. 24,
quoting Mental Development, pp. 323 ff.) that such a psychophysical account
of the rise of the general meaning attaching to concepts is sufficient without
more ado as a logic of ' general ' validity.

THE LIMITS OF PR A GMA TISM. 53

quite different question arises as to the meaning of certain of
these modes ot construing reality, together with the tests or
criteria for their valid and successful application. The ' gen-
eral' concept, for example, pretends to be valid as a vehicle of
real apprehension of the world ; but it would be impossible to
make a conclusive test even of the workableness of such a con-
cept by an appeal to a concrete or practical instance. In the
absence of further standards it would be impossible to separate
concrete consequences or other pragmatic marks from general
consequences. On the contrary, it is just the meaning of a
' general ' mode of thought that it stands for the particular case in
the sense of organizing it with other experiences. It introduces
organization, relationship, and systematization into experience
just to the dropping off of those aspects which are individual.
This is the sort of reality which such a concept claims to reflect ;
and its claim could be tested only by some principle which
could span the system in which the particular case in ques-
tion is organized. No appeal to a concrete situation can vali-
date an aspect of reality which is tpso facto a systematization
of various such situations or cases.1 There must be, therefore,
if such thinking is to have any control or positive validation,
certain principles of organization of logically apprehended
reality as such. This would throw us back upon the traditional
' laws of thought,' I suppose, or some analogous self-applying
criteria of sound thinking.

This may be a way of saying, with many modern logicians,
that only particular, not universal, judgments carry the affirma-
tion of reality ; if we limit ourselves to pragmatic tests, available
only in concrete experience, I see no way of avoiding such a
view. But such a position, it seems to me, allies pragmatism to
extreme nominalism, and it stands or falls with that as logical

1 Though from a logical point of view it might invalidate it. An applica-
tion of this is at hand in the genetic account of the development of self and
the dualism of self and not-self. These concepts are essentially general, and
no single act of a single self, no matter what its consequences, could validate
them as modes of reality, though by requiring a new psychological reconstruc-
tion of the material their logical meaning might be invalidated. It is inter-
esting to note that the word ' general ' does not occur in the rather full Index
to the Chicago Studies in Logic.

54 /• M. BALDWIN.

doctrine. Moreover, it is a pragmatic psychology which most of
all feels the need of some justification for universal truths and
judgments, for their ' utility ' is in some way to be reflected into
reality, if reality is to be no more nor less than the system of
judged truths . The difficulty arises, indeed, only when pragma-
tism aims to be a universal logic, and so essays the impossible.1
The case seems stronger still for the so-called normative or
ideal aspects of experience. In their origin these are functions
of the progressive organization of experience under pragmatic
rules : they illustrate the * prospective reference ' of thought to
the unfulfilled career and possibilities of reality. This we may
concede and defend. But we do not find — it is a contradiction
to conceive of finding — test cases, practical situations, which
exhaust the meaning or establish the validity of these modes of
reference in futuro atque in eternitate. How can we estimate
the practical consequences of ideal virtue, whereby there would
issue forth the * highest good ' ! * How can practical life ad-
equately test the validity of modes which essentially claim to
transcend the experiences of real life? If the normative modes
of apprehension or thought are of pragmatic origin, then it is
just the pragmatist himself who must give them validity as in-
terpreters of real aspects of things and events ; and he is the

1 As to the position itself regarding universals, I think, it survives simply
because nominalistic logic has not yet fully yielded to pragmatic psychology
(which really requires instrumental and real logic). Psychologically the
universal judgment refers to reality exactly as the particular does, i, e., by
the reference to the sort of real universe in which both judgments are made.
It is said that particular judgments are experiential ; but then universal judg-
ments are never made except as holding true in some experience. Certainly
pragmatism can not distinguish universal judgments from particular by that
mark. The reader may see in Professor Royce's treatment of the two sorts of
judgment with reference to their implication of reality ('outer meaning' of
ideas) an attempt to make the real reference of universal judgments to reality
negative only ( World and Individual, I., Sect. VII.). This is a logical way of
saying that a particular case may invalidate a general (which must be expressed
in a universal judgment) but may not suffice to validate it.

2 As intimated above utilitarian and hedonistic theories in ethics do attempt
something of this sort ; but they depart from their pragmatic formula in so far
as they seek a logical justification of their conception of the highest good rather
than an actual concrete experience, or situation, which would realize it. In
either case the individual has no practical test of conduct which is adequate to
establish its general or universal utility.

THB LIMITS OF PRAGMATISM. 55

last person to be able, when the practical criteria break down,
to throw over these categories and resort to subjectivism or pure
nominalism. They have utility, as he says, as ways of inter-
preting experience ; but they issue just by this interpretation in
modes of reality. The outcome is that these modes of thought
must carry in their exercise their own means of validating their
claim to organize experience essentially beyond its actual reali-
zation.

An interesting turn may be given to this point by asking for
the interpretation of the normative in logical values as such :
what is ideal or perfect thinking, and what makes it ideal ? Evi-
dently it must be flawless thinking, as tested by adequate rules
and criteria. The pragmatist would have to say that all think-
ing which fulfils the demand that it deal successfully with the
concrete situation which stimulates it is in so far flawless : that
is, that all thinking is flawless which does not issue in practical
embarrassment and confusion. It can not say, in the particular
case, that thought might have done better; for — to point again
a foregoing criticism — there is for this view no way to test rel-
ative or alternative solutions : such a test would involve the
application of general criteria of validity for which this view
makes no provision. So soon, however, as we do admit, in the
body of the logical processes themselves, certain criteria of the
valid organization of thoughts, we then have standards whereby
to determine a greater or less validity and conclusiveness with
reference to an ideal of flawlessness in the logical process. As
a matter of fact, we actually find many grades and modes of
inferential process — the disjunctive, the hypothetical, the cate-
gorical — having varying degrees of psychological determi-
nateness and of logical conclusiveness ; and there are also
various groundings of proof, as in universal or particular
judgments, which actually do issue in varying degrees of logi-
cal cogency and validity. This is in so far confirmation of the
presence of regulative or normative principles in the logical
process, as such, having this logical value no matter what their
origin.

It follows from these criticisms that in our final interpreta-
tion of reality as, in any sense general, universal, or normative

5 6 J. M. BALDWIN.

pragmatism does not take us far. It omits all such modes of
logical reality, although by its pragmatic account of their origin
for utilities of life, it gives them some presumptive value. It
must deny this presumption or — refuse to be a logic or philos-
ophy ! This latter is the only sensible course, if we are to re-
main pragmatists in our genetic psychology.1

V.

The requirements of a philosophy of reality based upon —
or at least not invalidating — the results of mental development
under the law of utility, are fairly plain ; at least, in the negative
sense of not incurring the criticisms urged in this discussion.
They come out with some clearness in connection with the
statement of the problem of error.

The problem of error as involved in that of truth has been
a theme in many of the discussions in pragmatism. The pos-
sibility and meaning of error are somewhat more clear on the
theory which holds that knowledge is a copy, an inadequate or
defective copy, of a separate system of realities. The problem
to this theory is not to account for the presence of error, but to
reconcile it with the validity of thought. Otherwise all guaran-
tee and reliability fail in respect to truth. If thought may re-
port reality erroneously, how are we to know that it ever reports
it truthfully? The positive constructive task therefore is to
eliminate error, or to make it an incident only is a generally
valid process. The limit and ideal of the apprehension of the

1 This is explicitly the course taken lately by one of the fathers of prag-
matism, Dr. Peirce ; and the considerations he sketches in the Dictionary
article already referred to are similar to those worked out here. He suggests a
philosophy called ' Synechism ' (see his art. of that title also in the Dictionary),
in which he makes ' reasonableness ' its own justification, and attempts to do jus-
tice to logical 'generals.'

The failure to explain these aspects of reality with constructive thorough-
ness stands out in such writings as Moore's detailed criticism of Royce already
cited. Royce makes a constructive effort to overcome the dualism of thought
and action. He reaches an absolute system of thought which may also be
looked upon as an absolute purpose systematizing and completing finite pur-
poses. Moore claims that this is no real reconciliation of the representative
(logical) and practical aspects of the idea; but Moore, in his turn, goes no
further than the re-assertion of the concrete action-thought process as genetic
function.

THE LIMITS OF PRAGMATISM.

57

real is an errorless experience : a system of logical values in
which reality is completely and finally revealed; this is the
ideal of intellectualist or logical theories as opposed to prag-
matic theories.

To the pragmatist, on the contrary, error presents a different
problem. It must arise by some variation or interference in the
process whereby truth, and with it reality, are normally consti-
tuted. As a fact there is no difficulty, for the method of dis-
covery, called in science that of ' trial and error,' is just that
which is taken over and made the typical method of mental
development. It is a matter of ' cases,' trials, efforts, only the
ex post facto inspection of which reveals some — those which
are unfruitful or embarrassing — as errors. Error then, to the
pragmatist, is a normal aspect of the process of the discovery
of truth.

To this theory also, the limit of the process of apprehending
reality would appear to be the elimination of error. The de-
velopment processes, at their limit, would issue in an errorless
system of cognitions and judgments — logical processes — which
would be the reflection into thought of a perfectly adapted and
satisfied conation. Such system would be, within its own point
of view, closed and self-maintaining. If pragmatism asks at
all1 for the meaning and limit of its process, this would be about
the answer : a state of equilibrium, or an errorless experience.

The problem set to both theories, intellectualism and prag-
matism alike, therefore, is the elimination of error, in the treat-
ment of reality whether by thought or by action.

But to depict an errorless experience is to depict one in
which the dualism of experience and reality is overcome.
Error is essentially a 'phenomenon of dualism.

By no fair definition can error be attributed to a thought-
system which is self-sustaining and has no reference to what is
outside itself. But errorlessness is also the resolution of the
active processes of adjustment to which the thought aspect of
ideas is ancillary. Embarrassment, conflict and he.sitation,
restlessness, are the motor or practical results of cognitive
error.

1 Perhaps it doesn't ! — see Moore, Existence, Meaning \ and Reality, p. 25.

58 /. M. BALDWIN.

The problem therefore of philosophy — as it arises from the
discussion of the demands of pragmatism — is no longer that of
the reconciliation of two logical categories, being and becoming,
identity and diversity, teleology and mechanism, thought and
reality ; it is, on the contrary, that of the reconciliation of two
opposed schemes of evaluation of experience in general, that of
logical systematization and that of practical manipulation, each,
by the very terms of their relation claiming to be valid. It
is possible that there may be no reconciliation ; but the con-
sistent development of personality as a whole by the activities
in which these two evaluations arise, would lead us to expect
that they reveal modes rather than diverse kinds of reality, and
that there is some possible experience which, while enriched by
this contrast, is not torn asunder by it.

Such an experience would be a deeper revelation of the
nature of all the real than is either of the partial modes, and it
would, at the same time, admit of the criteria by which each of
the alternative points of view establishes the claim it makes.1

It would seem that this sort of requirement should commend
itself to the Chicago thinkers who refuse to stop in the dualism
of thought and action. Professor Dewey says (loc. cit., pp.
80 f.) : " Both material and tool [matter and thought] have been
secured and determined with reference to * * * the maintenance
of a harmonious experience * * * life proposes to maintain at
all hazards the unity of its own process. Experience insists on
* * * securing integrity even through and by means of con-
flict." Certainly then the final unity is one to be experienced
or lived in, an experience which is the immediate unity of an
autotelic whole ; a whole which includes logical realities whose
ends are practical and also practical realities whose ends are
logical. This is what Moore's criticism of Royce points out,

1This, I take it, is essentially the reconciliation which Professor Royce
attempts, though without, I think, the clear apprehension that it requires a cate-
gory of experience not definable ultimately as either ideas or ' purposes ' (cona-
tions), nor yet by saying that it is both. Professor Royce's recent work is how-
ever a notable advance upon a certain dualism of value and fact to be found in
many recent writings (and also in his earlier papers ; cf. his art. in International
Journal of Ethics, July, 1895, and the present writer's examination of it in Frag-
ments in Philosophy and Science, V., reprinted from the same journal, Oc-
tober, 1895).

THE LIMITS OF PRAGMATISM. 59

I think (cf. his utterance as to dualism on p. 372, with his quite
sentimental conclusion on p. 382, Studies in Logic]. And in
the matter of its treatment of dualism this view is not inconsistent
with those of Mead (The Definition of the Psychical}. Even
James' penchant for pluralism is to Dewey a case of aesthetic
unity in the thinker's contemplation !

In such an issue, reached from the previous criticism, I find
one of the approaches to a type of philosophy to which other
considerations, developed in various recent studies, have also
pointed.1 These ' approaches' converge upon a position which
finds in aesthetic experience, at each grade in the develop-
ment of the dualism of fact and values, truth and practice,
inner and outer, just the union and reconciliation of the two
sets of claims.2

Our conclusions may be summed up as follows :

1. Having successfully depicted the genetic processes by
which consciousness reaches the dualism of the thinking prin-
ciple and reality, it is the * genetic fallacy ' to treat one term of
this dualism, the thinking principle, as valid in the sense it
claims to be, and to deny that the other is.

2. If either of the terms of this dualism is to be made pri-
mary as a philosophical principle, it would seem to be the log-
ical reality term ; since it is genetically, at each stage of mental
development, just the definite, general, and communicable term
in which pragmatic gains are reflected. The pragmatic ac-
count of thought fully justifies its function of having general
meaning as well as concrete. Pragmatism can not complete
itself until it issues in a logical account of reality.

3. The universal and normative modes of thought do not
get adequate logical justification in a theory which finds the

ts and criteria of reality solely in concrete experiences of
usefulness, workableness, etc. It is just the general and uni-

1 See especially the article ' Mind and Body ' in the PSYCHOLOGICAL RE-
VIEW, May, 1903.

*It may be said (cf. Urban, PSYCHOLOGICAL REVIEW, January, 1896) that
the self is the source of union of the two contrasting modes of experience ; and
that is of course true. But to use the thought of self is to resort to one of the
categories in question, which involves the genetic fallacy of pragmatism ; and to
point out an experience in which the self finds its attitudes and values free from
the dualism is just the question at issue.

6o /. M. BALDWIN.

versal aspects of such modes of thought whose meaning would
not appear in any set of practical consequences. General tests
of systematization or organization as such within the body of
logical data would alone accomplish this. This throws us back
upon such principles as consistency, contradiction, etc. — yet
without prejudice to a thorough-going pragmatic account of the
origin of the function of thinking.

4. The final demand is for a real reconciliation of the dual-
ism of logical truth and experienced value ; both making claim
to interpret reality. This reconciliation must not deny the claim
of logic wherever the material is logical, not that of value
wherever a valuation is made ; and no solution is possible except
as itself an experience in which the dualism is actually out-
lived. Any other solution would be hypothetical only, and
derive its support from one or other of the two modes of the
dualism which is to be explained.

5. The thoroughgoing application of the genetic method, as
illustrated in the foregoing point (4), requires that no member
of a genetic dualism, or other contrast, be taken as explaining
principle of the process in which that dualism or contrast
arises. This is held to introduce a new philosophical point of
view : that of finding the further genetic process by which the
dualism is itself overcome, and of interpreting the nature of
the reality which is then constituted.1

1 This has been insisted upon, as necessary in science generally, and
formulated in the theory of 'Genetic Modes,' in the work Development and
Evolution, Chap. XIX.

DISCUSSION.

THE SEXUAL ELEMENT IN SENSIBILITY.

Any theory of socialization must give first-rate importance to the
influence on the individual of the presence, behavior and opinion of
others ; and the quality of suggestibility to social influence, so impor-
tant in the formation of the character of the individual and in the for-
mation of society itself, seems to have two sources, one in the food
process and the other in the process of reproduction.

The life of any highly organized species depends on the quickness,
precision and adequacy of its reaction to stimuli. New and dangerous
or advantageous situations are constantly presenting themselves and
the species develops both the cognition and the emotional reactions
suitable to accommodate to these. Every such species has, in conse-
quence, a high degree of susceptibility. Perhaps the most remarkable
expression of susceptibility in the human species is seen in the sen-
sitiveness of man to the opinion in which he is held by others. Social
life in every stage of society is characterized by an eagerness to make
a striking effect. A bare reference to the ethnological facts in this
connection will suffice : The Kite Indians have a society of young men
so brave and so ostentatious of their bravery that they will not fight
from cover nor turn aside to avoid running into an ambuscade or a
hole in the ice. The African has the privilege of cutting a gash six
inches long in his thigh for every man he has killed. The Melanesian
who is planning revenge sets up a stick or stone where it can be seen ;
he refuses to eat, and stays away from the dance ; he sits silent in the
council and answers questions by whistling, and by other signs draws
attention to himself and has it understood that he is a brave and dan-
gerous man, and that he is biding his time.

This bidding for the good opinion of others has plainly a connec-
tion with food-getting, and with the conflict side of life. High
courage is praised and valued by society, and a man of courage is less
imposed on by others and comes in for substantial recognition and the
favor of women. It is thus of advantage to act in such a way as to
get public approval and some degree of appreciation ; and a degree of
sensibility on the score of the opinion of others, or at least a reckoning
upon this, is involved in the process of personal adjustment.

61

62 THE SEXUAL ELEMENT IN SENSIBILITY.

But the problem of personal adjustment at this point would seem
to call for more of intelligence than emotion; and we find, on the
contrary, an excess of sensibility and a mania for being well thought
of hardly to be explained as originating in the exigencies of tribal
organization, nor yet on the score of its service to the individual in
getting his food and living out his life. Why could not primitive man
live in society, be of the war-parties, plan ambuscades, develop his
fighting technique and gear, be a blood-brother to another man, show
his trophies, set a high value on his personality and insist on recogni-
tion and respect, without this almost pathological dependence on the
praise and blame of others?

Or if we approach the question from another standpoint and in-
spect our states of consciousness, we find signs that we have a greater
fund of sensibility than is justified in immediate activity. We have
the same mania to be well thought of; we are unduly interested when
we hear that others have been talking about us, we are annoyed, even
furious, at a slight criticism, and are childishly delighted by a compli-
ment (without regard to our deserts) ; and children and adults alike
understand how to put themselves forward and get notice, and equally
well how to get notice by withdrawing themselves and staying away
or out of a game. We have a tendency to show off which is not ap-
parently genetically connected with exploit or organization, and we
recognize that this form of vanity is not consistent with the ordinary
run of our activities when we argue with ourselves that the opinion of
this or that person is of no consequence and attempt to think our-
selves into a state of indifference. Intellectually and deliberately our
attitude toward criticism from others would often be, if we could
choose, represented by Tweed's query, 'What are you going to do
about it?' but actually it puts us to bed.

All of this seems to indicate that there is an element in sensibility
not accounted for on the exploit or food side, and this element is, I
believe, genetically connected with sexual life. Unlike the struggle
for existence in the ordinary sense of the phrase, the courtship of the
sexes presents a situation in which an appeal is made for the favor of
another personality, and the success of this appeal has a survival
value — not for the individual, but for the species through the individ-
ual. We have, in fact, a situation in which the good opinion of an-
other is vitally important. On this account the means of attracting
and interesting others are definitely and bountifully developed among
all the higher species of animals. Voice, plumage, color, odor and
movement are powerful excitants in wooing and aids both to the con-

DISCUSSION. 63

quest of the female and the attraction of the male. In this connection
we must also recogni/.e the fact that reproductive life must be con-
nected with violent stimulation, or it would be neglected and the spe-
cies would become extinct; and, on the other hand, if the conquest of
the female were too easy, sexual life would be in danger of becoming
a play interest and a dissipation, destructive of energy and fatal to the
species. Working, we may assume, by a process of selection and sur-
vival, nature has both secured and safeguarded reproduction. The
female will not submit to seizure except in a high state of nervous ex-
citation (as is seen especially well in the wooing of birds), while the
male must conduct himself in such a way as to manipulate the female;
and, as the more active agent, he develops a marvelous display of tech-
nique for this purpose. This is offset by the coyness and coquetry of
the female, by which she equally attracts and fascinates the male and
practices upon him to induce a corresponding state of nervous excita-
tion.1 This is the only situation in the life of the lower animals, at
any rate, where the choice of another is vitally important; and corre-
sponding with the elaborate technique to secure this choice we have
in wooing pleasure-pain reactions of a violent character. In a word,
extreme sensitiveness to the judgment of another answers on the sub-
jective side to technique for the conquest of a member of the opposite
sex. It seems, therefore, that we are justified in concluding that our
vanity and susceptibility have their origin largely in sexual life, and
that, in particular, our susceptibility to the opinion of others and our
dependence on their good will are genetically referable to sexual life.

This view would be completely substantiated if we could show
that the qualities of vanity and susceptibility in question are present
in any species where it is impossible to assume that they were devel-
oped in connection with the struggle for food and as the result of the
survival of types showing a tendency to combine and cooperate in the
effort to get food. And we do, in fact, have cases of this kind among
some of the lower animals. It cannot be said that the dog, for in-
stance, has survived in the struggle for existence because of his sensi-
tiveness to public opinion in his species nor on account of an interest
in being well thought of by the community of dogs at large which
would lead him to behave in a public-spirited or moral manner. At
the same time, the dog in his relation to man shows as keen a sensi-
tiveness to man's opinion and treatment as does man himself. The
attention which the master pays to one dog will almost break the heart
of a dog not receiving it. A neglected dog plainly suffers as much in

1 See Groos, The Play of Animals, p. 283.

64 THE SEXUAL ELEMENT IN SENSIBILITY.

his way as the soldier who is sent to Coventry by his messmates ; and if
neglected and jealous dogs do not commit suicide, as they are reported
to do, they are evidently in a state of mind to do so. This means
that the dog has highly developed susceptibility to the appreciation of
others, and that the species which he represents has had no history
except a sexual history capable of developing this mental attitude. In
connection with courtship he developed a fund of organic susceptibil-
ity, and this condition is involved in his more general relation to man :
the machinery set up in sexual relations is played on by stimuli in
general. A condition favorable to stimuli of a particular kind is fav-
orable to stimuli in general ; and it seems likely that this not very
prominent fact of a state of excitation in a sexual connection is an im-
portant factor in the formation of the mind and of society.

There are also certain conditions in the development of the indi-
vidual and of society where the sexual type of reaction is so near the
surface that it shows through in connection with political, moral and
other essentially non-sexual activities. Passing over the fact that the
period of adolescence is noticeably a period of ' susceptibility ' and
personal vanity, we may take as an example of the intrusion or per-
sistence of the sexual element in conditions of a non-sexual kind the
frequent association of sexual with religious excitement.1 The appeal
made during a religious revival to an unconverted person has psycho-
logically some resemblance to the attempt of the male to overcome the
hesitancy of the female. In each case the will has to be set aside,
and strong suggestive means are used ; and in both cases the appeal is
not of the conflict type, but of an intimate, sympathetic and pleading
kind. In the effort to make a moral adjustment it consequently turns
out that a technique is used which was derived originally from sexual
life, and the use, so to speak, of the sexual machinery for a moral ad-
justment involves, in some cases, the carrying over into the general
process of some sexual manifestations. The emotional forms used
and the emotional states aroused are not entirely stripped of their
sexual content. On the race side, also, there is a stage in develop-
ment where the sexual pattern is transferred almost unmodified to
public affairs. The following extracts from a lengthy description
given by Mr. Bowdich of his reception by the king of Ashanti, in the
year 1817, will illustrate sufficiently the employment of the turkey-cock
pattern of activity in political relations :

lSee e. g., Krafft-Ebing, Psychopathia Sexualis, 3 Aufl., p. 10; Adams,
' Some Phases of Sexual Morality and Church Discipline in Colonial New
England,' Proceedings of the Mass. Hist. Soc., ad Series, 1891, pp. 417-516.

DISCUSSION. 65

*' The sun was reflected with a glare scarcely more supportable than
the heat from massive gold ornaments which glistened in every direc-
tion. More then a hundred bands burst at once on our arrival, with
the peculiar airs of their several chiefs ; the horns flourished their de-
fiances, with the beating of innumerable drums and metal instruments,
and then yielded for a while to the soft breathings of their long flutes.
* * * At least a hundred large umbrellas or canopies, which could
shelter thirty persons, were sprung up and down by the bearers with
brilliant effect, being made of scarlet, yellow, and the most showy
cloths and silks, and crowned on the top with crescents, pelicans,
elephants, barrels, and arms and swords of gold. * * * The cabo-
ceers, as did their superior officers and attendants, wore Ashanti
cloths of extravagant price, from the costly foreign silks which had
been unravelled to weave them in all the varieties of color as well as
pattern ; they were of incredible size and weight, and thrown over
the shoulder exactly like the Roman toga ; a small silk fillet generally
encircled their temples, and many gold necklaces, intricately wrought,
suspended Moorish charms, dearly purchased, and enclosed in small
square cases of gold, silver and curious embroidery. Some wore
necklaces reaching to the waist, entirely of aggry beads ; a band of
gold and beads encircled the knee, from which several strings of the
same depended ; small circlets of gold, like guineas, rings and casts of
animals were strung round their ankles ; their sandals were of green, red,
and delicate white leather ; manillas, and rude lumps of rock gold hung
from their left wrists, which were so heavily laden as to be supported
on the head of one of their handsomest boys. * * * [The king]
wore a fillet of aggry beads round his temples, a necklace of gold
cockspur shells strung by their larger ends, and over his right shoul-
der a red silk cord, suspending three sapphires cased in gold ; his brace-
lets were of the richest mixtures of beads and gold, and his fingers
covered with rings ; his cloth was of a dark green silk, a pointed
diadem was elegantly painted in white on his forehead ; also a pattern
resembling an epaulette on each shoulder, and an ornament like a full
blown rose, one leaf rising above another until it covered his whole
breast. * * * The belts of the guards behind his chair were cased in
gold, and covered with small jaw-bones of the same metal ; the ele-
phants' tails, waving like a small cloud before him, were spangled
with gold, and large plumes of feathers were flourished among them.
His eunuch presided over these attendants, wearing only one massive
piece of gold about his neck ; the royal stool, entirely cased in gold
was displayed under a splendid umbrella, with drums, sankos, horns,

66 THE SEXUAL ELEMENT IN SENSIBILITY.

and various musical instruments, cased in gold, about the thickness of
cartridge paper ; large circles of gold hung by scarlet cloth from the
swords of state ; * * * hatchets of the same were intermixed with
them ; the breasts of the Ochras and various attendants were adorned
with large stars, stools, crescents, and gossamer wings of solid gold-
* * * ^» i

It is not surprising that the characteristically sexual method of
display and emotional appeal should be associated with the earlier
efforts at adjustment, both in the individual and in the state. This
method is based on the instincts, and just as inhibition and brain inte-
gration follow the instincts in point ofdevelopment, a rational mode of
control, individual and public, is developed later than the emotional
form, or, at any rate, is not at first independent of it.

The origin of mental impressionability seems to lie then not in one
but in the two general regions of activity — that connected with the
struggle for food and that connected with reproduction. The strain
on the attention in the food and conflict side of life involves the de-
velopment of mental impressionability, particularly of an impression-
ability on the side of cognition. But in addition we have the impres-
sionability growing out of sexual life which has been in question above,
and which is more closely related to appreciation than to cognition.
And of these two aspects of impressionability — the one growing out
of conflict and the one growing out of reproduction, the latter has more
social possibilities than the former, because it implies a sympathetic
rather than an antagonistic organic attitude. It is certainly in virtue
of susceptibility to the opinion of others that society works — through
public opinion, fashion, tradition, reproof, encouragement, precept
and doctrine — to bring the individual under control and make him a
member of society ; and it is doubtful whether this could have been
accomplished if a peculiar attitude of responsiveness to opinion had
not arisen in sexual relations, reinforcing the more general and cog-
nitive impressionability. Without this capacity to be influenced the
individual would be in the condition of the hardened criminal, and
society would be impossible.

This sex-susceptibility which was originally developed as an acces-
sory of reproduction and had no social meaning whatever, has thus, in
the struggle of society to obtain a hold on the individual, become a
social factor of great importance and together with another product at
sexual life — the love of offspring — it is, I suspect, the most imme-
diate source of our sympathetic attitudes in general, and an important

1 Ellis, The Tshi-speaking Peoples of the Gold Coast, p. 249 ff.

DISCUSSION. 67

force in the development of the ideal, moral and aesthetic sides of life.
It is perhaps not pushing the matter too far to suggest also that the
duality of motivation which characterizes our social system, and which
we designate as egoism on the one hand and altruism on the other, is
a natural result of the contrast in character between the states of con-
sciousness originating in the struggle for food and those originating in
courtship, and that the history of society on the moral and aesthetic
sides is in great part the history of an attempt to make the more sym-
pathetic attitude prevail over the more antagonistic.

W. I. THOMAS.
THE UNIVERSITY OF CHICAGO.

DR. MORTON PRINCE AND PANPSYCHISM.

In the last number of this REVIEW Dr. Morton Prince points out
that the panpsychist doctrine of my Why the Mind has a Body was
anticipated in his Nature of Mind and Human Automatism pub-
lished in 1885. The interesting quotations which he makes from the
latter and from an article in Brain for 1891 certainly bear him out in
this assertion. With the book I regret to say that I was unacquainted.
But I read his article some ten years ago, and I think that it contributed
to turn my thoughts in a panpsychist direction. Its clear teaching
that consciousness is the reality which appears as the brain-process
helped to fix that hypothesis in my mind, and I have no doubt that the
pages of Paulsen,1 to which I have always felt myself mainly indebted?
had a fuller meaning to me in consequence. I regret the more that, by
the time when I came to write, my memory of its contents had lapsed,
and I failed, rather inexcusably it now seems to me, to mention Dr.
Prince among earlier expositors of the theory. Had I been acquainted
with his book, I should have had a juster appreciation of his merits as
a pioneer of panpsychism, and this would not have happened.

I have lately had the pleasure of reading the book, and I find that
it contains an extremely clear and forcible statement of the panpsy-
chist hypothesis. It differs from Clifford's essay on the ' Nature of
Things-in-Themselves ' in being not merely an exposition of panpsy-
chism in the abstract, but a definite application of it to the problem
of the connection of mind and body.2 Readers who have difficulty in

lEinleilung in die Philosophie (1892), see esp. pp. 77-116, 137-149,
381-85.

1 Towards the close of his essay on Body and Mind Clifford does make a
rather definite application of it to the problem, as is shown by the sentence :
" If mind is the reality or substance of that which appears to us as brain-action
* * * " (Lectures and Essays, 2d ed., p. 269).

68 DR. MORTON PRINCE AND PANPSYCHISM.

understanding my account of the matter would do well to consult it.
In particular, the pointing out of the ambiguity of the term ' matter,'
the rejection of non-empirical views of mind, the definite use of the
conception of mental causality, the recognition that brain-events
are not merely symbols but effects of the ' accompanying ' mental
states, indicate the clearness with which the conception is grasped in
all its implications. That Dr. Prince should have worked this con-
ception out for himself, without knowledge of Clifford, simply (as it
appears) by way of criticism of the views of Huxley, Tyndall, Lewes,
Spencer, and Bain, is an achievement to be proud of. Should the
theory triumph, he will be entitled to an honorable place among its
earliest discoverers and defenders.

There is one slight misapprehension in Dr. Prince's paper which
I desire to set right, and that is as regards my attitude toward Clif-
ford. Dr. Prince seems to think that I disapprove of Clifford's views,
and regard them as somehow distinct from my own. I assure him
that this is a mistake. The fact that I describe my theory, in my pref-
ace, as " that which is implied in the panpsychism of Fechner and Clif-
ford " should, I think, have made any doubt as to the nature of my
feeling impossible. I presume his error is due to my having referred
to Clifford as a parallelist; whereas, according to Dr. Prince, parallel-
ism is an erroneous theory, quite incompatible with panpsychism.
This is really only a question of terms. Dr. Prince understands
by parallelism the assertion of a second real series, running parallel
with the psychical; and this, I agree with him, is a wholly erroneous
conception. But I think it is not expedient, as a matter of termi-
nology, to pin the parallelist down to the assertion of the independent
reality of matter. If the interactionist may remain still an inter-
actionist even though he conceive the matter on which the mind
acts idealistically, then the parallelist may still remain a parallelist
even though he conceive that the brain-process has no existence except
when an external observer chances to perceive it.1 The essence of
parallelism is the denial of causal relations between mental and phys-
ical; and this denial, as I have shown in my book (p. 345), remains
still valid on the panpsychist theory.

I want to take this opportunity of calling attention to a number of
other panpsychist discussions of the relation of mind and body, with
some of which I have only recently become acquainted. Professor

1 German critics of panpsychism have fallen into this error, and are pre-
vented by it from understanding the real meaning of the theory — see Heymans'
review of Basse in Zeitschrift fiir Psychologic, Bd. 33, Heft 3, esp. pp. 217-219.

DISCUSSION. 69

Lloyd Morgan, in his Animal Life and Intelligence, published in
1891, has a chapter on Mental Evolution, in which a panpsychist
theory is sketched out (Ch. XII., pp. 464-503). The late Joseph
LeConte hints at a panpsychist theory of the connection of mind and
body in the remarks contributed by him to Professor Royce's Concep-
tion of God (1897; see pp. 67-68). Professor Royce himself, de-
spite his early condemnation of ' mind-stuff' (see Mind, O. S., Vol.
VI., pp. 365 ff.), appears as a panpsychist in his essay on 'Self-
consciousness, Social Consciousness and Nature' (published orig-
inally in Philos. Rev., Vol. IV., pp. 465 ff., 577 ff., and reprinted
in his Studies of Good and Evil — see especially pp. 229, 230 of the
latter). Panpsychist principles seem to underlie Mr. Henry Rutgers
Marshall's acute discussion of the parallelistic view in his Instinct
and Reason (1898, pp. 19-67). Professor Walter Smith has crit-
icized the notion of interaction very judiciously from the same point of
view in Philos. Rev.,\o\. X., pp. 505-514, his little paper contain-
ing all the essential ideas of my book.

Professor Stout's chapter on ' Body and Mind ' in his Manual of
Psychology I have referred to in my preface. Professor Ebbing-
haus's discussion of the subject will be familiar to readers of his
Grundziige (pp. 27-47). Both of course are panpsychist.

No person interested in the question should fail to read Professor
Hey mans' article ' Zur Parallelismusfrage,' in Zeitschrift fur Psy-
chologic, Bd. 17, pp. 62-105. ^n a capital review of Busse's recent
book in the same journal (Bd. 33, pp. 216-222), Professor Heymans
defends the panpsychist theory against misapprehensions, in a series
of brief objections and replies which may be commended to the atten-
tion of those who think they see reasons for rejecting it.

C. A. STRONG.
COLUMBIA UNIVERSITY.

EDITORS' NOTE.

The customary sections devoted to ' Literature,' ' New Books,'
' Notes,' etc., are hereafter to be printed separately, appearing on the
fifteenth of each month, in form considerably enlarged and compris-
ing certain new features. For convenience of reference it will be
known as The Psychological Bulletin. It is to constitute a separate
volume, although still essentially part of the REVIEW.

Manuscripts for publication, books for review, and editorial matter
generally should be addressed hereafter to Prof. J. Mark Baldwin,
Johns Hopkins University, Baltimore, Md. ; business communications
and advertising matter to Prof. H. C. Warren, Princeton, N. J.

70.

N. S. VOL. XL No. 2

March, 1904

THE PSYCHOLOGICAL REVIEW.

THEORY AND PRACTICE.1

BY PRESIDENT WILLIAM LOWE BRYAN,
Indiana University.

I.

Two of my predecessors have discussed from this chair the
application of psychological theory to practice. Upon the prin-
cipal question considered by them, as you know, they did not
agree. However, difference of opinion upon this point is less
surprising than unanimity would be. For time out of mind
there have been not simply many divergent opinions as to the
relations of theory and practice, but several types of such
opinions persisting side by side century after century in collision.
In one case it is believed that there is a philosophy, which gives
a finally valid account of all reality and which lays down the
law for action in every field. In another case, it is not philoso-
phy, but empirical sicience which, as it develops, is to free us
from all the rules of thumb by which our ancestors groped and
fumbled their way, and which is to show us with certainty and
on rational grounds exactly what to do in every field. In a
third case it is not philosophy and not science, not systematic
learning of any sort, but intuition, tact, common sense, which
alone enable us to achieve success in any field.

The mention of these typical opinions brings to mind at once
many great names which could be cited for and against each of
them. I have sometimes tried to make the historic conflict of

1 President's address, American Psychological Association, St. Louis Meet-
ing, December, 1903.

72 W. L. BRYAN.

opinion upon this subject concrete for myself by imagining a
committee selected from the great philosophers, scientists, poets,
and men of affairs of history, the committee being directed, let
us say, to act together as trustees of a village school. Plato,
Cervantes, Comte, Prince Bismarck, Thomas Carlyle — whom
you please, — it would be easy to make up an interesting com-
mittee. The debates of that committee, the hopeless reciprocal
misunderstandings, the scorn or compassion of each man for all
the others, it would take Shakespeare to imagine. And the
scene would be worthy of Shakespeare for in a way, the most
fundamental conflicts of the history of culture with all their
humor and with all their gravity would be there. But even
Shakespeare, I fear, could not imagine what the committee
would decide to do. And yet decision as to what to do is the
unavoidable task of most of us who profess psychology or
indeed any science. For we are obliged to propose courses of
study and to advise students who have in view one or another
profession what courses they shall take. But what courses we
propose and what advice we give depend over and over upon
what we believe as to the practical usableness of our science. We
are therefore each of us bound in conscience to face the question
as well as we can, not as one of those questions which may wait
upon the leisure of science, but as an always immediate question
to which we can scarcely help giving daily some sort of answer
to those who look to us for guidance. We are in the position
of the conscientious physician who would like to wait for the
instruction of another hundred years of experimental medicine
but who must do as well as he can with the patient before him.
For better or for worse, therefore, I shall give the conclusions
which with time have come to me.

II.

THE FAILURE OF THEORY.

When a theory will not work, as so often the most promis-
ing theory will not, I believe the difficulty lies simply in the
fact that the theory is not true — not true, that is, with a suffi-
cient degree of approximation. An action is always necessarily

THEORY AND PRACTICE, 73

concrete, subject not only to certain known general laws and to
certain known definite conditions, but subject to the whole of
reality then and there effectively present. No theory com-
pletely embraces all the conditions determining any action.
Some conditions are omitted unintentionally because of igno-
rance. Some conditions are excluded intentionally, on the one
hand as disturbances which interfere with the accuracy of ex-
perimental results, on the other hand as complications which
interfere with the possibility of mathematical or logical treat-
ment. The intentional exclusion of disturbing or complicating
conditions is not a procedure which requires defense. Its defense
is found in the whole history of learning, and after that in the
history of the practical applications of learning. To make any
progress, we must focus for certain things and be temporarily
blind to environing things.

It may be, however, that in arriving at a theoretical result,
either because of my ignorance, or because of the very efforts
to be exact or to be logical, I shall leave out of account condi-
tions which are not in fact insignificant, which will not be absent
when my bit of theory is tried, which will be there to upset all
my previsions and to bring me to confusion. My air ship will
not fly. In such a case, the best fortune is immediate and de-
cisive practical trial. Decisive failure destroys our illusions, if
we have them, and sets us looking for conditions which have
been overlooked. Unhappily, however, decisive trial of theo-
retical results is often indefinitely postponed. In this case, the
scholar must be of extraordinary constitution if he escape the
historic disease of his kind, namely, blindness to realities which
his method has not embraced.

I wish to consider two types of this illusion of the scholar.
One of them, which may be called the illusion of consistency , is
generally recognized. The other, not so generally recognized,
I shall call the illusion of precision. I wish to show how in
both cases these illusions spring directly out of the painstaking
employment of methods which must be employed to discover
the truth, and how, when they have risen, they render the
scholar blind to certain aspects of truth which are not insignifi-
cant either in theory or in practice.

74 W. L. BRYAN.

The Illusion of Consistency.

I am, let us suppose, a scholar who is impressed above all
things with the necessary self-consistency of the truth. Ac-
cordingly, I have spent years in developing a system of greater
or less extent, which, to my mind, has the quality of complete
self-consistency. I have made its consistency explicit, by stat-
ing everything in exact logical or perhaps mathematical form.
Every term, every proposition or equation, every syllogism or
problem is perfectly defined and the whole stands, to my mind,
flawless and self-evidential. Everything in it hangs together.
Everything in it can be shown to be as certain as the most cer-
tain thing in it and that thing no sane man can doubt. Here is
the truth, final and clear, and here, within the field concerned,
is the law for action.

Whether such a system be finally credited with great value
or with small, it is sure to have certain characteristics which
limit its value. Its salient merit of exact logical or mathe-
matical consistency was bought at a price. That price was the
exclusion of conditions too complicated to be dealt with by the
logical or mathematical methods employed. This price was
paid by Spinoza in one field and by Newton in another. The
procedure requires no defense. It is necessary. There is no
definition without negation.

However, a life time spent in developing and contemplating
such a system makes it easy to forget and ignore altogether what
the method has excluded. Every clear idea, as we know ex-
perimentally, makes it harder to do justice to impressions just
unlike those which belong with that idea. A system of such
ideas is self-protecting somewhat after the analogy of a living
organism. Every item in the system is felt to be proof of and
proved by all the others. Everything in the system comes to
the point of attack, makes me abnormally sensitive for faint
experiences of the right sort and abnormally oblivious to salient
facts of the wrong sort. In a word, there is perhaps no hypno-
tic agent more powerful to sharpen the sight or to dull it than a
system of ideas which one has made for himself, and whose
truth seems \guaranteed at every turn by complete internal con-
sistency.

THEORY AND PRACTICE. 75

Very likely this hypnotic illusion of consistency is strongest
when the system concerned is believed to be all-embracing — a
philosophy of God, the world, man, what not; and the illusion
is the less likely to be broken because decisive trial is so diffi-
cult if not quite impossible. However, it is not simply the phi-
losophers who along with their systems of beliefs, develop the
illusion of consistency. No doubt every man does so in a de-
gree and men of science along with the rest. The history of
science is full of examples. It is seldom that a scientist is able
to do justice to facts which controvert his most important theories.
For this reason there is sober truth in the cynical remark that
the progress of science requires the death of scientists.

The illusion of consistency as I have said, is very well
known, for it springs out of conditions which have been legit-
imately and conspicuously present throughout the history of
learning. And so for centuries this illusion has been notori-
ous as a limitation of the scholar's knowledge and practical
judgment.

I turn to an analogous illusion which is less generally re-
cognized.

The Illusion of Precision.

To take a typical case, let us suppose that I am not a logician
but an experimental scientist. I cultivate a distrust for phi-
losophy. I am wary of all elaborate argumentation. Logic is
a trap. I have studied facts pure and simple. I have lived in
the laboratory. I do nothing except with instruments of preci-
sion. I have learned how to shut out disturbing conditions with
the last degree of refinement. My results are strictly quantita-
tive. Everything has been verified over and over and is verifi-
able by whom you please ad libitum. The outcome is not
poetry, not a guess, not a speculation. It is science and within
its field, it is the law for action.

It would be idle in this presence to insist upon the value of
such procedures and such outcomes. The chief merit of our
time lies doubtless in the fact that we have succeeded better
along these lines than men ever did before. And yet directly
out of the methods which science must employ, there rises over
and over again an illusion which stands between the scholar
and the truth and which may make him a failure in practice.

76 W. L. BRYAN.

Those disturbing conditions which were with infinite pains
shut out may be practically insignificant. Or the scientist may
take adequate account of them in a separate study. But some-
times they are not insignificant and sometimes after having
carefully shut them out of his laboratory the scientist forgets
them altogether and does not dream that they are waiting out-
side his laboratory door ready to take revenge when his formula
come to trial. Unhappily the necessary practical tests are often
long delayed or indecisive. This is true in every field of
science and there is no field of science where such delay does
not permit the illusion of precision to survive.

But when the phenomena concerned are very complicated,
when for example, we confront the complexities of human
nature in the individual and in society, when we attack by
exact scientific method the problems of psychology, ethics,
political economy, or any science dealing with human life and
thereupon undertake to tell men what to do, we have then the
best possible conditions for the development of the illusion of
precision.

For on the one hand it is possible in all these fields to be as
precise as one will. There are methods from the older sciences
to serve as analogical models. There are, if you like, instru-
ments of the highest precision. One has only to be scrupulous,
persistent, intolerant of errors. One will end by securing re-
sults, which whatever else may be true of them, are at any rate
exact. All this tends to establish in the man who does it a faith
which cannot be shaken. There is my machine. There is my
mathematical method. There are my statistics. There is my
sure concrete fact which no one can deny, which all the world
may verify. There is a bit of science which will stand till the
judgment day and take its place along with all the rest. How
can there be any illusion in this? Is not this precisely the
death of illusions? Is not this incoming of exact science the
beginning of the end of every erroneous conception of human
life?

So be it. There rises here nevertheless an illusion from
which few of us altogether escape. If I would remember just
what my scientific work has actually made known to me,

THEORY AND PRACTICE. 77

namely, a fragment, which exists never in isolation but always
in flux with innumerable other things which have not been
scientifically determined, that would guard me against serious
illusion — that would keep me, as a scientist, from believing
or from advising or from prophesying except within the safe
and narrow limits of my scientific knowledge.

But in fact it is fatally easy to forget how little I know, to
forget the whole tangle of things which I have left out through
ignorance or shut out in the interest of accuracy, to believe in a
word that the whole complex affair from which I have painfully
abstracted and defined a fragment goes on by the rules laid
down in my monograph.

If one wishes to see the illusion of precision in an extreme
and typically clear case he can find it sometimes in a young
man just become a doctor. The young man has to his credit
one dissertation upon some item of human experience. That
has made him an initiate. He has passed from the outside
world and is one of those who may speak to the outside world
with the authority of science. His work touches great affairs
in education, politics, ethics or religion. Time out of mind
men have dealt with these affairs by rule of thumb, by their
five wits, by what you please. The time for all that is past.
This is the age of science. Let all concerned read this disser-
tation and govern themselves accordingly.

I wish I could say that this illusion were confined to a few
unripe doctors of philosophy. In truth, the literature of the
sciences dealing with human life overflows with examples,
wherein men demand and expect a new education, a new
politics, a new ethics, the revolution of institutions, each man
assuming to speak with the authority of science, while yet no
six of them could agree upon the programme which science
requires.

In brief taking for granted that every sort of phenomena
admits of and requires exact scientific investigation, I am of
those who believe that every bit of knowledge, so far as it is
true, is actually or potentially practical. I see no reason why
any pulse of consciousness which reflects any aspect of reality
may not really and usefully affect action.

7§ W. L. BRYAN.

It is an obvious fact, indeed, that in some cases the state
of our knowledge permits us to formulate rules of procedure
such that the results may be foreseen in highly accurate detail,
while in other cases such precise prevision and prescription are
quite impossible. Naturally the cases where this is possible lie
in fields where the phenomena involved are simplest, most easily
defined, most thoroughly studied and therefore already most
completely understood. These conditions are doubtless found
best in the phenomena dealt with in the simpler chapters of
mechanics, chemistry, etc., and are most conspicuously absent
when we confront the subtle complexities of human behavior.
We know how to make soap but we do not know what Shake-
speare will say next.

Reflection on this obvious contrast has given rise to the doc-
trine that there is a difference -per se between nature and mind
such that exact theoretical and practical science is possible in the
one case but not in the other. In fact, however, the line be-
tween the simple well-understood phenomena where we have
exact theoretical and practical knowledge on the one hand
and the complex, little understood phenomena where we must
guess and fumble and grope is not at all identical with the line
which divides nature and mind. Our chemistry, mechanics,
physics, biology, etc., confront fields within which nearly every-
thing remains to be done and where we can still do nothing but
guess and fumble and grope.

On the other hand we are not without a body of definite
verified knowledge of human nature which gives us reliable
practical guidance. I can think of no reason why this theo-
retical and practical knowledge of human nature should not
continue to grow side by side with physical science both of
them becoming with the years more comprehensive, more exact
and more useful.

Nevertheless, the life of the scholar tends to unfit him to
succeed practically in any field, tends to make his advice inade-
quate in every field, unless his work as scholar is tested, cor-
rected and brought into due perspective with things outside his
specialty by thoroughgoing practical experience. A lifetime
spent in developing a system whose criterion of validity is its

THEORY AND PRACTICE. 79

internal logical or mathematical consistency, may bring about a
signal advance toward a finally valid view of all truth. In like
manner a lifetime spent in intelligent scientific research makes
its contribution to theoretical and in the long run to practical
knowledge. But never, I believe, does either of these proced-
ures or both of them combined determine all the conditions of
any action. Always some of these conditions are shut out
through ignorance or for the sake of consistency or for the
sake of accuracy. From these excluded conditions the eye of
the scholar is holden so that he cannot see them. And when
from the height of his learning he tells the foolish multitude
what to do, it is not simply the multitude which replies that he
also is foolish. It is over and over again the greater reality
which, speaking through the event, brings him to confusion.

III.

THE SUCCESS OF THEORY.

I turn now briefly to the question, how may we mediate
between abstract aspects or fragments of truth and the require-
ments of practice ? There are two answers to this question which
have weight beyond any individual opinion.

Concrete Science.

The first answer is given in one clear form by the higher
schools of technology. The professors in these schools are in
the best cases men who after thorough training in one or another
fundamental science, devote themselves to the study of concrete
problems for which a practical solution is required.

It is not to be overlooked that such studies have value as
contributions to scientific theory. The technical sciences are
not simply borrowers from the pure sciences. They exploit
new aspects of reality. They establish new facts which ' stand
in their own right, throw light upon the less and the more com-
plicated aspects of reality and so do their share toward a future
correlation of the sciences into science.'1 What concerns us
now, however, is not the contribution of such studies to scien-
tific theory, but the fact that such studies must be made as
bridges between abstract science and practice.

1 Bryan and Harter, PSYCH. REV., VI., 346.

8o W. L. BRYAN.

If we inquire for analogous studies within the field of psychol-
ogy, what showing can be made ? We have for one thing a
literature dealing with artificially isolated aspects of conscious
life, such as will, attention, association and the like. We have
another literature dealing experimentally with functions and
processes which are in themselves concrete but which in the
investigations are isolated from the complex stream of life in
which alone they normally occur. We have finally pseudo-
scientific literatures, phrenology, physiognomy, and the like,
which are concrete enough, and which tell all men specifically
what to do, but which science has disowned.

When we have told off these departments of our literature,
comparatively little remains, and yet something remains.
"Within the fields of comparative psychology, psychiatry,
criminal and industrial psychology, we have pictures of the
typical conduct of animals, children, melancholiacs, paranoiacs,
etc., which instruct us better than unscientific popular psychol-
ogy can, what to expect and what to do in dealing with individ-
uals of these sorts." 1

What the future will bring forth in the field of concrete
psychology, whether in time the studies in this field will approach
in importance the studies which now issue from the technical
schools, only the future can show. For myself I have grown
in the belief that in a great range of current psychological
problems it is good strategy for the experimental psychologist
to supplement his investigation of isolated activities and func-
tions by the investigation of concrete activities and functions as
they appear in everyday life. I believe that in this direction
there lie new chapters in the history of psychology.

Before leaving this point I should like to say that a large
part of the work of Professor James seems to me concrete and
also practical in the sense which I have indicated. Professor
James has not sought to develop a rigidly logical system within
which everything should interlock with everything else. He
thinks apparently that such a system sacrifices truth to logic.
On the other hand his book is not a bare list of findings based
upon laboratory statistics. No one has seen more clearly than

1 Loc. cit., 347.

THEORY AND PRACTICE. 8 1

he how the significant truth may evaporate through the finest
mesh of statistics, leaving behind it the illusion of precision.
Vet, notwithstanding his avoidance of an extremely systematic
psychology on the one side and of an extremely experimental
statistical psychology on the other side, his work is on all sides
recognized as of quite first rate rank in the history of psychol-
ogy- What he has done again and again is to satisfy men of
many sorts, psychologists and laymen, that he has hit off a bit
of life as it is. At the best his sentences are like Flaubert's
•phrases viables. A thing is at last once for all said. These
happy achievements which are I believe, as practical as they
are true, we owe to the fact that this author has the nearly
unique advantage of being a scholar who is also an artist.

Experience -with Affairs.

A second indispensable form of mediation between theory
(whether abstract or concrete) and practice is found only in
personal practical experience with affairs. Many scholars of
course never enjoy this experience. Some never wish to en-
joy it. It is easy as we know for a professor to become in
effect a monk, living apart in his university monastery with
cool and distant regard for the society from which himself and
the university derive — often with childlike ignorance even of
those practical affairs which his own specialty most nearly
touches. If such a man does his own business well his social
isolation is socially justified. He does one thing which he can
do best and all men profit by it. The practical results of such
work may, in the long run, prove to be incalculably great.
Faraday, Kant or Darwin works fifty years upon a problem
which seems remote. The busy public will believe anything of
him except that he will ever accomplish anything practical.
Yet we know very well that the ideas of such a scholar may in
another fifty years quite transform both the ideas and the forms
of business of the practical public.

It is equally certain, however, that learning does not effect
such results except through scholars who are also men of affairs.
It is the extraordinary good fortune of society to have had not a
few such men. A Kelvin becomes counselor to engineers. A

82 W. L. BRYAN.

Lecky or a Virchau serves in Parliament. A Lowell or a White
enters the diplomatic service. An Eliot becomes a university
president. In such a case the scholar does not confront society
with remote academic advice. With all his learning, experience
and will he grapples with men and affairs as they are. He is
not there to announce principles. He is there to secure results.
His principles are to be made flesh and dwell among us. His learn-
ing and his ideals throw their light about him as he works, but
in the stubborn and tangled realities with which he works there
is also light which in a life time may quite illuminate and trans-
figure his learning and his ideals. In a word the scholar may
at a great price become a statesman. When this occurs, whether
on a great scale or on a small one, whether at court or in a
village school, we have at last a solution of the ancient problem
of theory and practice.

ON THE ATTRIBUTES OF THE SENSATIONS.

BY PROFESSOR MAX MEYER,
University of Missouri.

When I was a small boy, I heard someone talk about sensa-
tions and their classification. He said that any sensation was a
sensation of one of the five senses, and that each sensation had
a quality and an intensity. This is the classification developed
by the demand of everyday life. What is or should be the
attitude of the psychologist to such a classification ?

When I was a little older and received some instruction in
psychology, I was told that it was not sufficient to distinguish
merely quality and intensity of a sensation, but that there were
two other attributes of a sensation, duration and extent, but that
some sensations had no extent; besides I was told that there
were probably more than five senses.

In recent years I read some books and magazine articles
written by professional psychologists, criticising the theory of
attributes and elements of consciousness. The criticism, how-
ever, was either purely negative or proceeded in a direction
which reminded me too strongly of the psychology of a hun-
dred years ago. It seems to me that the psychologists inter-
ested in this matter are not sufficiently aware of the fact that a
scientific terminology can never mean anything but what we
agree to mean by it ; that a scientific term does not lose its use-
fulness by being used in daily life in a sense disagreeing with
its scientific meaning.

Some principles of classification of psychological elements
are rejected by some critics because the principle is said to be a
physical or physiological or epistemological, not a psycholog-
ical one. I am unable to appreciate in this connection an argu-
ment which is based on the distinction of different special
sciences, /'. e., on a terminology. The terminology distinguish-
ing special sciences is as far from being an absolute truth as any

83

84 MAX MAYER.

more particular terminology. The only criticism of a classifi-
cation of elements of consciousness which I am willing to ad-
mit, is an inquiry into the scientific usefulness of the proposed
classification of elements.

It is a very serious mistake, from the scientist's standpoint,
first to adopt a certain terminology, and then to collect all those
facts which fit into this terminology, but neglect all those facts
of the same province of experience which do not fit into the
adopted terminological system. And yet I am inclined to say
that this mistake is almost without exception made by those who
write about psychological terminology.

It cannot be enough emphasized, that the principle accord-
ing to which we call a certain fact of experience either a single
sensation or an attribute of a sensation, must permit a modifica-
tion of our particular terminology whenever the progress of our
knowledge of facts demands an adaptation of our terminology
to these facts. However, some discussions of the terminologi-
cal problem of sensations and their attributes give one the im-
pression that the author was searching for such a thing as an
absolute truth. The problem, if it is to be a scientific problem,
is by no means this: what are the attributes of sensation, but
this : ivhat facts of experience should we describe under the
•present conditions of our knowledge by the help of the term
' sensation ' and what facts by the help of the term ' attribute' ?
If the progress of our knowledge demands it, the terminology
not only may change, but even must change. But, while the
principle used for establishing a terminology must permit in the
future the adaptation of our thought to those facts which will be
discovered in the future, the terminology itself must be the
means of describing our knowledge of actual facts of the pres-
ent, not of merely possible experiences of the scientist of the
future. We must not describe the known in terms of the ab-
solutely unknown. Therefore the principle according to which
we speak of sensations and their attributes must not make use
of any hypothesis. A hypothesis is neither at present a fact of
experience nor are we sure that it will ever be one in the future.
What science needs, is not a possible terminology of the future,
but an actual terminology adapted to the facts which we know

ON THE ATTRIBUTES OF THE SENSATIONS. 85

at the present time. For this reason I reject Mtinsterberg's use
of his hypothesis of a relation between a psychological atom
and the function of a single nerve cell. This relation is at the
present time perfectly hypothetical, indefinite, meaningless, un-
supported by any particular physiological knowledge, and for
this reason cannot be used as a principle of the terminology in
the science of the present time.

The principle which I propose in the following agrees to
some extent with Miinsterberg's views concerning the * elements
and atoms of consciousness.' But only to some extent. I be-
lieve that the application of my principle to the facts brings
about a terminology, which is more useful than Miinsterberg's
terminology, because it incites us to look first for the facts and
then for the terms in which to describe them, whereas Miin-
sterberg, it seems to me, offers a ready-made system of terms
and merely invites us to find the facts which fit these terms.1

Among the principles proposed for the classification of ele-
ments of consciousness is one which is particularly favored by
psychologists, namely the principle of independent variability ;
e. g-.j a tone may retain pitch and duration, but vary its inten-
sity. The reason why I am opposed to this seemingly very
beautiful principle (of independent variability) is, that it does
not possess scientific usefulness. I shall prove this at once by
referring to a special case. Stumpf pointed out many years
ago, that a single tone has the attribute of quality as well as
of pitch. This means of course neither more nor less than that
the tonal judgments which remain if the judgments of intensity
and duration are neglected, should be divided into two distinct
classes. For these two classes we must have names, scientific
terms, and the best terms seem to be pitch and quality. This
division is made for the single reason of its usefulness for the
description of the facts. It is self-evident, that, the more super-
ficial a psychologist's knowledge of the facts of hearing, the
less he will see the scientific usefulness of the division of judg-
ments into the two classes mentioned. That so few psychol-
ogists have paid any attention to Stumpf's division shows how

1 1 have in mind particularly his ' drei Qualitatenreihen, der Art, der Starke
und der Selbstandigkeit.' Grundzuge der Psych. , p. 285 ff.

86 MAX MA YER.

little interest the average psychologist takes in the facts of
hearing. I am sure that Stumpf s division is exceedingly useful
for a scientific description of the facts.

However, if we accept independent variability as the prin-
ciple of distinguishing attributes of sensations, then this distinc-
tion between pitch and quality of a single tone is impossible
since a stimulus of a given vibration frequency, producing a
single tone, does not permit to vary the pitch independently of the
quality, both being dependent on the vibration frequency. But
what should determine our acceptance of a certain terminology,
its merely formal beauty, or its scientific usefulness? I confess
that I do not hesitate to decide in favor of the latter. Inde-
pendent variability may be a very beautiful principle of classi-
fication, but it has ceased to be useful for the description of the
facts. It is scientifically sterile.

I shall use in the following the terms ' single sensation ' and
* attribute of a sensation,' and I shall call a single sensation a
representative of an ' element ' of consciousness, an attribute,
an * atom ' of consciousness, selecting the latter two words as
proposed by Miinsterberg. Now, I cannot in advance say
what I am going to mean by these four expressions ; that would
be unscientific. Their meaning can become clear only by their
application to the facts. But I will tell at once some things
which I do not mean by these expressions.

1. I do not mean, when I call a single sensation an element
of consciousness, and an attribute of a sensation an atom of
consciousness, that there are no other elements and atoms of
consciousness but sensations and their attributes. But, these
are the only elements and atoms of consciousness in which I
am interested at this time. There is no law prohibiting the
application of these terms, elements and atoms of consciousness,
to other facts of psychological experience (* affectional ' and
' transitional ' elements and atoms), provided that this appli-
cation is scientifically useful.

2. I shall limit the present discussion to peripherally aroused
sensations. The reason why I do this is that otherwise I fear
I could not make the matter perfectly clear in a brief paper.
I do not deny that there are also centrally aroused sensations.

ON THE ATTRIBUTES OF THE SENSATIONS. 87

I doubt, however, if this fact could have any influence on the
proposed terminology.

3. I do not admit at all as an argument for or against a
scientific terminology its agreement or disagreement with the
terminology of life. I do not believe that because * the words
element and attribute (or aspect) are vague and meaningless ' in
our daily life, they must for this reason be vague and meaning-
less to the scientist, the psychological theorist. If the theorist
would only make up his mind to mean by these words definite
facts and nothing suggested by these words in the affairs of the
day, these words as psychological terms would be perfectly
clear. I shall not begin, however, with a definition of these
terms according to the rules of formal logic. I merely wish to
say, that I regard an element as simpler in a certain way than
a complex, and an atom as simpler in a certain -way than an
element. In what way simpler? This question cannot be
answered in advance. We shall find out in what way as soon
as we use the terms for the description of the facts. If we
stated in advance what we mean by * simpler in a certain way,'
we should be compelled to adapt the facts to the terms. But,
as scientists, we wish to adapt the terms to the facts and to
mean by these terms, whenever we use them in this special
science, nothing whatever but these facts.

I shall now proceed to an analysis of a peripherally aroused
state of consciousness (affectionally neutral) and state the prin-
ciples according to which I propose to classify the theoretical
constituents of the actual complex state. I shall use the words
' simplification' and * elimination ' without meaning that elimi-
nated constituents are annihilated, but merely that they are
practically pushed beyond the threshold of psychological effec-
tiveness. They may be entirely annihilated, but the word
elimination shall not imply this. In the same manner I mean
by simplification, that practically our given consciousness must
be regarded as simpler, as less complex than another given
consciousness, what we mean in life by speaking of concentra-
tion of attention contrasted with diffusion of attention. I have
to state this in order to prevent any reader from applying argu-
ments of formal logic to problems of scientific terminology.

88 MAX MEYER.

A peripherally aroused complex state of consciousness can
be practically simplified :

1. By either simplification of the objective conditions or con-
centration of attention, with similar results. If there is such a
parallelism of effect, we speak of single sensations, of elements
of consciousness, in accordance with the classes of judgments
directly resulting from this simplification.

2. By concentration of attention only (not also by a simplifi-
cation of the objective conditions). In this case we speak of
attributes of a sensation, of atoms of consciousness, in accordance
with the classes of judgments directly resulting from this sim-
plification.

[3. By an alteration (but not simplification) of the objective
conditions. This case is insignificant for psychological ter-
minology.]

It is necessary to emphasize three points : (i) I did not speak
of a simplification of a complex state of consciousness by a
mere alteration of the objective conditions, but of simplification
by a simplification of the objective conditions. I did not speak,
however, of ' physical and physiological elements,' because our
views concerning such elements are too variable. No two
scientists would probably be found in perfect agreement on this
matter. But if of two facts, given in the simplest description
possible the one is simpler than the other or not, on this ques-
tion it is comparatively easy to agree. (2) When I speak of the
objective conditions (the stimulus) of a fact of psychological
experience, I mean those particular conditions which permit
the simplest and clearest definition of a stimulus producing the
effect in question. (3) When I speak of objective conditions, I
mean here, temporarily, physical or exterior chemical condi-
tions. Of course, physiological conditions of the nervous sys-
tem (like Munsterberg's ' function of a single nerve-cell ') are
also in a certain sense objective, but we cannot use them, be-
cause we cannot define the known by the unknown, because
physiological processes in the brain, to which we are accus-
tomed to refer, are as yet almost unknown, chiefly hypothetical,
not admitting any definite answer to the question, whether one
of them is simpler than another one. However, when phys-

ON THE ATTRIBUTES OF THE SENSATIONS. 89

iology shall have enabled us to answer this question, there
will be no objection to using also the physiological simplicity
for the classification of elements of consciousness.

I now have to apply the above terminological principle to all
the sensory facts known (so far as this is possible within a lim-
ited number of pages). I have to show what sensations and
what attributes we have to distinguish according to the princi-
ple. And in cases where I have reason to believe that the
reader may hesitate to admit the scientific usefulness of the re-
sulting classification, I have to point out wherein its usefulness
consists.

Let us imagine the peripherally aroused state of conscious-
ness of a very young infant, exposed to all the normal stimuli
of the average day. There can be no doubt that in some re-
spect this consciousness is much more complex than that of the
average adult under similar objective conditions. Of course,
the adult's consciousness is complicated by a great number of
associated ideas, of memory images, of centrally aroused sensa-
tions. But these we leave out of consideration. The complex-
ity of the infant's consciousness which I have in mind is re-
ferred to by some psychologists, when they call the functional
aspect of the infant's consciousness by the name of ' fusion,'
that of the adult's consciousness by the name of ' analysis.'
We may express this fact also by saying that the infant has not
yet learned to concentrate his attention.

Now let us see what judgments may directly result from
either simplification of this assumed consciousness of ours by
concentration of attention, or similarly from simplification of
this consciousness by simplifying the objective conditions. We
may imagine that all auditory stimuli are kept from the sub-
ject. Also all gustatory and olfactory stimuli. We may imag-
ine the subject as floating in a medium so as to prevent all
cutaneous stimulation. And so let us remove all stimuli but
those which we call optical. The practical difficulty of such an
experiment must not be offered as an objection. There is no
doubt that we can approach toward the limit (this word is here
used as it is used by the mathematicians) where only optical
stimuli are active. Every physicist will admit that the description

90 MAX MEYER.

of this limit is very much simpler than the description of the
condition of stimulation from which we started our discussion.
And every psychologist will admit that the accompanying con-
sciousness is simpler. But we have to continue our process of
simplification. Let us imagine that the optical stimulation was
that of a winter landscape, made up of nothing but dark trees,
white snow between, and the blue sky above. Or let us imag-
ine that our eye is exposed to nothing but the blue sky and an
infinite snow-covered plane ; or to the blue sky alone. Is not
a complete mathematical description of the stimulus producing
the latter experience simpler than the description of the stimu-
lus in the other cases ? And is not our consciousness simpler too ?
Nothing can directly result from it but a judgment concerning
this experience of sky-blue. A similar simplification of con-
sciousness can, of course, result from mere concentration of
attention. Without simplifying at all the original complex stim-
ulation, the adult mind may pay attention to the color of sky-
blue alone and pronounce a judgment thereon.

The problem now before us is this : can this parallelism of
the two methods of simplifying our consciousness be traced
farther than to this point? Let us agree that the only im-
pression which concerns us be the blue sky of the winter land-
scape. Can we not pronounce more than one class of judg-
ments concerning this sky-blue? Can we not say that it is
large (compared with the white and black remainder below), or
that it is bright, or that it is blue? No doubt, we can simplify
our consciousness by concentration of attention in such a way
that only one of the judgments mentioned directly results.
But can we also simplify the objective conditions so that the
directly resulting judgment can be one of, say, brightness
only? I shall show in detail that such a parallelism no longer
exists here.

How can we determine by the simplest physical definition,
under a given condition of sensitiveness of the eye, what we
mean by blue, or violet, or a neutral gray, or black? In a
more primitive stage of physical science the physicists used for
this definition the infinite number of homogeneous lights of the
common solar spectrum. We can use this primitive physical

ON THE ATTRIBUTES OF THE SENSATIONS. 91

definition of a stimulus producing a certain visual experience,
if we intentionally or unintentionally overlook its imperfection,
for the application of the proposed principle of psychological
classification. A difficulty arises in this case for the psycholo-
gist only with respect to the colors, which are not in the spec-
trum, i. £., the purples which are more reddish than the violet
of the spectrum. I shall actually show how we can use this
definition to this extent. However, those who are familiar with
the multitude of facts which we call physiological optics, will
urge at once, that the only objective definition of any visual
sensation, which is satisfactory to the physicist, is the definition
by means of an equation containing as three constants three
selected homogeneous lights. I copy a physical definition from
Helmholtz's Handbuch der Physiologischen Opttk, p. 341 :

** Wenn wir die Grundfarben und ihre quantitativen Einheiten
7?, G, V gewahlt haben, dann kann der physiologische Ein-
druck jeder andern Farbe F dadurch vollkommen beschrieben
werden, dass wir sagen, sie sehe so aus, wie eine Vereinigung
von so und so viel Einheiten 7?, (9, und V. Also, wenn wir
mit x, y, z Zahlen bezeichnen,

I shall now first use the physical definition of a stimulus
(producing any visual sensation) by means of the infinite number
of homogeneous lights. We have to answer the two questions :
(i) Is the sensation produced by homogeneous light of a certain
point of the spectrum (say, blue) simpler than the sensation pro-
duced by homogeneous light of another point of the spectrum
(say, violet)? The answer to this question must be that the sen-
sation blue is simpler in a certain way than the sensation violet.
For we can judge violet with respect to its duration, its extent,
its brightness, its bluishness, and its reddishness. Each of these
judgments implies a simplification of our state of consciousness
by concentration of attention. But we can judge the sensation
blue only with respect to its duration, its extent, its brightness,
and its bluishness. Blue therefore is simpler in this way just
mentioned than violet. (2) But can we produce a simplification
of our consciousness with a similar result also by simplifying

92 MAX MEYER.

the objective conditions? The answer to this second question
must be negative. The homogeneous light producing the sen-
sation blue is not simpler than the homogeneous light producing
the sensation violet. We have to say, therefore, in accordance
with our terminological principle, that violet is not a sum of sen-
sations, but a single sensation in the same sense, in which blue
is a single sensation. Does not the reader instinctively agree
with this result?

We must now answer the same two questions with respect
to ' blue ' and ' gray ' : (i) That the sensation blue is in a certain
way less simple than the sensation gray, cannot be doubted.
For we can produce by concentration of attention each of the
judgments concerning blue which we can produce concerning
gray, and one in addition, that of bluishness. (2) The physical
stimulus, however, which the physicist would use for the defini-
tion of blue, is simpler than the stimulus of gray. The stimulus
of gray would have to be defined as the stimulus of blue plus a
certain additional stimulus, or two additional stimuli, or more
than two additional stimuli; i. e., the stimulus of gray is the
sum of two, three, or more homogeneous lights. We do not
have, therefore, in this case, a physical simplification parallel
to the simplification by concentration of attention. Therefore
we must say, in accordance with our terminological principle,
that the sensation blue is a single sensation in the same sense in
which the sensation gray is a single sensation. I should think
that the reader will agree with this result.

Let us further answer the same two questions with respect
to white and black : There can be no doubt that the objective
condition of white is less simple than the objective condition of
black, the latter being merely negative. However, black does
not permit any simplification by mere concentration of attention,
which white would not permit also ; black is not simpler than
white. There is no parallelism of the two methods of simpli-
fying our consciousness according to our principle.

The result of this application of our terminological principle
to the facts which we know at present, is then this : The prin-
ciple requires us to call any visual sensation which is uniform
over a certain area of the field of vision, a single sensation, not

ON THE ATTRIBUTES OF THE SENSATIONS. 93

a sum of sensations. We must speak of two or more visual
sensations only in case we have two or more different areas
within the field of vision.

The next question to answer is now this : Which classes of
judgments are to be distinguished as directly resulting from a
further simplification of single visual sensations by concentration
of attention only? /. e., which are the attributes of visual sen-
sation ? The classes of judgments which we have to distinguish
are so far as my knowledge reaches, the following : duration,
extent, brightness, bluishness, yellowishness, greenishness, and
reddishness. These are the seven attributes of visual sensation,
in accordance with the terminological principle and the present
condition of knowledge.

I do not fear that anyone will raise the objection that the
element of consciousness which we call visual sensation, cannot
have seven attributes, since it is an a priori truth that any and
each sensation has four attributes (with the exception of some
which have only three). If a man pretending to be a scientist
would argue in that way, I should simply leave him alone. I
fear, however, that some psychologist might object to the above
result, because no particular sensation ever possesses all those
seven attributes. Yet this is not a scientific objection.

Does not the chemist call gold a chemical element, although
it is sometimes solid, sometimes liquid, but never solid and liquid?
Why should any particular experience of an element of con-
sciousness be required to possess all the possible attributes of
this element?

And further, this is not a singular fact, peculiar to visual sen-
sation. We shall later see, that in the auditory element of con-
sciousness too we find some particular auditory sensations which
do not possess all the attributes of the auditory element.

I believe that it is one of the fundamental tasks of experi-
mental psychology, to determine the laws governing the mutual
relation between the attributes of a single sensation as well as
the mutual influence of the attributes of several sensations.
Some of the most important laws of the attributes of visual sen-
sation are these :

i. The attributes yellowishness and bluishness do not coexist
in a single sensation.

94 MAX MEYER.

2. The attributes reddishness and greenishness do not coexist
in a single sensation.

3. The attributes yellowishness, bluishness, greenishness,
and reddishness may be absent altogether in a single sensation.
The other three attributes are always with a certain vividness
present in the conscious experience of a visual sensation.

Facts like these are not objections to the proposed termin-
ology, but simply the natural laws of the attributes of visual
sensation.

As children in school, we used to wonder why our teacher
told us sometimes that a solution of an arithmetical problem was
wrong in spite of the fact that the result was right. Now in the
above classification of the attributes of visual sensation, I believe
that the result is right, i. <?., scientifically useful, in spite of the
fact that I made tivo mistakes: (i) I started from a physical
definition which is too primitive, and (2) I used the trick of
omitting from discussion the purples which are reddisher than
the violet of the spectrum.1 I did this because psychologists
(and physicists) sometimes use and have to use this primitive
physical definition in elementary instruction ; when the student
hears about this matter for the first time. I shall now apply
my terminological principle to the facts while using the more
perfect physical definition of Helmholtz, as above stated. We
shall see that this results exactly in the above classification
of attributes, so that the above classification is indeed the one to
be accepted.

The equation F ' = xR + yG + z V can be simplified in no
other ways than by rendering either one or simultaneously two
or all three of the numbers x, y and z equal to zero. Now, let
us apply this fact to some particular experiences, the experiences
of J?, G, blue, purple, gray and black. jR, G, and purple
permit each five of the seven classes of judgments (above
mentioned as attributes) by means of concentration of attention.
E. g., R permits the judgments of duration, extent, brightness,

1 Someone might point out (justly) that a parallelism of simplification of
consciousness does exist in the case of purple and blue (' Urblau '), since the
latter can be defined by one homogeneous light, the former not by less than
two. When we use the scientifically more perfect physical definition of Helm-
holtz, no such difficulty of an apparent contradiction arises.

ON THE ATTRIBUTES OF THE SENSATIONS. 95

reddishness and yellowishness. Blue permits only four. But
the sum on the right side of our equation contains only one
member in the case of R or G, at least two members (F «= xR
+ z V\ but it is improbable that the brightness should happen
to be such, that/y is equal to zero) in the case of purple, and at
least two members (probably again all three) in the case of blue.
It is impossible, therefore, to speak in these cases of a parallel
simplification of our state of consciousness by either simplifi-
cation of the objective conditions or concentration of attention.

Further, the experience of gray permits only three judg-
ments resulting directly from a mere concentration of attention ;
the experience of blue permits four ; the experience of V per-
mits five. But the physical definition by Helmholtz's equation
is simplest in the case of V\ less simple in the case of blue;
and least simple in the case of gray. There is no parallelism
of the sort we are looking for.

Further, the experience of gray permits by concentration of
attention three classes of judgments, concerning duration, ex-
tent and brightness. So does the experience of black. That
the stimulus in the case of black is defined in the simple way
f=O, does not establish any parallelism of the sort in question.

We must repeat therefore what we stated above :

Any visual sensation which is uniform over a certain area of
the field of vision, must be called a single sensation, not a sum
of sensations. The attributes of visual sensation are : duration,
extent, brightness, bluishness, yellowishness, greenishness, and
reddishness.1

Let us now consider a case in which the objective conditions
of stimulation have approached the limit at which we need not
speak of any but acoustical stimulation. Let us imagine that
the stimulus consists of three sine waves of the frequencies 300,
400 and 500. We say that we hear several tones. No one

1 1 have been criticized by Mrs. Ladd-Franklin for saying : Die Heringsche
und die Helmholtzsche Theorie erganzen sich gegenseitig. I did not mean by
this that both of them were psychological theories of color-vision. This name
can be given only to the Hering theory. What I wished to express by the
words above quoted is my conviction that we cannot get along in psychology
without the Helmholtz theory. And I hope to have made clear now, in what
sense this is true.

9 MAX MEYER.

denies that a stimulus consisting of only one of these sine
waves is physically simpler. We hear in this case one tone
only. We can produce a similar result, without simplifying
the stimulus, by merely concentrating our attention. We con-
centrate our attention on one of the several tones ; and only
judgments concerning this one tone result directly. A further
simplification of our state of consciousness by simplifying the
objective condition is impossible, since one sine wave is as
simple as any other. We therefore say, that the tone we hear
is a single tone, not a sum of sensations.

But a further simplification of our state of consciousness
by concentration of attention only is entirely possible. Four
classes of judgments may directly result : judgments concern-
ing duration, intensity, quality,1 and pitch. These are there-
fore the four attributes of auditory sensation.

Since the average psychologist takes so little interest in
auditory sensation, I shall have to point out in some detail, in
what respects it is useful to distinguish judgments of quality
and judgments of pitch as two classes. I shall state a number
of facts, most of which seem to be quite unrelated facts as long
as we fail to make the above distinction but become interesting
as particular cases of more general facts as soon as we refer
them to the one or the other class of judgments.

There are two reasons why this distinction of the attributes
of quality and pitch of auditory sensation is not generally
made. One reason is that this distinction does not agree with the
old-fashioned, but nevertheless absurd theory, that any sensation
has either the four attributes of quality, intensity, extent, and
duration, or the three attributes of quality, intensity, and dura-
tion. The other reason is that it does not agree with the prin-
ciple of independent variability, quality as well as pitch being
dependent on the vibration frequency ; but psychologists should
keep in mind that the terminological principle of independent

ll use the term 'quality,' because in English this is the term which is
daily used by every maker of musical instruments when he is conscious of
that sort of judgment. The reason why the psychologists do not like this
term, is no other than its not fitting into their artificial systems. However,
it is the best term I know. In German we should use Stumpf 's term ' Tonfarbe '
( ' Klangfarbe ' for a mixture of qualities).

ON THE ATTRIBUTES OF THE SENSATIONS*. 97

variability is not a divine revelation. It has no claim for exist-
ence beyond its usefulness.

1. We can theoretically understand the difference between
a * pure noise ' and a * tone,' if we regard a tone as an auditory
experience under such conditions material for the function of
attention, that all four classes of judgments may directly result;
a pure noise as an auditory experience made up of brief tone
sensations under such conditions for the function of attention,
that no judgments concerning pitch can result. /. e., we may
judge, that this noise is * higher ' (referring to the mixture of
qualities, the mean quality) than another noise ; but we cannot
say that this noise is in unison with the other noise, or that it is its
fifth, or a mistuned fifth. More details about the physical con-
ditions of stimulation producing a pure noise are to be found in
my paper ' Zur Theorie der Gerauschempfindungen.'1

2. In demonstrating to students very low and very high
tones, say below 30 and above 8,000 vibrations, I have found
that quite commonly my hearers refuse to call these sensations
* tones.' They incline to call them noises. This is not won-
derful to the psychologist who is aware of the fact, that these
tones do not possess the attribute of pitch (they cannot be music-
ally employed), while they possess the attributes of intensity,
duration and quality. Such a tone may be said to be lower or
higher (referring to quality) than another tone, but it cannot be
said to be its third or its fourth. The difference between the
experience of such a tone and a pure noise consists merely in the
fact that theoretically we regard the latter's quality as a mixture
of qualities, the former's quality as a single quality.

3. The theoretical distinction between tone and noise is
entirely different in kind from the theoretical distinction between,
say, blue and yellow. We may compare, in some respects, a
tone with a uniformly colored piece of paper, a noise with a
painter's palette, or rather with a flickering color-wheel. The
only respect I can see, however, in which the distinction of tone
and noise can be identified with the distinction of blue and yel-
low, is its practical significance. To distinguish between tone
and noise is of equal practical importance in life as to distin-

1 Zeitschrift fur Psychol. u. Physiol. d. Sinncsorgane, 31, p. 233.

9° MAX MEYER.

guish between yellow and blue. But who ever thought of bas-
ing the terminology of pure science on such a consideration !
Why not say, then, that the distinction between a horse and a
steam engine is the same as the distinction between ice and
glass? With respect to practical importance it certainly is.
But scientifically?

4. The technical problem which the organ builder has to
solve is this : He is required to construct single sources of sound
which impress us as possessing the same pitch, but different
quality. He solves this problem by making use of this impor-
tant law of auditory attributes in a plurality of auditory sensa-
tions : We can easily pay attention simultaneously to several
pitches so that several judgments directly result, but with great
difficulty only to several qualities so that several judgments
directly result, unless we pay attention to their corresponding
pitches too. And we can easily concentrate our attention on a
single pitch, but not on a single quality ; if we try to pay atten-
tion to the corresponding quality, the resulting judgment of
quality is as a rule a judgment determined by all the qualities
of the several tone sensations present. This is the law which
the organ builder uses in order to obtain the desired effect (but
which seems to be unknown to most psychologists, who take
only a slight interest in the fundamental laws of hearing). By
using the physical fact of partial vibrations of elastic bodies,
he combines several tone sensations of which one has a far
greater intensity than the others. This intensity attracts our
attention to this one sensation more than to the others, and we
judge to hear one pitch because our attention is concentrated
on one pitch. Our judgment of quality, however, is, we may
say, an auditory illusion^ entirely comparable with a certain
class of geometric-optical illusions.1 It is self-evident that the
mellowest quality which an organ builder can give to a single
source of sound is the quality of its fundamental tone (without
overtones), since an elastic body may produce overtones, but
does not produce undertones.

1 Compare Schumann, Zeitschrift f. Psychologic u. Physiol. d. S. v. 24, p.
7, 190x3. " Die eigentlich zu vergleichenden raumlichen Grdssen bestimmen
nicht allein das Urteil, sondern die Ausdehnungen benachbarter Eindriicke
wirken mit."

ON THE ATTRIBUTES OF THE SENSATIONS. 99

5. Musical effects depend on the hearer's paying attention
to the pitches. Our congenital ability in this respect differs in-
dividually. There are some individuals, however, who are not
normally affected by musical relationships in spite of a maxi-
mum practice and effort to pay attention to whatever there may
be in their auditive consciousness. We describe their condition
by saying that they are ' pitch-deaf ' (similarly as we call some
individuals ' green-blind '). We are all of us pitch-deaf for the
lowest and highest auditory sensations (similarly as we are all
of us green-blind on the peripheral parts of our field of vision).
For the individuals mentioned auditory sensation possesses only
three attributes, duration, intensity and quality. To their class
belong probably those rare cases, reported in psychological
literature, of individuals who could sing a tune in the key in
which they had learned it, but not in another key. They sang
by a memory for quality, not by a memory for pitch. Melodic
relationships are the relations (on a certain unknown physio-
logical basis) of the pitches of the auditory sensations.

We need not wonder why the attribute of pitch in auditory
sensation is sometimes lacking, while the attributes of intensity,
duration and quality are always present. We have found
analogies in visual sensation.

The terms ' high ' and ' low ' are ambiguous. When we
speak of tones as being high and low, we refer sometimes to
pitch, sometimes to quality. This ambiguity is the cause of
some disagreements in experimental practice as well as in
theoretical discussion.

6. The ' absolute memory for pitch ' is actually much less a
memory for pitch than a memory for quality. It is a well
known fact that quite frequently individuals are able to name
the tones of a certain musical instrument (e. g., the piano), but
not those of another musical instrument (<?. g., the human voice).
If we call this memory an absolute memory for quality, as we
have a right to do from other reasons too, the fact mentioned
does not require any particular explanation at all, since the
* mean quality ' of the sum of auditory sensations produced by
singing c is very different from the mean quality of the sum of
auditory sensations produced by striking c on the piano. The

100 MAX MEYER.

pitch, if we pay attention to this attribute, of the strongest sen-
sation in either case is the same, but the name c is absolutely
associated not with this pitch, but with the quality of the piano c.

The physiological processes in our nervous system, which
underlie the experiences of pitch and quality, are probably no
less different than the physiological processes underlying the
experiences of extent and brightness in visual sensation.

7. There has been a discussion in psychological literature
(I will mention only the names of Stumpf and Ebbinghaus)
whether we have a right to call a tone more similar to its Octave
than to its Third, after we have agreed to call a tone more
similar to another tone by three vibrations higher, less similar
to another tone by six vibrations higher. It is here as in so
many other cases in science, that a problem is discussed with
many arguments in this or that direction, while there is no prob-
lem at all. It is clear that there is no scientific problem of this
sort left for discussion as soon as we adopt the distinction be-
tween the two attributes of pitch and quality. Or should we not
have the right to say, that a circle is more similar to an ellipse
than to a rectangle, after we have agreed to call a circle of
medium gray more similar to one of dark gray than to one of
black?

The above facts seem to me to express the most important
natural laws governing the attributes of auditory sensation.

Let us now consider a case in which the objective conditions
of stimulation have approached the limit at which we need not
speak of any but gustatory stimulation. We can physically de-
fine any given gustatory experience by saying that it is the
same experience which we have when the stimulus is made up
of certain quantities of four substances in standard solutions, as
shown by the equation

G = xAc + yBi + zSa + uSiv,

e. g., of tartaric acid, quinine, common salt, and sugar. By
concentration of attention we can simplify such a state of con-
sciousness so that only the judgment sweet, none of the judg-
ments sour, bitter, or salt, directly results. The same is possible
by simplifying the stimulus according to the equation.

ON THE ATTRIBUTES OF THE SENSATIONS. IOI

Can we further simplify our state of consciousness which we
call sweet by concentration of attention? We can; and two
classes of judgments may directly result, judgments concern-
ing more or less sweetishness and judgments concerning dura-
tion. But is also a parallel simplification possible by simplifying
the stimulus according to our equation? This is impossible, for
the equation does not become simpler by a variation of u. We
must say, therefore, that the state of consciousness produced by
a certain solution of sugar is a single sensation, and that its
duration and its intensity are its two attributes.

The sensation of sweetness, then, has the attributes of inten-
sity (sweetish, sweet, luscious) and duration, but not of ' quality.'
Sweetness is not an attribute of sweetness. Else, we should
have to call visuality an attribute of visual sensation, auditiveness
an attribute of auditory sensation, acidity an attribute of sour-
ness. We need not say that what distinguishes a certain single
tone from the visual sensation of the blue sky, is their different
' quality,' pitch and blueness. For we have just as much right
to say, that what distinguishes them, is the tone intensity on the
one hand and the brightness on the other hand. This con-
sideration seems to have been the basis of Miinsterberg's dis-
tinction of ' drei Qualitatenreihen, der Art, der Starke, und der
Selbstandigkeit.' These * Qualitatenreihen ' seem to be a very
simple expression of the laws of sensation. However, Miin-
sterberg's terminology, as well as the terminology which speaks
of the two attributes of quality and intensity of any and each
sensation, makes us believe that simplicity of psychological law
reigns over a province of facts, where actually there is a great
diversity of laws. Simplicity of description is the aim of
science. But the pretension of simplicity, where there is no
simplicity, by means of adapting the facts to the terms is worse
than no science. I do not see the necessity of having a general
term (' quality ') for sweetness, visuality, auditiveness, etc. Let
us simply speak of different departments into which our sensa-
tions as such, without respect to their attributes, may be classi-
fied. And let us say, that some four of [these sense depart-
ments, namely sweetness, sourness, bitterness and saltness, are
in some respects (with respect to certain reactions of ours) so

102 MAX MEYER.

closely related, that we may call these four departments by the
common name of a sense of taste. If we do this, we do not
make the mistake of asserting simplicity where there is none
and using complicated description where the facts are simple.

One fact concerning the sensations of sweet, sour, bitter and
salt deserves to be mentioned in this connection. Those who
speak of ' the quality of taste ' and « the quality of color ' and
other ' qualities ' have to admit the peculiar fact that most of
their ' qualities ' are of the form of a continuous series, while
' the quality of taste ' consists of four discrete points only.
Those who are accustomed to mathematical thought must sus-
pect under such circumstances, that the distinction of these
' qualities ' of the psychologists is not the result of the consistent
application of a definite scientific principle. And indeed it is
not. Those ' qualities ' are the outcome of practical necessities,
not of scientific thought. They are mainly those attributes of
the several sensations, which happen to be particularly im-
portant in man's struggle for life. Of course, I do not deny
that some one attribute of a sensation is practically more impor-
tant than others. But I do not see why such a fact should be
expressed by the terminology of a pure science. Unfortunately
the number of psychologists is still too great who regard psy-
chology as an art rather than as a science.

I might go on and apply the terminological principle to all
the other sense experiences. I merely fear to tire the reader.
The following table shows the result of the application of the
terminological principle to the facts, so far as the facts are at
present generally agreed on by the psychologists. If our
knowledge of the facts progresses, the application of the prin-
ciple to the facts may lead to the addition of elements and
atoms or to changes in what I now propose to call in each case
a particular sense department. However, a principle which
did not permit such modifications of our thought, would not be a
scientific principle. By ' element,' of course, I do not mean
the special sensation of a special individual at a special time,
but an abstraction. Similarly as the * element gold ' of the
chemist does not mean the liquid gold in a certain smelting
furnace or the gold in a gold bath of a certain photographer,
but an abstraction.

ON THE ATTRIBUTES OF THE SENSATIONS.

TABLE OF THE SENSATIONS.

103

Sense Departments Distin-
guishable at Present. (Per-
ipherally Aroused Elements
of Consciousness.)

Groups into which
some Sense Depart-
ments are Collected
at Present. (Senses).

Attributes Known at Present. (Possible
Atoms of Bach of the Peripherally
Aroused Elements of Consciousness.
Their Number Differs in the Several
Elements.)

i. Visual sensation.

Vision.

Duration, extent, brightness, blu-

ishness, yellowish ness, reddish-

ness, greenishness.

2. Auditory sensation.

Hearing.

Duration, intensity, quality, pitch.

3. Sweet sensation.

Duration, intensity.

4. Sour sensation.
5. Bitter sensation.

• Taste-sense.

Duration, intensity.
Duration, intensity.

6. Salt sensation.

Duration, intensity.

7. Warmth sensation.
8. Cold sensation.

Cutaneous

Duration, extent, intensity.
Duration, extent, intensity.

9. Pressure sensation.

sense.

Duration, extent, intensity.

10. Pain sensation.

Duration, extent, intensity.

n. Muscular sensation.

Duration, intensity.

12. Tendinous sensation.
13. Articular sensation.

Organic

Duration, intensity.
Duration, intensity.

14. Sexual sensation.

sense.

Duration, intensity.

15. Static sensation.

Duration, intensity.

16. Olfactory sensation x.

Duration, intensity. Further at-

tributes ?

17. Olfactory sensation y.

Duration, intensity. Further at-

Sense of

tributes ?

1 8. Olfactory sensation z.

smell.

Duration, intensity. Further at-

tributes ?

19. Other olfactory sensa-

Duration, intensity. Further at-

tions ?

tributes ?

? Other sense depart-

?

?

ment*?

Some one may object to speaking of * visual sensation ' as
an ' element ' of consciousness, because ' blue 'is as * elemen-
tary ' as ' yellow.' To him I have to say, that a chemist speaks
of mercury as an 'element, although it may be harder soft or
liquid or gaseous. He calls mercury an element nevertheless.
Why should the psychologist not call visual sensation an ele-
ment in spite of the fact that it is sometimes blue and some-
times not blue ? The chemist justifies his terminology by point-
ing out the scientific usefulness of this particular usage of
language. And I believe that it is equally justifiable to speak
of visual sensation, without reference to any of its particular
attributes, as an element of consciousness. That the natural
laws of chemical elements and atoms should be identical with
the natural laws of the elements and atoms of consciousness, is
an arbitrary requirement. For scientific terms need never mean
anything beyond what we agree to mean by them.

AN INQJJIRY INTO THE NATURE OF
HALLUCINATION.

II.

BY BORIS SIDIS, M.A., PH.D.,

Director of the Psychopathic Hospital and Laboratory of the New York
Infirmary for Women and Children.

I.

A peripheral process often of a pathological nature, a state
of dissociation and a subexcitement of secondary sensory and
ideomotor elements constitute the main conditions of hallucina-
tions. The peripheral pathological process and the state of
dissociation are requisite to the formation of the hallucinatory
percept, while the content of such percepts are given by the
systems of sensory-motor and ideomotor elements. A periph-
eral process alone even if it be pathological in character does
not give rise to hallucinations. Similarly a state of dissociation
by itself or a state of subexcitement of secondary and repre-
sentative elements cannot give rise to hallucinations. It is only
when these conditions cooperate, it is only then that hallucina-
tions arise. The state of dissociation and that of subexcitement
of * central ' systems may be regarded as the ' central ' condi-
tions of hallucinations, while the peripheral process is the factor
that supplies to the systems the primary sensory nuclear ele-
ments round which the secondary elements crystallize and form
a hallucination.

States of dissociation, provided the other conditions are pres-
ent, are preeminently favorable to the formation of hallucinatory
percepts. In sleep, when the mind is immersed in darkness,
isolated isles of systems may stand out of this general night of
consciousness and give rise to dreams of various degrees of in-
tensity. Dreams are sleep hallucinations, while hallucinations
are waking dreams. Both hallucinations and dreams develop
under the same conditions of dissociation. The nature of

104

THE NATURE OF HALLUCINATION. 105

dreams and hallucinations are essentially the same. An isolated
dissociated system of secondary sensory and representative ele-
ments predisposed to function become awakened by a special
peripheral stimulus or by a summation of series of stimulations
and gives rise to hallucinations or dreams according to the gen-
eral state of consciousness, waking or sleeping. The halluci-
nation of the comparatively waking state stands out alone, it
remains more or less isolated and becomes obliterated by the gen-
eral inrushing flood of peripheral sensations and perceptions of
the waking consciousness. The dream is made up of a series
of hallucinations going sometimes to form a complicated halluci-
nation expanded into a whole life history. From this stand-
point we may say a hallucination is an abbreviated dream,
while a dream is an expanded hallucination.

In sleep the primary sensory nucleus of the dream halluci-
nation is supplied by the peripheral processes coming either
from external stimuli or from internal stimulations, from
changes taking place in the organism. The psychophysiologi-
cal threshold is raised in sleep, the resistance to the entrance of
sense impressions is increased, the rise being proportionate to the
depth of the sleep state. The peripheral sensory channels are
closed to external stimulations. External stimuli, however, as-
sail the peripheral sense organs from all sides and now and then,
whether on account of the intensity of the stimulus or of the
summation of a series of stimulations or of the temporary rise
of the sleep level and consequent fall of the threshold and de-
crease of resistance to the influence of external stimuli, sense
impressions force an entrance and awaken to activity some
slightly slumbering systems thus giving rise to the dream hal-
lucination. Under such conditions the sense impressions have
but small chance to awaken its appropriate systems and hence
become incorporated into any chance system they happen to
awaken thus giving rise to the phantastic combinations char-
acteristic of dream life. The sense impressions form the nu-
cleus around which cluster systems of secondary sensations and
representations all tinged with the sensory color derived from
the original primary nuclear sense impressions. The systems
of secondary sensory and representative elements once awakened

106 BORIS SID IS.

may go on expanding and developing, awakening other groups
and systems, assimilating them or being assimilated by them as
much as the nature of their content permits and being further
reinforced by incoming stimulations. During the whole course
of its expansion the aroused groups and systems maintain their
sensory or rather their perceptual character. For, if a system
is once awakened to activity, the threshold, the resistance to in-
coming stimulations is lowered and many more sense impres-
sions gain access to the functioning systems and become in-
corporated and assimilated. This assimilation of chance
systems and sense-impressions often give birth to highly
elaborated phantastic dreams and visions.

II.

Systems awakened by stimuli must have some relation of
familiarity to the nuclear sense impressions. If perception is to
take place, there must be some congruence between the sense
impressions and the stimulated systems. Only on such condi-
tions can assimilation take place. Similarly the awakened sys-
tems in sleep assimilate congruent sense impressions, the latter
becoming so transformed as to fit the system and the system is
modified by the incoming impressions. This congruence in the
dream state is often strained and remote and consequently often
of a phantastic and irrelevant character. Thus the taking off
a plaster may give rise to a dream of being skinned alive, or of
being scalped by an Indian. A change to an easier position
and a freer respiration may generate a dream of flying. In one
of my experiments of dream hallucination the uncovering of
the feet in a cold room gave rise to the dream of walking on
the frozen surface of a river and the impeded respiration awak-
ened the feeling of fear of falling into the water.

The internal sensations such as arise from the different func-
tions of the bodily organs are very important factors in the gen-
eration of dream haL icinations. Every one knows the fact that
indigestion often gives rise to nightmares and unpleasant
dreams, but not many realize the fact that coenagsthetic sensa-
tions, sensations that come from our internal organs play a very
important role in the production of dream hallucinations. The

THE NATURE OF HALLUCINATION. 107

circulation of the blood, the secretion of the various glands, the
peristaltic movement of the small intestines, the action of the
stomach, the changes in the muscles, the metabolism going on in
the various organs of the body, in the cells of the organism, all
these give rise to sensations which, though obscure and con-
fused, go to make up the general sense of organic life activity.
The sense of crenaesthesis may in fact be regarded as the basis of
our physical being or of our physical personality. A change of
this sense is frequently an important factor in the formation of de-
lusions, when mental life becomes dissociated and disaggregated.
Hypoaesthesia or anaesthesia of the leg, for instance, may form
the nucleus for the formation of the delusion that the leg is made
of glass or of putty or is totally gone. Anaesthesia of the body
or of the internal organs may develop the delusion of being
dead, the patient asking to be buried. Similar conditions are
also present in dream life. Changes of coenaesthesis play no
doubt an important role in the activity of the dream conscious-
ness. Changes in the various component elements that go to
make up the obscure but highly complex life of organic sensi-
bility affect profoundly the rich exuberant play of the. dream
consciousness. Since the channels to external stimulations are
closed, the coenaesthetic sensations that form the obscure basis
of waking consciousness become the sole possessors and guides
of whatever mental activity is present in sleep. These internal
sensations are woven by the dream consciousness into phantastic
images of all shapes and forms.

The dream consciousness presents many characteristics
found in states of mental dissociation and disintegration. Moral
tone is lowered, attention is greatly reduced, logical thought is
enfeebled and the sensory-motor and ideomotor elements are
thrown out of gear, often resulting in the formation of illusions,
hallucinations and delusions. In the dream state there is pres-
ent the mental degradation of dementia, the sordid delusions of
hypochondria and melancholia, the delirious states of mania, the
delusions of grandeur of general paralysis, and even the persis-
tent systematized delusions of paranoia. The dream conscious-
ness is extremely unstable, it forms no definite type of men-
tal disintegration and has no determinate course, it is extremely

108 BORIS SIDIS.

fluctuating in its states and its background is usually shifting
ceaselessly. From this standpoint it may be said that the dream
consciousness is a normal form of mental alienation and that
mental alienation is an abnormal form of dream consciousness.
A very characteristic diary brought to my notice in which a
retrospective and introspective account is given by a patient in
the normal condition of the experiences lived through in the
state of mental aberration opens with the suggestive title :
Memories of my Dream Life and with the following interesting
introductory remarks :

"Where shall I commence? How shall I begin to recall
and record this to me mysterious life I have been living? So
beautiful, so strange, and in some way so terrible. Yet I would
not forget, for it seems as though I must have been in commu-
nication with intelligences above — spirits of the air, if it were
possible.

" When did it commence? How long has it been with me?
are questions I cannot solve. For weeks before coming to the
hospital I must have been living this * ideal life ' as in an * ideal
world.' I have jotted down what I have thought, though they
are not one hundreth part of the thoughts which passed through
my mind during this strange time of dreaming." In one of my
cases of katatonia the frightful dreams of the year preceding
the disease became hallucinations of the maniacal stages and
appeared again as dreams during convalescence. The dreamer
dreams with, his eyes closed, the insane dream with their eyes
open.

In both the dreamers and the insane the disaggregated states
under the influence of external and especially of internal stimuli
give rise to illusions, hallucinations and delusions. Dissociated
states grouped round nuclei of primary sensations form the in-
ternal organizations of hallucinations and delusions so often
characteristic of dream life and insanity. Coenaesthetic sensa-
tions are important agents in the formation of insane delusions
and hallucinations, there are so many fermentation nuclei
among masses of dissociated states. Irritation of the ovaries
may in the insane awaken hallucinations and delusions of a
sexual character; constipation and heaviness in the intestinal

THE NATURE OF HALLUCINATION. 109

tract may generate delusions and hallucinations of rats and pigs
in the stomach ; rumbling in the stomach and the intestines may
give rise to the delusions and hallucinations of devils in the
body or of electric discharges of powerful batteries placed in
the abdomen. The hallucinatory delusional dream conscious-
ness works on similar lines — thus the first stages of migraine
with a heaviness of the head may in sleep give rise to the dream
hallucination of the head being opened, the brain swept away
and chalk substituted ; pain in the abdomen may form the
hallucination of mice gaining an entrance into the abdominal
cavity and gnawing at the intestines.

The difference between the walking life of the insane and
that of dream consciousness is the mode of activity, the dream
consciousness works in images, in sensory percepts, while in the
insane mind the activity is largely representative. This dif-
ference is due to greater dissociation present in dream con-
sciousness. The awakened dissociated systems in dream life
become tinged with a perceptual sensory color by the process
of absorption and assimilation of all the incoming sense im-
pressions. Pathological states of rapid mental dissociation, such
as the acute states of maniacal excitement or in states of psy-
chopathic functional dissociation, such as the ' Dammerzustande '
of psychic epilepsy and other states of functional psychosis,
closely approximate to the condition of dream consciousness,
though the former are more stable and far more consistent,
being narrowed to the active functioning of definite mental sys-
tems, conditions rarely to be met with in dream states.

The dream consciousness lacks unity of logical thought,
certainly fails in critical judgment and is sometimes brutally in-
different to immoral situations and acts. The credulity of
dream consciousness is well known to every active dreamer.
Changes of time, place and of objects are often instantaneous
ixnd the most incongruent situations as well as transformations
of personality are placidly and credulously accepted. The
dream consciousness is entirely at the mercy of incoming sense
impressions which spin the dream experience regardless of
truth and reality and steadiness of logical purpose and moral
ideals of "he race. From this standpoint it may be claimed that

HO BORIS SID IS.

the dream consciousness is to some extent a reversion to the
earliest forms of mental life, when the race was as yet undisci-
plined by the accumulated experience of ages of social life.

The teleological aspect of the dream consciousness may
possibly lie in the fact suggested by some that the many trains
of thought started in the activity of waking life and arrested
and suppressed by the selective thought and logic of things and
events of waking life find their vent and completion in the
activity of dream consciousness. This vent relieves us from
the high pressure of suppressed thought and makes it easier to
sustain the rigid selection of sequences of mental states required
by the struggle of existence and social life in our adjustments
to the conditions of external environment. This view, how-
ever, is not strictly correct. For the dream consciousness
follows not only along the lines of thoughts started in waking
life, but more often forms new lines of associations giving rise
to highly dramatic situations and far from relieving waking
thought impedes and depresses it, since the mind feels unrefreshed
by the sleep and in many cases serious mental troubles arise
due to the disturbing influence of active dreams on the course of
waking thought. It is more likely that there is little teleology
to dream life and if any teleology there be, it may consist in the
freedom and ease in which the mind finds itself in the dream
state, fettered as the mind is by the rigid relations of the external
environment. In dream life the routine of waking life is inter-
rupted and new associations are formed. This possibility of
forming new associations and thus breaking through the routine
of life, a possibility maintained and fostered by the dream con-
sciousness, might have possibly proved of the highest conse-
quence to the human race. The dream consciousness may thus
be regarded as an important factor in the progress of human
thought, as an agent in the breaking up of habits of thought due
to the routine of life and calling the attention of man, absorbed
as he is with the interests and requirements of the needs of his
physical world, to another life existence and strange universe of
reality.

\

THE NATURE OF HALLUCINATION. Ill

III.

The sense of reality and belief in external existence of the
hallucinatory objects are quite strong in hallucinations and in
some of the more vivid and intense dream states. In order to
explain this seemingly anomalous sense of reality, it may be well
to revert to our general principle of subsuming both the normal
and the abnormal under the same general laws and processes.
Although the abnormal is of the highest importance in revealing
new relations which the customary and habitual normal seems to
hide, as it is found for instance in the growth and development of
physiology largely due to pathological research, still we must
clearly remember that from a strictly scientific standpoint the
normal and abnormal are but teleological concepts which are of
importance for the practical purposes of our habitual life activity
and possibly for classification of various types of phenomena,
but which science is to reduce to the same laws and processes.
The abnormal is the normal out of place. In mental life as in
the phenomena of life in general the atypical, or the variation,
helps to explain the typical, the normal and the latter in its
turn explains the atypical, the abnormal. We may therefore
turn to the criterion of the normal sense of reality and validity
of experience as explaining the same relations in abnormal
mental life and the latter in its turn may throw light on the
* reality and validity ' of * normal ' experience. A brief re-
view will suffice for our purpose. It may look as if we attempt
to make an excursion into a domain not belonging to normal or
abnormal psychology proper, but to epistemology. This may
be so, but the nature of our subject brings us so closely to this
problem that a brief discussion may help us to see the facts in
a clearer light. Abnormal psychology with its various forms
of mental aberration, such as are to be found in the phenomena
of insanity, functional psychosis, hallucination, delusion, som-
nambulic states, hypnoidic states, is so intimately connected
with aberrations of the * sense of reality and validity ' of expe-
rience that not only the abnormal psychologist, but also the clin-
ician must take it into account from a purely practical standpoint.
We shall view the problem only in so far as it directly concerns
and illustrates the general subject of our discussion, namely
hallucination and illusion, or fallacious perception.

H2 BORIS SID IS.

The objective reality of the physical world is sometimes
defined, and with best of reasons, as social experience, as ex-
perience common to ourselves and our fellow men, as experience
which men share in common seems in contradistinction to the
psychic experience which is essentially of an individual character.
The tree yonder can be seen by everyone who possesses eyes,
but my perception of the tree, or my idea of it, can only be ex-
perienced by myself. It may be said that this difference
between the physical object and psychic state is a valid and
valuable one. It is, however, neither general enough, nor
specific enough. For on the one hand it may be claimed that
from a more general philosophical standpoint even the physical
object belongs ultimately to the individual only and on the other
hand it may be claimed that psychic experience is communi-
cated to our fellow men not only in terms of the physical object,
but far more often in terms drawn directly from our psychic
experience. Neither the physicist nor the psychologist will be
quite satisfied with this point of view as both physical objects
and psychic objects are entirely emptied of their specific con-
tents and must remain at best in the dubious regions of epis-
temology. Still this social aspect of the physical object is sig-
nificant and valid and is even used by the psychiatric clinician
as a practical standard in the valuation of abnormal mental life
in general and of insanity in particular. It may, therefore, be
of great value even if we do not agree with the extreme way
in which this view is sometimes put.

It is true that at the first glance we cannot help being struck
by the import of the common or social aspect of external reality.
We are well assured of the existence and presence of an ex-
ternal object, if we have the assurance of our fellow-beings, and
what is accepted by our fellow men assumes the dignity and
authority of actuality. A fact is regarded as existing beyond
the shadow of any dispute, if every one can verify it in his own
experience. The categorical necessity of our modern science
rests entirely on this principle of validity : The social object is
the valid object. This criterion of validity of the external ob-
ject stands out specially clear and distinct in our standard of
abnormal mental life. A belief is regarded as insane and delu-

THE NATURE OF HALLUCINATION. 113

sional, if it is in opposition to social beliefs and experience and
is emphatically rejected by all other men. An object is regarded
as illusory or hallucinatory, if it is treated as non-existent by
other people ; a desire, an action is considered immoral, if it is
spurned by our neighbors. The real object is the social object,
the valid belief is the social belief, and the social will is the
moral will. The individual object, the individual belief, the
individual will are treated as insane. One can not help notic-
ing, the semblance of truth in the assertions of those pathological
anthropologists who put genius in the same category with in-
sanity. What is social is alone true, valid and real, the indi-
vidual is false, non-existent. The individual can buy the reality
and truth of his being on condition of becoming social. Soci-
ality is verity.

Let us now, however, try to break through, if for a moment
only, the traditions of social regime with its criteria of social
reality and validity. When being pricked or in getting a blow,
or when cut or scratched, along with the experience of the
sensation, the experience of the external stimulus is also given.
In looking out of the window and seeing the tree with its green
leaves moving in the wind, along with the perception of the
sensory elements, primary and secondary, the external exist-
ence of the object tree is also given. Similarly in listening to
the sounds of a familiar and dear voice and listening to the
words as they form into phrases and sentences is not the sense
of reality of the external object given along with the series of
sound sensations? Sensation carries along with it the sense, the
reality of its stimulus. It is not that the sense of reality is dif-
ferent from the sensation, it is given in the sensation itself.
Similarly the percept and the sense of reality of the external
object are not two different things ; they are given together in
the same process of perception and are identical. The percept
tree is the perception of the reality of the objective tree yonder.
The sensory process is also the process of the sense of reality.
As Spinoza puts it in his Ethics : « If the human body is
affected in a manner which involves the nature of any external
body, the human mind will regard the said external body as
actually existing.' In seeing or perceiving the chair yonder we

"4 BORIS SID IS.

do not perceive it as real, because of its social or common aspect
— the reality of its existence is given directly in the sensory
processes of the percept itself. Sensory elements involve the
reality and existence of their stimuli ; the percept involves the
existence of the perceived objective content.

The sense of reality of the external stimulus or object is
strengthened by association of the original sensory systems with
other sensory systems, and the intensity rises in proportion to the
number of systems of sensory elements, brought into relation with
the functioning sensory system. If on perceiving an object, we
wish still further to assure ourselves of its reality, we verify it by
means of other sense organs. If one sees an apple and wishes
still further to assure himself of the real presence of the object,
he goes to it and examines it with his other sense organs, he
touches it, presses it, bites it, tastes it. Kinassthetic elements,
being the most important in adaptations and reactions to the
stimuli coming from the external environment, are possibly of
all sensory elements the ones that give the keenest and most
intense form of sense reality. Facts warrant us to assert with
some show of probability that the sense of reality is chiefly
centered in the sensory motor or kinsesthetic elements which
serve as nuclei for other sensory elements. Whether this be
correct or not, it remains true that the sense of reality is given
directly by sensory elements and their combinations and organi-
zations. The more systems of sensory elements are pressed into
service, the stronger is the sense of reality and the more assured
is the reaction to the stimuli of the external environment. In
the evolutionary process of man's adaptation to his environment
he becomes extended in being and grows more developed be-
cause of his social relations with other men. Man presses into
active service the systems of sensory elements of his fellow be-
ings. Adaptations and hence successful reactions to the exter-
nal environment are now more assured and the sense of reality
is still further emphasized and intensified. Throughout the
course of intensification of the sense of reality the principle
remains unchanged in a nature. The sense of reality is given
by and consists in nothing else but the sensory elements.

Social experience may be regarded as more real or as giving

THE NATURE OF HALLUCINATION. 115

a more intense sense of reality, because of the greater number
of sensory systems involved, but an object is not felt as external
and real, because of its social aspect merely, the sensory aspect
is by far the more fundamental. If one's perception of the
house yonder is of a purely * individual ' character, not shared
by his fellow men and even emphatically denied by them, the
visual preception as such still directly perceives it as real, ex-
ternal and physical. Should furthermore this experience be
intensified or confirmed by all the other senses — should he be
able to touch it, to press it and feel its resistance, knock against
it and feel concussion and pain, and have a series of tactual and
muscular sensations by walking into the perceived house and
around it, and should he further have this purely * individual '
experience of all the senses each time he comes to the same spot,
the perceived object would then be a real, external, physical
object and no amount of social contradication and lack of the ear
marks of community could make it less real objective and physi-
cal. Epistemologically regarded, community may be sufficient
for the purpose of reality ; psychologically regarded, the real,
existent physical object is essentially the perceived sensory
object given by the * community ' of sensory elements. Sensory
elements give the objective ' reals.'

IV.

From this long digression we may turn again to the ques-
tion : "What is it that makes hallucinations in general and
dream hallucinations in particular appear real, objective?" The
solution is given in the question itself. We have shown in our
analysis that hallucinations are essentially peripheral and sen-
sory in character and do not differ in their make-up from sensa-
tion and perception in general which furnish the very founda-
tions of our sense of reality. Hence hallucinations are real and
objective, because of the constituent sensory elements. Strictly
psychologically considered, percepts do not differ from halluci-
nations as far as process is concerned. Normal percepts differ
from hallucinations mainly by the fact that the former are the
habitual, the customary, confirmed by other systems of sensory
elements and that in the struggle for life, they proved to call
forth the fittest reaction.

n6 BORIS SID IS.

Dream hallucinations, like hallucinations in general, are
initiated by peripheral stimulations ; even the so-called ' central '
hallucinations are really peripheral in origin, the dream halluci-
nations naturally falling under the same category. The en-
trance of external peripheral stimulations being difficult in
proportion to the depth of sleep and extent of hallucinatory
dissociation, the internal sensations predominate in the function-
ing systems of dream life. For in sleep the activity of the
internal organs, though depressed, still goes on uninterruptedly ;
the glands continue their function of secretion and excretion,
the heart continues to contract and dilate, the blood goes on
circulating through arteries and veins ; the liver, the spleen,
the stomach, the intestines, the lungs and other organs carry on
their functions without a moment's arrest; the whole sympa-
thetic nervous system, the vasomotor, the spinal cord, the
medulla and other basal ganglia, all, contributing to the vast
mass of internal sensations, can hardly be regarded as being
asleep. All these peripheral internal sensations go to form
nuclei of primary sensations around which secondary sensory
elements become crystallized and organized and give rise to
hallucinatory percepts — to dreams. To these must be added
the external peripheral sensations coming from touch and pres-
sure of bed clothes, from changes in the muscles, joints, liga-
ments, and synovial surfaces, from changes in the superficial
temperature of the extremities from chemical changes in the
olfactory and gustatory organs, from summation of minimal
acoustic stimulations, and above all from changes in the visual
apparatus and especially from the masses of light in the retina
and macula lutea.

With the obscuration and dissociation of the mind the in-
ternal sensations along with the external peripheral minimal
sensations come to the foreground of mental life. The dreaming
consciousness stands in closer relation to the bodily functions
than the waking consciousness, absorbed as the latter is with
the intense stimulations coming from the external environment.
The intense external peripheral sensations of waking con-
sciousness obscures the weaker, but more constant internal
sensations, as Hobbes puts it, much ' as the light of the sun

THE NATURE OF HALLUCINATION. 117

obscureth the light of the stars.' This intimate relation be-
tween internal sensations was clearly seen and pointed out by
Hobbes ; ' and because ' he says ' the brain and nerves which
are the necessary organs of sense, are so benumbed in sleep as
not easily to be moved by the action of external objects, there
can happen in sleep no imagination and therefore no dream, but
what proceeds from the agitation of the inward parts of man's
body ; which inward parts for the connection they have with
the brain and other organs, when they be distempered do keep
the same in motion.'

Dreams often reveal in a symbolic form and frequently in
phantastic and grotesque images the conditions of bodily func-
tion, conditions which the waking consciousness cannot detect,
because they lie in the subconsciousness and cannot overstep
the threshold of waking consciousness. It is here in the deeper
regions of ccensesthesis, that we have to look for those * pro-
phetic ' dreams which seem to foretell some future event, some
future state of the organism. An incipient irritation of the
nerve endings in the teeth, an irritation not yet felt in the wak-
ing consciousness, may become the nucleus of a dream and give
rise to a dramatic vision of sitting in a dentist's chair and being
operated upon, a prevision that may actually become fulfilled
soon after. The growth of a malignant tumor may be repre-
sented in a dream under the form of a savage dog making an
attack and setting his teeth into the place where the tumor is to
develop. An incipient affection of the stomach may appear
under the vision of being eviscerated alive, or of having swal-
lowed a mouse which gnaws at the intestines. Incipient or-
ganic affections, not yet felt in the waking consciousness, may
thus become the starting point of a highly dramatic prophetic
dream. Dreams of such a 'veridical' character often appear
highly mysterious and their fulfilled prophecy seems nothing
short of the miraculous and supernatural. There are many
such cases on record, but the following may be regarded as
typical.

A lady, a relative of mine, had a very vivid vision which
proved * veridical ' and seemingly could only be accounted for
on supernatural grounds. One evening, on being left in a room

n8 BORIS SID IS.

all alone, she suddenly saw the apparitions of her deceased
parents. The lady became very much frightened, but the
parents quietened her and told her not to be afraid as they came
to bring her good tidings. "You will give birth to twins, a
girl and boy, name them after us, they will be strong and
healthy." With this the apparitions vanished. The lady be-
came very much agitated and, although she did not suspect to
become a mother, still, being religious and a firm believer in
spirits, she had implicit faith in the actual appearance of her
parents, who appeared to her in order to bring her glad tidings
from another world, and naturally she even began to prepare
clothes for the promised twins. As this happened in a remote
country place this prophetic vision soon circulated among all
the neighbors and expectations were aroused as to the fulfill-
ment of the prophecy. It soon became apparent that at least
a portion of the prophecy was being fulfilled. The lady soon
discovered that she was going to become a mother — the scep-
tics were somewhat confused, still they maintained their front,
but they were completely silenced, when after a few months the
lady gave birth to twins and that a boy and girl. The vision
then did prove to be of supernatural origin.

If, however, we analyze the vision somewhat more closely,
we find that it can easily be resolved into elements which
admit of a perfectly natural explanation. The vision first
of all occurred during the state of repose and was really a
dream hallucination. Still this does not explain the fact that
the hallucination appeared in such a dramatic form which
turned out to be so strikingly prophetic. On further exami-
nation of the lady's history it was found that she lost both her
parents but a few months before the occurrence of the hal-
lucination and that this loss deeply affected her. This mental
system was an important factor in shaping the course and devel-
opment of the hallucination. At the same time there was an-
other factor at work in the elaboration of the dramatically effec-
tive dream hallucination. The lady before she gave birth to
the twins was already a mother of fourteen children. It is
quite possible that, although in her waking state she did not sus-
pect of being pregnant, still in her dream state, being cut off

THE NATURE OF HALLUCINATION. 119

from the intense external stimulations, she could more easily real-
ize her condition from symptoms and changes in the internal or-
ganic sensations which now alone reigned supreme in the dream
consciousness. These symptoms and changes in the organic
sensations during the incipient stages of pregnancy could all
the more be easily appreciated by the lady as she had ample
experience of them before. Some special changes in the or-
ganic sensations such as the arrest of the menses, changes in the
circulation, in the metabolism of the generative organs and other
changes of similar nature served as so many peripheral stimu-
lations which, in states of dissociation such as occur in the light
states of sleep, favored the occurrence of a dream hallucination
that took the form of apparitions of the deceased parents, be-
cause of the subexcitement of this particular system and because
for the time being the system played a dominant role in con-
sciousness. Moreover, the organic changes differed greatly
from the previous experiences of similar kind, and it was there-
fore quite natural that the dreaming consciousness should sus-
pect the coming of twins, a circumstance which connected itself
all the more closely with the formation of the dream and was
no doubt a factor in the determination of the appearance of the
apparitions of the parents, which in turn, helped her further to
confirm the intuition that she was to be a mother of twins. All
this was represented in the dramatic form characteristic of dis-
sociated states in general.

V.

States of dissociation, light sleep and especially the interme-
diary states occurring in the course of falling into deep sleep or
coming out of it are especially favorable to the formation of hal-
lucinations. Such conditions occur in abnormal mental states
in hypnosis, in somnambulism, in hypnoidal and hypnoidic
states, in the so-called psychic equivalents of epilepsy, in pure
psychic epilepsy, and, generally, in states of functional psycho-
sis. In the intermediary states between waking and sleeping,
dissociated systems awake and become accessible to the influ-
ence of external stimuli. This is clearly shown in the hypna-
gogic hallucinations, as well as in the frequent dreams often
taking place in the lighter sleep states usually before waking.

120 BORIS SIDIS.

I have often observed in myself, when being fatigued and be-
coming drowsy and closing my eyes, how fast phantoms and
scenes flit before the mental gaze, most of them being formed
by the flitting masses of light in the field of vision. Often in
closing my eyes and keeping quiet, so as to become somewhat
drowsy, and watching the field of vision, not directly, but, so to
say, from the corner of the eye, animals, figures, faces, can be
seen forming and dissolving into mist. These phantoms can be
directly traced to specks of light and masses of color coming
from the retina and especially from the macula lutea. In many
psychopathic cases, not only vision, but also sounds and voices
are experienced as in some of my cases that have hypnagogic
auditory hallucinations of voices. The dimly lighted up regions
that lie on the borderland of sleep and waking states are peo-
pled with phantoms, ghosts and apparitions.

Statistics seem to confirm this point of view, since about 50 per
cent, of cases of hallucinations may be classed as 'borderland hal-
lucinations.' Some recent critics in this field of inquiry strongly
favor the view that hallucinations occur in dream states, hallu-
cinations being nothing else but vivid dreams, the percipient not
being conscious of having fallen asleep. This view is not new, it
is favored by Hobbes. ' The most difficult discerning,' Hobbes
tells us, 'of a man's dream from his waking thoughts is then when
by some accident we observe not that we have slept.' Many
cases no doubt admit of such an explanation. I myself had an
experience of such a character. While sitting and studying one
evening, I felt myself suddenly transported into my father's
house and looking out of the window, seeing the scenery char-
acteristic of the locality and hearing the voices of my parents in
the next room, but I could not discriminate the words. The vision
was so real that I was surprised to find myself again at my book
and in a place hundreds of miles away from home whither my
hallucinatory state had so suddenly transported me. The hal-
lucination was so strong and real that had I not critically
analyzed the conditions of its occurrence I should have been
fully certain that the hallucination appeared in the waking
state. As a matter of fact, I was fatigued from my studies
and dropped off. The actual surroundings, the room, the

THE NATURE OF HALLUCINATION. 121

table, the book, the voices of my friends present, all disap-
peared from my view during the intermediary state, and when
I came out of it I remember the start I gave in realizing once
more the actual situation. According to records, hallucinations
take place when the percipient is in bed, just after retiring, or
about to wake up, or after waking. The percipient is really
asleep, only he is not aware of it, so brief is the state and so
intense and vivid is the hallucination. It may, therefore, be
maintained with some show of truth that hallucinations are
dreams and take place in sleep states.

We must guard, however, against carrying a generalization
too far. This contention that hallucinations occur in dissociated
dream states is somewhat overstated. It is true that hallucina-
tions require states of dissociation, but this does not necessarily
mean sleep states. Not all states of dissociation are dream
states taking place in sleep, although it may be safely asserted
that all dissociative states have many traits in common and are
at bottom of the same nature. Hallucinations and dreams may
be analogous, may be of the same structure requiring the same
general conditions, but it does not for that reason follow that
they occur in the same states, in sleep states. Dissociation
with consequent hallucinations may also take place in waking
states. Those who have studied hallucinations in different
forms of mental diseases know that most of the hallucinations
occur under widely different conditions and they further know
that it is precisely in the waking states that hallucinations are
most commonly present, while in the sleeping states they are
more frequently absent. Insanity may be compared with dream
states, but they are by no means identical. The important con-
dition requisite for the occurrence of hallucination is dissocia-
tion and this condition often occurs in waking states, such as
the hallucinations found in many forms of insanity, as for in-
stance paranoia, hebephrenia, katatonia, general paralysis and
other states of mental aberration. Even hypnotic and post-
hypnotic hallucinations can hardly be claimed to have been
really induced in dream states. They who have devoted time
and labor to hypnosis know that the hypnotic state can by no
means be identified with sleep and that in the very deepest stages

122 BORIS SfDIS.

of hypnosis the subject is to all intents and purposes fully awake ;
he is full of activity, his eyes are open, his senses are on the
alert — he is far more awake to external stimuli than even in
his normal state. The mind is very active and the subject car-
ries on long trains of reasoning, argumentations and discus-
sions with the people around him ; in short, the subject in the
deep somnambulic state is in a condition the very opposite from
that of the sleeping state. Hallucinations occur both in the
waking and sleeping states and require dissociation as an indis-
pensable condition.

VI.

If we inspect more closely the relation of the stimulus to the
hallucination, especially to the dream hallucination, we find that
the intensity of the content is disproportionate to the intensity of
the initiating stimulus, to the peripheral sense impression. A
comparatively slight stimulation often gives rise to a dream of a
highly dramatic character. This exaggerated character of the
dream hallucination is well known. Thus a prick of a pin may
give rise to a dream of being attacked by robbers and finally
being run through by a thrust of a dagger. The application of
a warm bottle to the feet may develop a dream of ascending a
volcano and walking on molten lava, while a cold stimulus may
give rise to a dream of participating in a dangerous expedition
to the North Pole. Pain in the head, impeded respiration and
pressure in the region of the neck may develop, as in the case
of a friend of mine, the horrible dream of being dragged into
a narrow dungeon and then beheaded.

To explain this dream exaggeration a theory is advanced
based on dissociation. It is claimed that dissociation tends to
convert the physiological 'ideational currents' into sensory
* currents ' and intensify and exaggerate the psychic states.
Before discussing the theory it may not be amiss to examine
the facts which the theory is called to explain. It is question-
able whether the general relation of dream stimulus is quite cor-
rectly stated. It appears that the generalization is stated some-
what in the form of the well-known question : Why do great
rivers flow by great cities? It is by no means generally true
that the characteristic of dream consciousness is to exaggerate

THE NATURE OF HALLUCINATION. 133

stimuli received and work them up to a pitch so as to convert
4 ideational into sensory currents.' The relation is far simpler.
The dream does not necessarily as a rule exaggerate incoming
stimuli and make of them exciting and sensational dream hallu-
cinations. What happens is this : the commonplace non-exag-
gerated, unaffective dreams tend to fade away almost immediately
on waking, while the impressive dreams are usually remem-
bered. I have observed a number of dreams in my own case
as well as in others and have found that the number of or-
dinary commonplace dreams far predominates over the striking
and extraordinary dreams. Even in psychopathic cases in which
subconscious dream life is often well developed, even in such
states I have found in the cases which have been under my ob-
servation and experimentation that the commonplace dreams far
predominate over the dramatic and extraordinary ones. The
only way to convince oneself of it is to try to write down the
dream immediately on waking. I find that the ordinary dream
is very hard to hold in memory, it is elusive and is constantly
slipping away from us, a special effort of attention is requisite
to hold on to them ; they are usually hazy, vague and con-
fused. On the whole, the indifferent dreams really predom-
inate, but it is only the impressive ones that remain in memory.
Even the freshness and recency of the dream do not save it from
falling into oblivion. Now while commonplace and indifferent
dreams are forgotten older dreams, but more impressive, more
awakening our emotions, especially emotions of fear, will be
clearly and vividly remembered.

Still the fact that exaggeration and intensification of the
sense impressions received by awakening a greater volume of
secondary sensory elements and representations more often than
in the waking state requires an explanation. This intensifica-
tion may partly be due to the fact that in sleep sensory impres-
sions often enter consciousness suddenly. This brings about a
shock, awakening emotions which are conducive to a greater
stimulation of a greater volume of secondary sensory elements
and their accompanying representations. Even in the normal
waking state sense impressions suddenly introduced into con-
sciousness may cause a shock and give rise to an illusion, the

124 BORIS SID IS.

object appearing as something strange and formidable. We
can often observe it in ourselves, when falling into a drowsy
condition, a slight stimulus which we otherwise ignore will give
us a sudden start. I often observed in myself when in a
drowsy state and ' dropping off ' how an ordinary stimulus such
as a cough, for instance, will produce a shock affecting the
visceral organs, the feeling being somewhat similar to the con-
dition commonly described as a ' sinking sensation in the pit
of the stomach ' ; the shock seems to reverberate all over the
organism.

To this must be added another important factor, namely,
the emotion aroused. When an object is perceived under con-
ditions that do not permit its recognition or its assimilation and
consequently its customary reaction, an emotion of fear, or that
of fright is produced. Such is the case, for instance, when
some objects impress us in the dark or when we get hold in the
dark of some slimy, slippery and especially of moving objects.
These two factors often work together inasmuch as an object
suddenly introduced into consciousness is also not speedily
assimilated so that the shock and emotion due to non-recognition
or non-assimilation go together. Now in sleep stimuli entering
into consciousness effect it in a sudden way and from the very
nature of the sleeping consciousness the external stimulation is
but imperfectly assimilated ; both factors, shock and emotion,
due to non-assimilation are present and sometimes give rise to a
highly wrought up emotional state which is so apt to transform
objects by arousing different systems of elements and at the
same time to impress the memory powerfully.

It is claimed that the very fact of dissociation brings about
an intensification of ideational states converting them into sen-
sory states. Physiologically, the assumption is made that the
sensory nerve cells can be set into activity not only by peripheral
stimuli, but also by central * currents ' going from center to
periphery. The sensory centers are like a bucket with water,
the upsetting of the bucket being likened to the upsetting of the
sensory centers, giving rise to sensations. This upsetting can
be affected by peripheral ' currents.' Small intracellular idea-
tional currents flow freely through the centers without upsetting

THE NATURE OF HALLUCINATION. 125

them. Now when an obstruction occurs in the sensory centers
the ideational currents which otherwise flow out and disperse may
accumulate, and aided by a chance activity of central character
may upset the nerve cell in the same way as our bucket may be
upset by the accumulation of water from the small incoming cur-
rents (like the ideational currents), when the holes and interstices
through which they usually flow out are stopped up. The hy-
pothesis as far as explanation goes is good enough, the drawback
is that it explains too much. For it is hard to understand why
intense dreams of this character do not occur more often.

Besides it is hard to realize how an idea can give rise to a
sensation of any intensity by the mere agency of ideas, the sen-
sation and its intensity being entirely a function of peripheral
stimulation and consequent sense impressions. An idea, a rep-
resentation, may be very vivid, but does not become a presenta-
tion or sensation. A sensation is not an ' intense ' idea, nor
is an idea a weak sensation. A series of sensations arranged in
ascending or descending gradation of intensity may be likened
to the continuous series of the spectrum in which there is a
qualitative difference from line to line, a difference that admits
of no substitution. A sensation the intensity of which is changed
is a fallacious percept, a hallucination. A thunder clap per-
ceived as a whisper, a whisper perceived as a thunder clap may
be equally regarded as fallacious perception as any other change
in the content of the percept. The rustling of leaves perceived
as an explosion is as much of fallacious perception as when the
paranoiac, for instance, hears in it curses and threats of his ene-
mies. Sensations and percepts cannot change in content or in-
tensity without giving rise to illusions or hallucinations. The
changes that may occur in regard to sensations and percepts
without their being qualitatively changed and becoming falla-
cious can only be in vividness belonging to the representative
elements which cluster round the primary and secondary sen-
sory elements. A less intense sensation may be more vividly
represented than one of greater intensity. A weak sound, a pale
color, a light pain may be more vividly represented than the
ones the intensity of which is far greater. This vividness, how-
ever, is not at all a characteristic attribute of the sensor}' ele-

126 BORIS SIDIS.

ments, it is rather an attribute belonging to the functioning sys-
tem of representative elements into which the given sensory
elements enter as constituent nuclei.

Keeping to facts as closely as possible we may venture with-
out much risk on the following generalization which may be
regarded in the light of a working hypothesis. Just as sensory
primary or secondary sensory elements vary in intensity and can
be arranged in a continuous series of gradations of intensities, so
do the representative elements vary in vividness and may be
arranged in a continuous gradated series of vividness. Sen-
sory elements have intensity, but no vividness, -while repre-
sentative elements have vividness, but no intensity. Repre-
sentative elements may refer to the same presentative content
with different degrees of vividness. Vividness of representative
elements like intensity of sensory elements may pass through
all degrees of variation from maximum to minimum and finally
reach a vanishing point. In this respect vividness is like sense
intensity and as a matter of fact the two are usually interrelated.

Under ordinary conditions of psychic activity sensory inten-
sity and representative vividness vary together. An intense
sensation is vividly represented and a weak sensation less so,
the vividness varying directly with increase or decrease of sen-
sory intensity. This direct variation, however, is not always
constant ; there are conditions under which the two may part
company such, for instance, as are found in states of distraction
or in states of dissociation. Under such conditions a strong
stimulation giving rise to sensory elements of great intensity
may give rise to representative elements of but slight vividness.
In states of distraction as well as in various states of mental
dissociation sensations of great intensity may meet with so little
vividness in the representative elements as to fall so to say
below the threshold of consciousness, may be submerged into
the twilight region of the subconscious and ' not be perceived
at all.' From this standpoint we may say that the depth of
dissociation varies inversely as the degree of vividness. When
vividness is at its minimum, dissociation is at its maximum, and
inversely. Briefly stated, dissociation and vividness are inversely
interrelated variables.

THE NATURE OF HALLUCINATION. 127

VII.

Functional -psychosis, the basis of which is dissociation, may
also psychologically be regarded, according to the gravity of the
psychopathic affection, as a decrease or even loss of "vividness of
representative elements. The diminution or total loss of vivid-
ness may be of different systems of representative elements and
will thus give rise to various forms of psychopathic amnesias,
which play such an important role in functional psychosis,
which in the main is a disease of representative life consisting
in a decrease of functional activity of representative elements
and which from the present point of view may be regarded as
the tendency towards a minimum of the most important attri-
bute of ideational elements, namely, vividness.

From this standpoint, the degree of vividness of ideational
elements can no more confer on them sensory intensity than the
idea of riches, however vivid, can confer upon one the power
of wealth. Dream hallucinations, like hallucinations in gen-
eral, are sensory in character, not because of the intensive na-
ture of the central elements or ideas, but because of the primary
and secondary sensory elements present, directly and indirectly
peripherally initiated, as it is in the case of all sensory and per-
ceptive processes. Hallucinations are peripherally induced and
are started either in the same sense organ, or indirectly in some
other sense organ, the secondary sensory elements form so to
say the hypertrophied portion of the hallucinatory percept, but
they are always sensory in character and peripherally initiated.
The more closely one investigates hallucinations, the more he
learns to trace cases of supposed mysterious hallucinations to
external peripheral sources. A pure central hallucination is as
rare as the fabulous phoenix. A central hallucination means
an unanalyzed psychic state. Whenever an analysis of such
hallucinations is made, the peripheral sensory character, primary
and secondary, stands out distinctly in the foreground. In the
so-called ' purely central hallucination ' the nuclear primary
sensory elements remaining in the background of consciousness
cannot easily be traced to their appropriate peripheral sense
organs and their external stimuli and are on that account re-
garded as * centrally initiated.' Dream hallucinations, hyp-

128 BORIS SIDIS.

notic, hypnagogic and pseudo-hallucinations, if closely analyzed,
can be clearly traced to peripheral origin, — to peripheral stimuli
that give rise to primary sensory elements that form nuclei
round which secondary sensory elements become organized as
cytoplasm.

These so-called central hallucinations form the stumbling
block of the psychologist and the psychopathologist. To ac-
count for them the theory is commonly advanced that the irri-
tability of the ideational centers may reach such a pitch as to
give rise to such intense ideational states as to amount to a full-
fledged sensation or perception and thus bring about a pure
central hallucination. It is strange that such a theory should
be maintained at all and that it should gain currency. The
theory does not accord with the facts, and its very principle dis-
regards facts. For no matter what strength an idea may attain
it is still far from becoming a sensation. An idea of a bell does
not sound and an idea of a blow does not strike. The fact is,
as we have pointed out before, ideas or representations are qual-
itatively different from sensations ; an idea can as little be con-
verted into a sensation as the sour taste of vinegar can be turned
into violet color of the spectrum. Ideas and sensations differ
fundamentally, they differ in kind and no amount of ideational
activity can ever be made to become sensory in nature. A
higher pitch of ideational activity will make an idea more vivid,
but can nowise confer upon it sensory qualities, just as all the
immensity of space and infinity or eternity of time can not make
them weigh as much as a grain.

A further modification of the same theory is given by those
who maintain that central hallucinations are due to the irritability
of the higher ideational centers from which ' ideational ' currents
are propagated to the lower sensory centers. In other words,
it is not the idea that becomes by its intensity or by its vividness
directly transformed into a sensation, but an intense or vivid
idea may give rise to a corresponding sensation without the
presence of an external stimulus, or of a peripheral sensory
process. Psychologically as well as biologically regarded, the
theory is untenable. For it is not in accordance with observed
facts that an idea, however vivid, should give rise to a corre-

THE NATURE OF HALLUCINATION. 129

spending sensation or percept. Were that the case the course
of internal and external worlds would have become confused
and confounded, man would have become the dupe of his own
ideas, the world a gigantic madhouse, and the process of idea-
tional activity would have long ago become eliminated in the
struggle for existence.

From a physiological standpoint, the theory can hardly be
considered, inasmuch as it is in direct opposition to the known
physiological laws. Sensory excitation, ideational processes
and motor reaction form, so to say, a sensory-ideo motor arc, —
the excitation going from peripheral sense organs to central sys-
tems and thence to the muscles. Now the conditions postulated
by the central theory are such as to have the processes reversed-
Sensory processes work upward, from periphery to center, while
motor processes work downward, from center to periphery.
On the modified central theory, the sensory process in halluci-
nations is reversed, it goes downward instead of upward. There
is not a particle of evidence for such reversal, the assumption
being in contradiction to the principles of physiology. The
claim of special structures for effecting such a reversal is
entirely unfounded. As far as can be ascertained, the neuron
works ' cellulipetally ' in the direction of the sensory ganglia
and central neuron systems, while the neuro-axon works ' cellu-
lifugally ' that is from sensory ganglia and central neuron sys-
tems to the periphery to the muscular apparatus. There is on
the other hand not the least bit of evidence that the functions
of neuron systems can be reversed in their course.

The central theory then cannot stand the test of critical ex-
amination as it is neither in accord with the facts it is called to
explain, nor does it fall in line with the facts and principles of
physiology. We are therefore forced to fall back on the pe-
ripheral origin of hallucinations under the condition of central
dissociation. According to the theory advanced in this paper,
the origin, and structure of hallucinations, of dream hallucina-
tions as well as of pseudo-hallucinations and hypnotic halluci-
nations do not differ in the least from those of normal percep-
tion, a difference unwarrantly claimed by the theories of cen-
tral origin of hallucinations. Hallucinations are peripherally

13° BORIS SID IS.

initiated, hallucinations are abnormal percepts occurring" under
the conditions of central dissociation 'with primary and secon-
dary elements as their central nuclei.

VIII.

The phenomena of so-called * double thinking ' are ex-
tremely interesting from our point of view. The patient hears
his own thoughts uttered aloud. He has the hallucination of
his thoughts uttered when engaged in writing or in reading,
though loud reading may check the hallucinatory voices. These
hallucinatory voices may be of an imitative character and simply
repeat what is spoken or read by the patient ; or they may be
of an anticipatory character and utter the patient's thoughts be-
fore he himself utters them. The usual explanation of such cases
is found in the theory of the so-called ' overcharged centers/
Where the voices follow and repeat the patient's words and phrases,
it is assumed that the auditory centers are highly irritable and
overcharged so that stimulations from other centers bring about
a discharge into the * ideational ' auditory centers and auditory
hallucinations result. In the case of reading, for instance, the
visual image of the word awakens also an auditory image, but
when the auditory centers are overcharged the visual images
awaken directly an auditory image before the spoken word
takes place. Now this auditory image is so intense, on account
of overcharge, that it becomes an auditory hallucination and the
patient hears his own thoughts uttered aloud. This reflex action
from one ' ideational * center into another occurs while the
patient reads or writes, and that is why he has the experience,
the hallucination that there is a voice often regarded as ' inner '
which repeats his own words and phrases. Cases where the
voice utters the words and phrases before they are written are
explained on the hypothesis that the central discharge into the
overwrought auditory centers occurs before the words are
written down or before the motor discharge takes place. When,
however, the patient hears the voice repeat the phrases soon
after he has uttered them, the phenomena are explained on the
supposition that the centripetal currents from the speech centers
into the auditory centers give rise to the voices, the patient hear-

THE NATURE OF HALLUCINATION. 131

ing his own words shortly after he has uttered them, the efferent
discharge from the graphic centers into the auditory centers
will give rise to an auditory hallucination of hearing the words
and phrases he has just written. In the phenomena of * double
hearing ' the patient has the hallucination of hearing his own
voice while talking or reading aloud, and then again another
voice due to the centripetal discharge from the speech centers
to the overcharged auditory centers. Thus in some patients
these hallucinations of hearing are brought about by the volun-
tary suppression of speech, the patient then hears a voice utter-
ing his own thoughts. This is claimed as confirming centra
initiation — the currents from the word images in the speech
centers not having a free outlet run into the overcharged ' idea-
tional auditory centers ' and give rise to inner speech heard by
the patient.

In opposition to this central theory of double thinking or of

* inner speech ' held in various forms by psychologists and psy-
chopathologists, there are some who maintain the view that these

* double thoughts ' hallucinations are not of central, but of
peripheral origin, being due to hypersesthesia of the centripetal
paths. The apparatus employed in speech carries out not only
the requisite delicate movements, but also forms the sensitive
apparatus for information of the movements executed. The
sense of movement may be regarded as originating in the mus-
cles, especially in the joints and articular surfaces. Sensory
stimulations coming from these structures to their appropriate
central systems give rise to kinaesthetic sensations and motor
ideas. Now if the peripheral sensory tracts of the muscle sense
or of kinaesthetic sensations become hyperaesthetic, kinaesthetic
sensations and motor ideas are aroused automatically and may
give rise to hallucinations of positions, movements and acts ;
movements which have not been performed are thus experi-
enced. If now the centripetal sensory tracts of the speech cen-
ters are hyperaesthetic then involuntary kinaesthetic sensations
and respective ideas arise which go to form the hallucinations
known as * double thought.' The patient experiences 'inner*
speech, a voice repeats after him his own thoughts, his own
words and phrases. When the speech centers are overcharged

I32 BORIS SID IS.

and give rise to automatic centrifugal discharges, then the hy-
perassthetic centripetal paths bring it back in the form of spoken
words and the patient experiences his own thoughts uttered by
an inner voice which is foreign to him. In speaking the inner
voice comes after the speech and reverberates like an echo and
persists as an ' after image ' of the spoken word or phrase.
When the patient is engaged in writing, the voice usually pre-
cedes the written phrase, because the spoken word image pre-
cedes the written word image, the inner voice thus anticipates
the patient's writing by uttering his thoughts. This theory
seems further to be confirmed by cases in which such halluci-
nations take place. If one observes closely cases of ' double
thinking' or of 'inner speech,' he often finds 'involuntary
whispering' present — the patient whispers to himself. These
whispers come back to him, on account of the hyperaesthesia
of the peripheral paths he hears it as speech of some inner voice.
A close examination of the two theories, of the central and of
the peripheral, reveals their inadequacy. The central theory, as
it is generally put forth and commonly accepted, may possibly be
regarded as the more inadequate. For the central theory rests
on the psychological fallacy, so prevalent in psychopathology
that it maybe regarded as the psychopathologist's fallacy, namely
that an idea may reach such a high pitch of intensity as to
become sensory in nature and give rise to a percept. The per-
cepts formed by the visual perception of reading awaken, ac-
cording to this theory, also accompanying ideas of sound inti-
mately related associated with visual word reading, and it is these
ideas that reach such a high intensity as to give rise to halluci-
nations of hearing, the words are read aloud, as if by a strange
voice. This explanation, as we have already pointed out, is psy-
chologically incorrect and rests on the fallacy that ideas have
intensity and that an intense idea becomes a sensation, or that a
sensation is but an intense idea and an idea is a weak, a faint
sensation. To modify this view and assume that an intense idea
stimulates and gives rise to the formation of a percept is to as-
sume a supposition not warranted by facts that an idea is equiv-
alent to the action of external stimuli or objects with their
requisite physical structures and processes. In either case, the

THE NATURE OF HALLUCINATION. 133

central theory as it stands is not in accord with psychological
and physiological data and, as such, cannot possibly be accepted
at least in the shape as it is usually put forth.

Furthermore there is an inherent difficulty in the central
theory itself. For if it be correct, as the theory claims, that the
visual image calls forth an intense auditory image amounting
to a hallucination, the hallucinatory voice should precede and
not follow the patient's reading. In order to explain the hal-
lucinations of double thinking or of double hearing in the case
when the voice follows the reading, it would have to be as-
sumed first that the visual image of the written or printed word
stimulates the speech centers, which, innervating the muscular
apparatus of speech, give rise to reading, which in turn stimu-
lates the peripheral auditory apparatus, awakening activity in
the auditory centers, giving rise to the hearing of the read
words, and that then only do the indirect stimulations of the
visual image coming from the visual centers awaken once more
the same central connections, thus bringing about a repetition
of the self-same words heard. We have to assume that the
action of the visual centers in stimulating the motor speech
centers with the resulting acoustic stimulations and functioning
activity of the auditory centers are enacted before the direct
central stimulations from visual center to auditory center take
place ; in other words we must assume central retardation.
Now what does this central retardation mean? It means that
the phenomena of double thinking or of double hearing are
brought about by some form of central inhibition, of central
blocking of pathways as it is usually put : in other words, the
requisite condition of double thinking is reduced to the psycho-
pathological state of central dissociation.

The inadequate side of the central theory as it is commonly
advanced lies in the supposition of its referring auditory hal-
lucinations in the phenomena of double thinking or hearing to
the intensification of the auditory image or idea, but no strain-
ing of an auditory image can get a sound out of it. Moreover,
were the central theory correct it would really involve a double
auditory hallucination, one preceding and the other succeeding
the reading. For in the process of reading the visual image of

134 BORIS SID IS.

the word awakens the auditory image along with its kinaesthetic
image, stimulating the centrifugal motor apparatus and giving
rise to the spoken word. Now this awakened auditory image
preceding the spoken word, on account of the assumed irrita-
bility of the auditory centers and the consequent ' intensification '
of the stimulated auditory images, should necessarily give rise
to a full-fledged hallucination. When the voice also follows the
reading, a dissociation of the visual from the auditory centers is
assumed, a dissociation that gives rise to a secondary succeeding
hallucination of the words and phrases read and spoken. The
fact that the central theory requires the presence in all phe-
nomena of double thinking that the voice should necessarily pre-
cede the reading ; that when the voice follows the reading, an-
other hallucinatory voice must have also preceded and that there
is also a double stimulation from the visual into the auditory
centers, that the hallucination first appears under conditions of
association of visual and auditory centers, while the succeeding
hallucination occurs immediately under the opposite conditions,
namely dissociation, — all these assumptions make the central
theory wholly unsatisfactory and unacceptable.

IX.

The peripheral theory of double thinking is on general
grounds more acceptable as it falls more in line with psycho-
logical and physiological principles and facts. Unfortunately
the special facts which the theory is called for to explain do not
exactly tally with it and may even be said to contradict the
hypothesis. For if the hallucinations of double thought are
due to hyperaesthesia of the centripetal sensory-motor tracts,
then reading aloud should intensify the hallucination, but the
case is quite the reverse, — reading aloud makes the hallucina-
tory voice to disappear altogether. On this theory again, the
voice should follow the reading. We are thus confronted with
the opposite difficulty met with in the central theory. On the
central theory the hallucination should precede, while on the
peripheral theory the hallucination should follow the reading.
The central theory cannot account well for succeeding hallucina-
tions, while the peripheral theory does not account well for pre-

THE NATURE OF HALLUCINATION. 135

ceding hallucinations. On the central theory there should be
double hallucinations in cases where the voice follows reading,
while on the peripheral theory there should be double halluci-
nations, when the voice precedes the reading. Besides « hyper-
aesthesia ' alone should rob the perception of its hallucinatory
character, the patient should be the more conscious of his own
utterance.

A closer examination of the peripheral theory discloses a
fundamental fallacy which it primarily involves, a kind of
ignoratio elcnchi. The theory is probably correct in principle,
but it misses the essential point of the whole problem ; it may
be an adequate explanation for motor, but not for auditory hal-
lucinations. Hyperaesthesia of the central motor speech tracts
would at most give rise to pure kinaesthetic hallucinations. The
patient may have hallucinations of action, tension, or of move-
ments in his peripheral speech organs, but he will have no hal-
lucinations of hearing. To have an auditory hallucination, as
to have an auditory perception in general, the auditory periph-
eral and central apparatus should be stimulated. No other
organ but the acoustic apparatus can possibly supply sensations
and percepts of an auditory quality, unless the hallucination be
of a reflex secondary character, but then it may be induced
through any other peripheral source than that of kinaesthesis of
the speech organs.

Although each theory taken by itself proves to be inadequate
and leads to contradictions and puts us out of accord with facts,
still the two may be regarded in a certain sense as supplement-
ing each other, if modified by supplementary conditions. Now
the central theory emphasizes the aspect of the central charac-
ter of the phenomena, while the peripheral theory lays stress on
centripetal factors ; both, however, can be brought in line with
facts, if assuming centripetal factors of kinsesthesic and specially
auditory hyperaesthesia we also refer to the central conditions
of dissociation. The patient in double thinking is subject to
subconscious states, to states of dissociation ; this dissociation is
of central character and specially affects the visual and kinaes-
thetic systems. Impressions, on account of dissociation and
peripheral hyperaesthesia, are subconsciously received and sub-

I36 BORIS SID IS.

consciously reacted upon. The visual impressions of the written
and printed characters are subconsciously perceived and sub-
consciously uttered in a whisper and sometimes quite loud, as I
had occasion to observe in a case of mine. This subconscious
utterance, unperceived by the patient, 'comes back to him
as a strange external voice proclaiming the patient's thoughts
or repeating his words and phrases. The hallucinations of
' double hearing ' are due to subconscious whispering which
comes back to the patient as an auditory hallucination. I had the
occasion to verify this phenomenon of subconscious whispering
in a case in which functional dissociation was quite marked and
in which auditory hallucinations and double thought were quite
persistent.

In cases where the auditory hallucinations precede the read-
ing or writing it is the subconscious whispering along with
kinsesthetic and auditory hyperaesthesia that give directly rise
to the phenomena of ' double thought,' or of ' double hear-
ing.' The dissociation being in the kinaesthetic systems the
patient does not experience consciously the peripheral incoming
sensations due to his subconscious whispering. More often the
patient continues to whisper subconsciously what he has just
read consciously. Such a habit is common with many people
in the normal state and is due to the result of the persistence of
the peripheral sensory impression, to a kind of verbal after-
image. The absence, however, in the normal condition of dis-
sociative states prevents the formation of subconscious whisper-
ing with its consequent auditory hallucinations partly due to
hyperaesthesia of the auditory tracts.

If cases of ' double thinking ' are closely examined one finds
in them the presence of subconscious states with their psycho-
motor reactions, the patient in walking on the street, for in-
stance, may hear a voice telling him words and phrases that can
be traced to signs and advertisements which he has read sub-
consciously though he himself has not been aware of it. What
happens in such cases is this, the patient whispers or even
utters aloud the words h ~ subconsciously sees on the signs. It
is this subconscious whispering that comes to him back as an
auditory hallucination of a voice. In one of my cases in which

THE NATURE OP HALLUCINATION. 137

the patients suffered from auditory hallucinations I found on
close examination the phenomenon of unconscious or subcon-
scious whispering, which became very much aggravated in
proportion to the state of distraction in which the patient was,
ranging from an almost inaudible whisper to a loud talk, the
patient being entirely ignorant of it and could not be made
aware of it, even when the attention was fully called to his
talking. One of my patients suffering from pronounced audi-
tory hallucinations, but in whom the dissociation is not deep,
aptly describes his experiences as * autovocalization.'

Similar conditions can be induced in hypnosis thus con-
firming our point of view by experiment on otherwise nor-
mal people. If a post-hypnotic suggestion of subconscious
whispering is given, the subject experiences an hallucination
analogous to that of * double thought,' — the subject hears a
voice telling him the words and phrases which he himself
whispers, but of which he personally is entirely ignorant. The
peripheral character of the * double thought ' or hallucination
under condition of central dissociation may thus be regarded as
an efficient working hypothesis in accord with facts.

From the whole course of our discussion it appears that we
remain more closely in touch with facts, if we accept the view
that hallucinations require states of dissociation as central con-
dition and that they are primarily peripherally initiated having
secondary sensory elements as their main content; in other
words, hallucinations are dissociated secondary -percepts.

DISCUSSION.

THE MECHANISM OF IMITATION.

The object of this paper is to explain the psycho-physical process
by which acts of imitation are performed. We often hear imitation
spoken of as an instinct, and of course, in the loose popular sense of
the term, there may be no objection to speaking of the tendency to
imitate as instinctive in mankind. But if we use the term instinct in
a technical sense as applying to those acts which are made possible by
hereditary paths of discharge in the nervous system, it seems impos-
sible to bring our acts of imitation under this head. We can under-
stand how a definite reflex might become hereditary. The act of
swallowing when something is placed in the mouth, is a perfectly
definite sensori-motor reflex, and we can easily conceive that the nerve
arrangement necessary for this act should be prenatally established.
But in the case of imitation we have no single definite act, but an
indefinite variety of actions. No one definite set of nerve adjustments
could explain such different activities as the imitation of sound, the
imitation of a movement of the hand, and an imitation of the shake
of the head. The very use of the single term imitation for such a
large variety of actions, has a tendency to mislead. Many writers
speak of imitation as if it were a sort of faculty, and this leads to the
usual error of faculty psychology — the individual act is thought to be
sufficiently explained when it is shown to be an instance of the general
activity of the faculty. As James says there is no such thing as mem-
ory, but only memories, so we might say there is no such thing as
imitation, but only imitative actions.

Our instincts are sensi-motor, our imitations ideo-motor. Since
our ideas are not hereditary, but are always individual acquisitions, it
seems necessary to regard the actions especially associated with these
acquired ideas as also acquired. A movement which depends upon
the perception of another's doings cannot be more original than the
act of perception itself. In speaking of perception, Professor James
says, " Nature is frugal in her operations, and will not be at the ex-
pense of a particular instinct to give us that knowledge which expe-
rience and habit will soon produce. Every perception," he says, " is
an acquired perception." l Now this, it seems to me, applies to imita-

1 Principles of Psychology, Vol. II., p. 78.

138

DISCUSSION. 139

tion as well as to perception. Nature does not furnish us with par-
ticular instincts to give us those actions which experience and habit
will soon produce. Every imitation is, I believe, an acquired imitation.

Many recent writers seem, however, to regard imitation as hered-
itary and instinctive. Thus Professor James himself includes imita-
tion in his enumeration of the prominent instincts of man.1 Yet he
shows an apprehension of something other here than a single instinct
by distinguishing ' the instinct to imitate sounds,' from ' the instinct to
imitate gestures.' Had he carried this differentiation further he would
soon have seen that the great variety of imitative acts precludes all
thought of their reference to one or two hereditary instincts. Royce,
in his recent Outlines of Psychology^ speaks of the sources of the
social interest, as instinctive and hereditary, and then goes on to say,
" On the basis of the general social interests there appear more special
instincts amongst which the most prominent is the complex of instincts
suggested by the name imitation." * Tarde, in his Laws of Imitation,
calls imitation the 'action at a distance of one brain upon another.'1
He speaks of * a will to imitate ' as being handed down,* and he quotes
from Maudsley with apparent approval the statement, " It cannot be
too clearly apprehended that there is a sort of innate tendency to
mimicry in the nervous system." 5 Baldwin's view is not entirely clear
to me. He seems to regard imitation as an acquisition of the race,
but as an instinct in the individual. At any rate he speaks of the
* imitative instinct,' and again refers to * the instinct to imitate ' as
found in the child and in animals. 6

In contrast with these writers who more or less definitely pro-
nounce imitation an instinct, we may note this statement of Stout in
his just published work, The Groundwork of Psychology. " Both
spontaneous and deliberate imitation," he says, " presupposes a motor
association between the perception or idea of the act to be imitated,
and more or less similar movements which the child has already
learned to perform."7 Stout, it seems to me, is on the right track.
He does not explain, however, the way in which these motor associa-
tions first arise. These motor associations, it seems to me, may be
explained as follows : In same cases they are based on instinct. The

1 Ibid., p. 408.

•P. 275.

* G. Tarde, Laws of Imitation, trans, p. 199.

*Ibid.t p. 193.

s/M/.,p. 88.

6 Baldwin, Mental Development, p. 290.

» P. 82.

1 4° THE MECHANISM OF IMITATION.

child instinctively cries at its own pain, and thus associates the sound
of the cry with the cry movements. This association established, it
becomes then perfectly natural, on the principle of ideo-motor action,
for the child to cry when he hears another cry. In other cases the
association between the idea and the movement may have been estab-
lished by purely random activities. The spontaneous overflow of ner-
vous energy makes the child wave his hands — he then sees the hand
movement, and thus establishes a path of connection between the idea-
tional process and the corresponding motor process. With this con-
nection established, it is easy to understand how the child comes to
wave his hand when he sees another perform the same act. In some
cases probably the associations on which the imitative act depends may
depend upon the imitations of others. Take the case of the imitative
smile. The child cannot see his own smile and thus form a visual
image of it, as he can of his hand movements. But the mother smiles
when she sees the smiling face of the babe, and the child thus asso-
ciates its own motor feelings of the smile with the visual idea of the
smile as seen reflected in its mother's face. The necessary association
is now established, and thereafter the child readily smiles when he
sees others smile. In the case of vocal sounds while the imitation of
the child by others may not be an absolutely indispensable factor, it is
yet of great service in establishing the motor associations. To hear
its own sound uttered by another, calls attention to it, makes its audi-
tory image more definite and vivid and thus associates it more firmly
with the movements of articulation. By means then of his original
instinctive reflexes, his random movements, and the imitations of his
movements, which he sees made by others, the child forms a sufficient
fund of motor associations to explain his imitative acts, without the
need of assuming any special instinct, endowment or faculty of imita-
tion. Thus far we have considered only the first simple imitations of
the child. But if these are not instinctive certainly none are. The
more elaborate and complex imitations of later life are simply com-
binations and coordinations of the elemental movements already
learned. As imagination can create nothing for which the ideational
experience of the past does not supply material, so imitation can per-
form no act for which earlier motor experience does not furnish the
elements.

Closely connected with imitation is sympathy. This connection is
illustrated by an instance of my own son's babyhood. Whenever we
repeated to him the nursery ditty — ' This little pig goes to market,'
etc., he would always break out crying when we said the last line —

DISCUSSION. 141

* This little pig cried Wee, Wee, Wee 1 I can't find my way home.'
The 4 wee, wee, wee ' uttered in a somewhat pathetic tone of voice
produced a cry which was certainly imitative, and perhaps in some
measure sympathetic. One other example may be given — an incident
related to me by a friend of his little two-year-old daughter. One day
as the father was holding the child in his arms playing with her, she
got quite a hard bump against his forehead and began to cry. To dis-
tract her attention, he took her hand, patted his forehead with it, and
said, ' poor papa, poor papa ! ' The next day the child got a bump
against a chair, and her father was much surprised to see her pat the
chair and hear her say * poor chair ! ' Certainly no one would claim
that we inherit an instinct to sympathize with injured chairs and I
believe there is as little ground for making sympathy an instinct in
any case.

James again counts sympathy among the instincts,1 and Baldwin
regards organic sympathy as instinctive.* But much the same diffi-
culties arise when we try to think how the vast variety of our sym-
pathies can be accounted for by an inherited nervous mechanism as in
the case of our imitations. In order to explain the first sympathetic
feelings we need only to add to imitation the James-Lange theory of
emotions. As already explained, the child cries when he hears
another cry. He sees another in an attitude of pain or grief and imi-
tatively assumes that attitude, but the assumption of the attitude excites
the corresponding emotion and the child thus acquires the sympathetic
feeling. But even if we do not accept the theory that the ' expression '
makes the emotion, it is still unnecessary to regard sympathetic
feelings as instinctive. The child does not feel the woes and joys of
others until he has already experienced at least elemental feelings of
the same sort in his own person. Having established then a certain
fund of association between feelings and their outward expression, the
imitation of the expression of feeling that he sees in others will
awaken the kindred emotions in his own breast.

One objection to this experiential theory of imitation should be
considered. It seems to be the social psychologists, and those who
have approached the subject from the sociological side, who are espe-
cially inclined to regard imitation as an instinct, an endowment, or a
quasi-faculty. When we think of the tremendous r61e which imitation
plays in the development and education of the individual and in the
entire organization and structure of society, can we regard it as any-

1 Principles of Psychology, Vol. II., p. 410.
8 Social and Ethical Interpretations, p. 222.

H2 THE MECHANISM OF IMITATION.

thing less than a native endowment ? Or, to put the question more
concretely, if imitation is only one type of ideo-motor action, why
should it acquire such a preponderant influence? Why should the
ideas enacted before us in the doings of our fellows take precedence in
their influence on conduct over the numberless ideas of all sorts that
crowd the mind ? In answering this question we may note first that
when a number of motor ideas are in the mind it is the one which is
most interesting, the one which is most vividly apprehended, or which
most forcibly strikes the attention, that is sure to be executed. From
the first experience of parental care, to the last days of life, our social
relations are of the utmost importance to us. There is therefore a
constant and accumulative development of social interests. Where our
interest is there our attention is. The child might wave its arms in
imitation of the waving branches of a tree were it as interested in the
tree as in the movements of its parents, brothers and sisters. Then
too the movements actually enacted are much more clearly and vividly
apprehended than those that are suggested by mere words or ideas
that arise in mind. In fact many activities that can readily be under-
stood when seen, can hardly be described at all. The gymnasium
instructor may be quite unable to tell in words how he performs a cer-
tain feat, but by performing the feat in the sight of his pupils, they
get a far clearer apprehension of it than words can possibly give.
Moreover the very infertility of our minds as to ideas of conduct,
makes us prone to follow the examples presented before us. We are
endowed with a store of energy, we are bound to act, and given the
proper motor associations the doings depicted before us offer the line
of least resistance to our conduct. Lack of originality is as manifest
in actions as in ideas, and it is therefore not to be wondered at that
the doings made vivid by the acts of others, and enforced by the grow-
ing power of our social interests should have a dominant influence on
our conduct. There is nothing then, it seems to me, in the impor-
tance of the role which imitation plays in individual and social life to
invalidate this simple experiential theory of its workings.

F. C. FRENCH.
UNIVERSITY OF NEBRASKA.

N. S. VOL. XI. No. 3 May, 1904

THE PSYCHOLOGICAL REVIEW.

THE LAW OF ATTRACTION IN RELATION TO
SOME VISUAL AND TACTUAL ILLUSIONS.

BY PROFESSOR HAYWOOD J. PEARCE, PH.D.,
From the Psychological Laboratory of Brenau College.

In working over the results of some experiments which were
published under the title, * Ueber den Einfluss von Nebenreizen
auf die Raumwahrnehmung,'1 in which was demonstrated a
tactual illusion similar in nature to the Miiller-Lyer visual
illusion, I observed that the influence of a secondary stimulus
(Nebenreiz) in producing an elongation of a primary linear
stimulus was directly proportional to the intensity of the secon-
dary stimulus and inversely proportional to the square of the
distance between the two stimuli. The number of specific
instances upon which this observation was based was somewhat
small and the number of variations in distance and in intensity
of secondary stimuli were very limited. Moreover, data for
determining the influence upon the result which might be oc-
casioned by the variation in intensity of the primary stimulus
were almost wholly lacking. On account of these and other
similar deficiencies, it was not thought wise at the time of the
former publication to propose a hypothesis of such apparently
far reaching significance.

Starting, however, with this hypothesis in mind I have
directed numerous other experiments, designed to reveal the
exact relation between primary and secondary stimuli and the
effect in perception of the one upon the other. Inasmuch as it
was so clearly shown 2 that the tactual illusion was similar in

1 Archivfur die gesamte Psychologic, Vol. I., pp. 31-109.
'Ibid.

'43

144 HAYWOOD J. PEARCB.

almost every detail to the visual illusion, I selected the latter for
the investigation. The visual figure lends itself more readily to
small and numerous variations, the mean variation in any series
of judgments is much smaller and it seems to me that the opera-
tion of a law such as that indicated would be more easily detected
because more uniform in its manifestations.

The present paper, therefore, is a report of some experi-
ments made for the purpose of determining quantitatively the
influence exerted by secondary visual linear stimuli upon a
primary visual linear stimulus, or perhaps more accurately
stated, the attempt is to determine in mathematical terms the
attractive force operative between two visual stimuli. It will be
understood, of course, that the limitations of language confine
us to the use of the term « stimulus.' If there be, in truth, any
attractive force operating between the actual objective stimuli,
it is certainly not our present purpose to determine that. Stimu-
lus, as I have used the term, represents an * impression ' made
upon a sense organ. It is not yet, necessarily, sensation ; it is
certainly no longer stimulus. It is rather a middle state, viz.,
a state of the nervous system occasioned by objective stimulus
and the conditioning element in sensation.

I began, first, a series of experiments with a figure of the
Miiller-Lyer type. The projecting arms were turned outward,
and separated from the central line, or primary stimulus, by
small open spaces as represented in Fig. i (Plate I.).

In general, the results were of the sort which I had ex-
pected, but it soon became manifest that another factor in ad-
dition to distance and sensation intensity was playing a part in
the results. It is a well-known fact that the Miiller-Lyer illu-
sion varies with the cosine of the angle formed by the project-
ing arm and the central line.1

In order to eliminate this third variable factor, I abandoned
the Miiller-Lyer figure entirely and constructed a figure with
one central linear stimulus and two other simple linear stimuli,
which I have termed the secondary stimuli. The secondary
stimuli were constructed exactly in the line of direction of the

1 Heymans, Quantitative Untersuchungen ueber das optische Paradox in
Zeitschriftf. PsychoL, Vol. IX., p. 221.

REVIEW, Vol.. XI., 1904.

Pi. AT* I.

FIG. i.

FIG. 2.

FIG. 3.

LAW OF ATTRACTION. 145

central, or primary, stimulus but separated from it by small
open spaces.

The accompanying Fig. 2 makes clear this construction.
AB is the primary stimulus, EC and ZJ/^are the secondary stim-
uli. The distance between primary and secondary stimuli is
measured of course from center to center and is xy in the
figure.

It will be observed that several radical variations of this
figure may be made. We may vary (i) the distance xy, or (2)
the stimulus AB> or (3) the stimuli EC and DF, retaining in
each case all other factors constant. The results of our experi-
ments, therefore, fall naturally into three groups : (i) The
effect of secondary stimuli at different distances, (2) the effect
of secondary stimuli when the primary stimulus is varied, (3)
the effect of secondary stimuli of different intensities upon
primary stimuli of a constant intensity and at the same relative
distances.

METHOD OF THE EXPERIMENT.

Preliminary experiments were conducted in order to deter-
mine the best method, (i) A card upon which was drawn the
figure to be judged was given the subject and, in addition, a
series of cards containing each a single line but of different
lengths. The subject was required to select from the series of
cards the one containing the line apparently equal in length to
the primary stimulus in the figure. (2) Instead of a series of
cards, a single card upon which was drawn a series of lines of
different lengths was used. The subject was required to desig-
nate in the series of lines the one which appeared equal to the
primary stimulus. (3) Instead of a series of lines, a single
long line was drawn upon a piece of cardboard, and the card-
board was adjusted to slide back and forth through a slit cut in
another piece of cardboard. By this means it was possible for
the subject to make the line longer or shorter until it seemed to
him equal in length to the primary stimulus.

The method finally adopted and which I think will be recog-
nized as the most convenient and accurate of the four methods
which were tried, was as follows : A frame, three feet square,
was hung upon two upright posts which projected three feet

H6 HAY WOOD J. PEARCE.

above a low table. In this frame was fixed a square of card-
board containing near the center two rectangular openings. On
the rear side of the frame and parallel to the openings were
tacked wooden runners or grooves, so adjusted that one could
slide certain cards, containing the figures to be judged, into their
proper positions filling the rectangular openings just referred to.
When in position and ready for the experiment, the frame ap-
peared to the subject as shown in Fig. 3. The upper single
line could be lengthened or shortened by sliding the card back
and forth. On the back of this card was a millimeter scale, so
arranged that the experimenter could read immediately the
length of the line as it appeared to the observer. The experi-
menter, seated behind the screen at the table, could move the
card easily back and forth and record immediately the reading
of the millimeter scale, which registered the judgment of the
observer.

The observer was seated in front of the screen at a distance
of 80 cm. His task was to observe the moving upper line and
the lower stationary figure at the same time and to say * stop '
as soon as the difference between the upper line and the lower
primary stimulus ceased to exist. This form of instruction to
the observer was adopted because it was noted that if told to
say stop when the two lines appeared to be equal, the subject
adopted, somewhat irregularly, either of two courses : (i) She
said * stop ' when the difference ceased, or (2) having allowed the
variable to pass the point where the difference ceased, she said
* stop ' not until she began to perceive a difference in the other
direction. In order to secure relative constancy, the former
type of reaction was insisted on. In half the experiments
constituting a series, the moving, variable line was gradually
lengthened and in the other half of the series this line was
gradually shortened.

METHOD OF ESTIMATING THE INFLUENCE OF THE

SECONDARY STIMULUS.

It seemed natural at first thought to estimate the influence
of the secondary stimuli as equal to the difference between the
length of the primary stimulus as given by objective physical

LAW OF ATTRACTION. 147

measurement and the length of a second line which is judged
by the subject to be of the same length.

Preliminary experiment, however, showed clearly that when
a subject attempts to estimate the length of a single line (with-
out secondary stimuli), using the method above described, the
judgment is always too small, t. £., the line is always judged to
be shorter than it actually is. Consequently, in order that the
secondary stimuli may produce subjectively an elongation of the
line objectively given, the tendency to shorten just observed, must
first be overcome. Inasmuch as the addition of a secondary
stimulus accomplishes this, we must include this in our estimate
of the influence of the secondary stimulus. Accordingly in
every series of judgments of a line accompanied by secondary
stimuli, I have required evenly distributed judgments of the
same line without secondary stimuli. The results, therefore,
which appear in the tables as « influence of the secondary stimu-
lus ' always represent the difference measured in centimeters
between the judgment of the line without secondary stimuli and
the judgment of the length of the same line with secondary
stimuli.1

The foregoing is in general the method of experimentation
and computation of results employed in each of the three groups
of experiments which follow. In connection with each group
some further details of method must be pointed out.

'Professor Judd (Genetic Psychology for Teachers, p. n) has attempted to
explain the fact that if one tries to draw upon paper a line equal in length to a
copy upon the blackboard, he invariably makes it too short, as due to the larger
environment represented by the blackboard as compared with the smaller
environment represented by the sheet of paper. The facts brought out in my
experiments seem to throw serious doubt upon the adequacy of Professor Judd's
explanation. In my experiments, the environment of the two lines judged to
be equal was the same, and moreover, the error remained the same when the
relative position of the two lines was reversed.

The explanation of the error is probably as follows : When I am comparing
two lines, one standard and the other variable, the latter is the one which is
kept most prominently in the foreground of attention. The eye wanders to the
standard only to renew the memory of its length. What actually happens is a
comparison of a present vivid, intense sensation with a fading memory image.
I may fixate the standard but by the time my eye reaches again the line which
I must make equal to the standard, the latter has become a memory image, or
at best appears upon the periphery of vision and consequently has less sensation
value than the same image upon the fovea centralis or in the focus of attention.

148 HA YWOOD J. PEARCE.

THE EFFECT OF Two SECONDARY STIMULI UPON A PRIMARY
STIMULUS WHEN THE DISTANCE is VARIED.

I present first three tables, L, II. and III., showing results
for a primary stimulus of 16.0 cm., 17.0 cm. and 18.0 cm.,
respectively, in length. The experiments with the three dif-
ferent primary stimuli, though recorded in different tables, were
conducted simultaneously. For example, a card containing a
primary stimulus 16.0 cm. and secondary stimuli 9.5 cm. distant
was presented to the subject. Five successive judgments of
this same line were required, the variable line being first length-
ened and then shortened and so on alternately for the five judg-
ments. Then a second card containing primary stimulus 17.0
cm. was presented and five successive judgments of this line in
a similar order were required. Then primary stimulus 18.0
cm. was presented and the same judgments required. Now we
return to primary stimulus 16.0 cm. but one in connection with
which the distance of the secondary stimuli has been slightly
increased, viz., 10.0 cm. Then the series 17.0 cm. and 18.0
cm. with similar increase in distance of secondary stimulus are
taken, and then back again to primary stimulus 16.0 cm. with
distance of secondary stimuli still further increased and so
through the entire series of five variations in distance for each
of the three primary stimuli. Including the three cards which
contained only a single line each, to which reference has previ-
ously been made, there were eighteen different cards and five
judgments of the stimulus on each were required. Such a
series could be made in about a half hour, which was the length
of a setting for each subject.

At the second sitting, the experiment was conducted in a
similar manner, except that the detail in every particular was
reversed. Two sittings afforded a series of ten judgments each
for each of the eighteen primary stimuli. The tables show
results for ten different subjects and each result given is the
average of ten individual judgments made at two different sit-
tings. There are two such series for each subject and the
general average for the ten subjects represents in each case two
hundred individual judgments.

LAW OF ATTRACTION.

149

The subjects used were of varied age and character. One
is my colleague, Professor Essary, of the department of biology,
to whom I am under especial obligation ; another was a student
in the department of psychology ; a third was a special student
of painting, and a fourth was a special student of music. The
remaining six were taken indiscriminately from the preparatory
school of Brenau College and vary in age from ten to fourteen
years. All except the first mentioned are female.

TABLE I.

INFLUENCE OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG UPON A

PRIMARY STIMULUS 16.0 CM. LONG AT DISTANCES 9.5,

10.0, 10.5, ii. o AND 13.0 CM.

Distances.

9.5 cm.

10.0 cm.

10.5 cm.

n.o cm.

13.0 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

of Secon-
dary Stim-

in cm.

in cm.

in cm.

in cm.

ulus in cm.

E.

I

1-49

1.18

I. II

.46

.04

2

1.61

1.16

.81

•32

.08

Pa.

I

1.17

.81

.72

.18

— .01

2

1.18

•71

.90

.06

—.18

B.

I

1. 06

I.IO

.78

•46

•35

C.

2
I

.86
1.49

•54
1.46

•31
i-35

£

—.04
•47

2

1-75

1.47

1.27

.92

•56

H.

I

2.16

i-59

I-5I

1. 00

•73

2

1.92

1.90

1.81

1.56

1.08

Pi.

I

1-57

1.42

.92

•52

.29

2

2.OO

.82

.69

.98

.18

Pr.

I

3-30

3-°7

2-57

1.78

I.IO

2

1.29

-87

•55

.09

— .22

G.

I

2.09

1-56

1.29

1.03

.24

2

1.88

1.74

1.29

1.06

.77

Pp.

I

2. 02

1.81

1-53

•95

•74

2

1-97

1.64

1.16

•95

.71

Hn.

I

1.98

1.88

1.50

1-30

.89

2

2.03

1.55

1.43

-56

Average.

1-74

1.41

1.18

.80

•44

£X&

I57-03

141.00

130.09

96.80

74.36

The first three mentioned had some knowledge of optical
illusions and the first two were acquainted in part with the
hypothesis upon which I was working. The others had no
knowledge of the nature or object of the experiment except that
which was gained as a result of their own observation in the
progress of the same.

HAY WOOD J. PEAR CM.

Turning to an examination of the results shown in the
tables we find that, with a very few exceptions, there is a uni-
form decrease in the influence of the secondary stimuli corres-
ponding to an increase in the distance between the primary and
secondary stimuli. The majority of the exceptions to be noted
will be found in Table II., in which are shown the results for
primary stimulus 17.0 cm.

TABLE II.

OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG, UPON A
PRIMARY STIMULUS 17.0 CM. LONG AT DISTANCES 10.0,

10.5, II.O, II.5 AND 13.5 CM.

Distances.

10.0 cm.

10.5 cm.

n.o cm.

11.5 cm.

13-5 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

of Secon-
dary Stim-

in cm.

in cm.

in cm.

in cm.

ulus in cm.

B.

I

I.I5

•75

•93

.25

•44

2

1.42

I. II

.53

•15

.00

Pa.

I

•73

•97

•57

.20

•19

2

.96

.68

.72

.40

—.16

B.

I

1.06

•95

.46

.05

—.18

2

1-13

•93

.61

.28

.27

C.

I

1.74

1-45

.64

•23

-.60

2

1-54

i-3i

1.42

1. 12

•50

H.

I

1.70

1. 60

1.18

1.04

.18

2

1.68

1.49

1. 60

1.28

.89

Pi.

I

1.05

•57

.56

'37

.18

2

1.41

1. 01

•44

.07

—.09

Pr.

I

i-75

1.61

1-49

I.OO

.66

2

1.67

i. 08

i. ii

.41

.46

G.

I

1.94

1.27

.90

•71

.20

2

1.41

1.22

1.16

.81

.98

Pp.

I

1.83

1.58

i. 06

.83

.46

2

1.62

i-39

1.17

1. 02

.40

Hn.

I

1.79

1.41

i-45

•99

•47

2

1.76

1.26

1.17

.62

•45

Average.

1-47

1.18

.91

•59

.29

EX&

147-00

130.09

IIO.II

78.03

52-85

The only explanation which I can offer for the greater ir-
regularities manifest in Table II. is the fact that a stimulus
17.0 cm. cannot be distinguished with certainty from either 18.0
cm. or 16.0 cm. and inasmuch as the 17.0 cm. stimulus in the
order of the experiment follows sometimes the 16.0 cm. and
sometimes 18.0 cm. stimulus, the judgment when it related to
the 17.0 cm. stimulus was unequally influenced by the preceding

LA \V OF ATTRACTION.

'5'

TABLE III.

INFLUENCE OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG, UPON A

PRIMARY STIMULUS 18.0 CM. LONG AT DISTANCES 10.5,

II.O. II.5, 12.0 AND 14.0 CM.

Distances.

10.5 cm.

ii.ocm.

1 1. 3 cm.

is.o cm.

14-0 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

of Secon-
dary Influ-
ence in cm.

E.

I

1-43

•71

.28

•19

—.17

2

1-53

1-23

• 77

.27

—.23

Pa.

I

.64

.45

.29

.41

.16

2

.76

.62

.41

.29

—.28

B.

I

1-32

1. 12

1-25

.60

.58

2

1.50

I.IQ

.67

•50

.OI

C.

I

1.76

1.58

1.27

.86

.OI

2

1.33

I.I7

.91

.90

•45

H.

I

2.18

2. II

1.60

1.16

.18

2

2-55

2.4O

1.80

1-37

.98

Pi.

I

.82

.66

.35

.26

—.15

2

1.99

1.78

1.19

.78

.02

Pr.

I

3-15

3.02

1.61

i. ii

1.34

2

1.23

•92

•49

•59

—.07

G.

I

2.40

1.96

1.66

1-37

.16

2

2.06

1.41

.89

•51

•53

Pp.

I

2.30

2.46

1.98

2.OI

i. 60

2

1.49

I.I3

1.08

.46

•25

Hn.

I

1.38

1.19

I-5I

-57

.63

2

1.42

1.02

.68

.52

•30

Average.

1.66

I.4I

1.07

•75

•32

E^D1

183.01

I70.6I

141.51

108.00

62.72

judgments relating to the 18.0 cm. and the 16.0 cm. stimuli. The
subject was particularly liable to such confusion, because she was
not informed as to the number of primary stimuli which were
used, nor that a primary stimulus of different length was always
introduced when the cards were changed.

Considering the general averages of all results for each dis-
tance, we find that the irregularities referred to have been
eliminated and a consistent decrease in influence corresponding
to each increase in distance is manifest. For example, the in-
fluence of two secondary stimuli, each 2.0 cm. in length, act-
ing upon a primary stimulus 18.0 cm. in length at a distance of
10.5 cm. is found to be 1.66 cm. ; at distance u.o cm. the in-
fluence of the same secondary stimuli is 1.41 cm. ; at distance
11.5 this influence has diminished to 1.07 cm.; at 12.0 cm.
distance the influence is 0.75 cm. and at 14.0 cm. it is 0.32 cm.

152 HAY WOOD J. PBARCB.

influence. This same degree of regularity is manifest in each
of the other tables.

An attempt to establish anything like an exact proportion
between the decrease in influence and the increase in the square
of the distance was a failure. It became at once apparent that
the decrease in influence was far more rapid than the increase
in the square of the distance.

The foregoing fact directed attention to another principle,
viz., the intensity of visual stimuli decreases as the stimulus is
moved toward the periphery of the retina. There are not,
within my knowledge, any recorded experimental data which
directly confirm this last statement, and, indeed, the well known
device of the astronomer of using the periphery of the retina
in order to bring to view an otherwise indiscernible star seemed
at first thought in direct contradiction to such a statement.
This astronomical devise, however, only shows really that the
periphery of the retina may under favorable conditions, be
more sensitive to wry faint stimuli than the fovea centralis and
can be explained by the fact that the fovea centralis being con-
stantly bombarded by intense stimuli becomes insensitive to
very weak ones.

On the other hand, the common facts of every day ex-
perience that we see most distinctly when the stimulus falls
upon the center and less distinctly when it is moved toward the
periphery, together with the well known facts of nerve distri-
bution upon the retina afford sufficient confirmation of the
statement that the intensity of the same objective stimulus de-
creases as the stimulus is moved toward the periphery.

Applying this principle to the case under consideration, we
see that when the secondary stimuli are removed to a greater
distance from the primary stimulus, they are removed at the
same time towards the periphery of the retina, inasmuch as the
eye maintains the same position relative to the primary stimulus.
We have, therefore, at the same time, an increase in distance
and a decrease in intensity of the secondary stimulus, although
it remains objectively the same length. Both of these factors
thus entering into the conditions of our experiments call for a
decrease in the influence of the secondary stimulus, according

LAW OF A TTRA C TION. 1 53

to our hypotheses, and the very rapid decrease to which atten-
tion was called is, so far, in confirmatory of rather than contra-
dictory to this hypothesis.

If our hypothesis is valid, it follows that the influence ex-
erted by a stimulus A multiplied by the square of its distance
would equal the influence of a stimulus B multiplied by the
square of its distance. In other words, E x Z>1 = C, in which
E represents the influence of any secondary stimulus, D is the
distance of that stimulus and C is a constant.

As has already been shown, the value of C in the results
previously recorded is not constant. For example, in Table I.
the value for the five distances decreases from 183.01 to 62.72.
This rapid decrease was due to the very rapid diminution of
the value of E, and this last we have attributed to the decreased
intensity of the 2.0 cm. stimulus occasioned by its removal
towards the periphery, in addition to the increased distance.

In order to compensate for this decrease in intensity of the
secondary stimulus, I prepared a new series of figures in which
the same primary stimulus and the same distances were em-
ployed as in the former experiments, but the length of the
secondary stimulus was altered. The amount by which the
secondary stimulus should be altered in length was determined
as follows: I selected arbitrarily one value of C1, viz., that
shown in the second column of each table of results. I then
determined for each distance what the value of E should be,
using the value of C selected as a constant. I was thus en-
abled to determine what effect a secondary stimulus of the same
subjective intensity should have at different distances.

Now at a given distance, we know by experiment the effect
of a secondary stimulus 2.0 cm. in length; we also know for
the same distance, by computation as above shown, what the
effect of a secondary stimulus of a certain standard intensity
ought to be. The problem is to determine how much the second-
ary stimulus shall be lengthened or shortened in order that it
may have the same subjective intensity as the standard.1

For lack of a better, I adopted the purely objective method
of solving this problem, using increase in objective length as

1 Compare Weber's Law.

154 HAYWOOD J. PEARCE.

equivalent to increase in subjective intensity. For example,
referring to Table I., we select the influence of a secondary
stimulus of 2.0 cm. at a distance of n.o cm. as the standard.
The value of C (see second column) in this case is 170.61. We
have assumed that this value should be a constant, if the inten-
sity of the secondary stimulus remained constant. But we find
that the value of C when the distance of the 2.0 cm. stimulus is
11.5 cm. (see third column) is only 141.51, the actual influence
of secondary stimulus being only 1.07 cm. Now if the value
of C were constant the actual influence of secondary stimulus
ought to be 1.29 cm., provided the intensity of our 2.0 cm.
stimulus had remained the same. This conclusion is reached
as follows: The value of C should be 170.61, but as a matter
of fact it is only 141.51. This indicates that the influence of
secondary stimulus (1.07 cm.) is less than is to be expected of a
secondary stimulus equal in intensity to that one which we have
selected as the standard (second column) and, indeed, 1.07 cm.
is as much less than the influence of a secondary stimulus of
standard intensity ought to have been, as 141.51 is less than
170.61. In other words (170.61 x 1.07 cm.) -5-141. 51 = 1.29011.
which is what the influence of a seconday stimulus of standard
intensity ought to be at the distance 11.5 cm. Further, if a
secondary stimulus 2.0 cm. in length has produced an effect of
1.07 cm., how long must the secondary stimulus be in order
that it may produce an effect of 1.29 cm. ? Proceeding accord-
ing to the objective method, this question is answered by the
following arithmetical operation : (1.29 x 2.0 cm.) -k- 1.07 cm. =
2.41 cm., which last is the length which our secondary stimulus
must have at distance 11.5 cm., in order to be equal in intensity
to the standard, which is a 2.0 cm. stimulus at distance of n.o
cm. Proceeding according to this method, I calculated, upon
the basis of results given in the three precding tables, what
the length of the secondary stimulus should be in our new series
of figures, in order that a standard intensity might be maintained
throughout. The method of procedure is unquestionably
crude, and is justified only on the ground that it was used
merely as an empirical device. It is doubtless possible to deter-
mine definitely the relation between increase in subjective inten-

LAW OF ATTRACTION.

'55

shy and increase in objective length of visual stimuli. When
this is done it will doubtless be possible to construct a series of
figures, in which the secondary stimulus at different distances
remains of the same subjective intensity. The time at my
disposal did not admit of such a determination.

The validity of the objection just raised against the method
of constructing the new series of figures was fully justified by
the results of the experiments made with these figures. These
results are shown in Tables IV., V. and VI. In the case of
Table VI., primary stimulus 18.0 cm., the addition to length of
secondary stimulus has produced a result which gives to C a
practically constant value. But in Tables IV. and V. the value
of C appears in a constantly diminishing ratio, showing that

TABLE IV.

INFLUENCE OF SECONDARY STIMULI OF VARIOUS LENGTHS, BUT ESTIMATED
TO BE OF THE SAME SUBJECTIVE INTENSITY OR VALUE, UPON A
LINE 16.0 CM. IN LENGTH AT THE SAME DIS-
TANCES SHOWN IN TABLE I.

Lengths of the Sec-
ondary Stimuli.

1.79 cm.

2. oocm.

3.15 cm.

2.90 cm.

3.70 cm.

Distances.

9-5 cm.

10. o cm.

10.5 cm.

n.ocm.

13.0 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

of Secon-
dary Stim-

in cm.

in cm.

in cm.

in cm.

ulus in cm.

E.

I

1. 10

•71

.69

.68

—.06

2

1.36

I.I7

1. 08

•95

.70

Pa.

I

I-I5

•93

.88

.70

.36

2

.88

.82

.48

•54

.07

B.

I

1.28

•93

• 75

•67

.48

2

1. 21

•9i

.62

•63

•35

C.

I

1-37

1.05

•77

.60

2

1.36

• 14

•72

.69

•48

H.

I

1.27

.04

.80

•67

—.09

2

1.3°

.96

•73

•43

.21

Pi.

I

1.40

.34

i. 20

1.06

.89

2

1.60

.61

1.42

1.22

1. 08

Pr.

I

1.92

.66

1.22

I. II

•50

2

1.72

•39

1.25

•99

.80

G.

I

2.00

.64

1-55

1.43

.51

2

1.29

•56

.69

•39

Pp.

I

1.63

•44

1.25

.88

•63

2

1.40

•34

1.20

1.06

.89

Hn.

I

1-55

1.36

1.05

1. 01

.89

2

1.79

1.80

1.67

1-34

•96

Average.

1-43

1.20

I.OI

.86

.53

£X&

129.06

1 20.0O

113.52

104.06

89-57

HAYWOOD J. PEARCE.

the addition in length was not sufficiently large. In other words
we have a somewhat new verification of Weber's Law, viz.,
equal increments in objective length of visual linear stimuli do
not imply equal increase in subjective intensity of the visual
stimulus.

Comparing Tables IV., V. and VI. with I., II. and III.
respectively, we find that the increase in the length of the

TABLE V.

INFLUENCE OF SECONDARY STIMUW OF VARIOUS LENGTHS, BUT ESTIMATED
TO BE OF THE SAME SUBJECTIVE INTENSITY OR VAI,UE UPON A
LINE 17.0 CM. IN LENGTH, AT THE SAME DIS-
TANCES SHOWN IN TABI^E II.

lengths of the Sec-
ondary Stimuli.

1.77 cm.

2. oo cm.

3.35 cm.

3.32 cm.

4.90 cm.

Distances.

1.00 cm.

10.5 cm.

n.ocm.

11.5 cm.

13-5 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

Secondary
Stimulus
in cm.

of Secon-
dary Stim-
ulus in cm.

E.

I

1.22

1. 12

1.03

•97

.42

2

1.23

.90

.81

.72

•43

Pa.

I

.80

.84

.41

.23

.02

2

.89

.65

.40

•25

• 15

B.

I

1. 2O

.85

.86

.58

.48

2

1.24

1.08

•77

•57

•33

C.

I

1-47

1.28

i. 02

•93

•30

2

•77

•87

.62

.64

.20

H.

I

1.70

1-54

1.50

1.50

1.16

2

.88

•75

.72

•7i

•51

Pi

I

1.14

1.07

1.03

•97

•95

2

I.OO

I.OI

.80

•45

.27

Pr.

I

1.61

1-39

1-35

1.04

.96

2

I.OO

•75

.86

•58

•50

G.

I

1.76

1.49

1.29

1.38

.84

2

1. 10

•79

•77

.48

•35

Pp.

I

1.77

1.46

1.28

i-35

• 71

2

i-53

I.O2

i. 06

•9i

•34

Hn.

I

1.98

1.63

•93

•98

.68

2

i. 20

I.I9

•94

.69

•58

Average.

1.28

1.08

.92

.80

•5i

EX&

128.00

119.07

111.32

105.80

92.95

secondary stimulus has greatly increased the constancy of C, in
other words the effect of the secondary stimulus of increased
length has been uniformly greater. It appears, therefore, more
than probable that if the length of the secondary stimulus were
increased according to subjective rather than objective standards,

LAW OF ATTRACTION.

'57

the value of C would become really constant as is demanded
by our hypothesis.

TABLE VI.

INFLUENCE OP SECONDARY STIMULI OP VARIOUS LENGTHS, BUT ESTIMATED
TO BE OP THE SAME SUBJECTIVE INTENSITY OR VALUE, UPON A
LINE iS.o CM. IN LENGTH, AT THE SAME DIS-
TANCES SHOWN IN TABLE III.

Lengths of the Sec-
ondary Stimuli.

1.85 cm.

2.00 cm.

2.41 cm.

3-54 cm

5 43 cm

Distances.

10.5 cm.

xx.ocm.

11.5 cm.

ta.o cm.

14.0 cm.

Influence of

Influence of

Influence of

Influence of

Influence

Subject.

Series.

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

Secondary
Stimulus

of Secon-
dary Stim-

in cm.

in cm.

in cm.

in cm.

ulus in cm.

E.

I

.41

1-35

I.I9

1.29

.48

2

•39

1.25

1.22

1.30

1.18

Pa.

I

•52

1.40

1-25

I.I5

1.08

2

•34

1.23

1.06

•97

•72

B.

I

.61

1-36

1.26

1. 10

•95

2

.40

1.24

i-59

1. 08

.98

C.

I

.02

1.92

1.76

1-57

2

-39

1-30

1. 14

•83

.84

H.

I

.32

1.36

1.30

1.18

1. 10

2

.81

.64

.48

.40

•*3

Pi.

I

.28

1-34

1.27

1.18

1.02

2

.66

1.47

i-37

1-25

1.28

Pr.

I

•93

1.64

1.72

1.65

1.47

2

•99

.80

•58

•57

.64

G.

I

.48

1.07

1-35

1.06

1.02

2

.11

1.18

.84

.68

•43

Pp.

I

.28

1.40

1.36

i-34

1.07

2

•9i

.1.87

1-45

1.18

Hn.

I

2.O2

1-55

1. 21

1.02

1.18

2

1-37

1.26

•94

1. 12

•63

Average.

1.46

1-32

1.22

I.II

•94

EX&

160.96

159-72

161.34

I59-84

184.24

For purposes of comparison, I present three tables, VII. , VIII.
and IX., showing the results of some experiments conducted by
Misses E. Dickson and B. Brock, students in my laboratory
course. In these experiments only four subjects were used and
ten tests of each was made with each figure. Thus each of the
general averages represents forty judgments.

The primary stimulus in these experiments was 24.0 cm.,
25.0 cm. and 26.0 cm., respectively. The secondary stimulus
was objectively 2.0 cm. in length, the same as in the experi-
ments previously recorded, but inasmuch as the increased length
of the primary stimulus makes it necessary to remove the

i58

HAYWOOD J. PEARCE.

secondary stimulus towards the periphery, the subjective intensity
of the secondary stimulus is materially decreased as compared with
the secondary stimuli of the experiments previously discussed.

Comparing these results with those of Tables I., II. and
III., we find the same general features, viz., decrease in inverse
proportion to distance, but a decrease more rapid than is
demanded by increase in square of the distance.

Detailed comparison of results in the several tables, bring
out some interesting relations. For example, in Table I., where
we have primary stimulus 18.0 cm., secondary stimulus 2.0 cm.
and distance 14.0 cm., the influence of secondary stimulus is
0.32 cm. In Table VII. for the same distance, but a primary

TABLE VII.

INFLUENCE OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG UPON A PRI-
MARY STIMULUS 24.0 CM. LONG AT DISTANCES 13.5, 14.0,
14.5, 15.0 AND 16.0 CM.

Distances.

13.5 cm.

14.0 cm.

14.5 cm.

15.0 cm.

16.0 cm.

Subject.

Influence of
Secondary
Stimuli

Influence of
Secondary
Stimuli

Influence of
Secondary
Stimuli

Influence of
Secondary
Stimuli

Influence of
Secondary
Stimuli

in cm.

in cm.

in cm.

in cm.

in cm.

A.
Kg.
Ch.

1.4
2.2
1.2

.80
1.2
I.I

I.O
.8

•9
— .2

.8

•7
—•7
.6

Cl.

I.I

•9

.8

.6

.0

Average.

1-5

I.O

•9

•5

•15

stimulus of 24.0 cm. the influence is i.o cm. Similarly, com-
paring the effect when primary stimulus is 25.0 cm., and the
same distance, 14.0 cm., we find, in Table VIII. an effect of

TABLE VIII.

INFLUENCE OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG UPON A PRI-
MARY STIMULUS 25.0 CM. LONG AT DISTANCES 14.0, 14.5,
15.0 AND 15.5 CM.

Distances.

14.0 cm.

14.5 cm.

15.0 cm.

15.5 cm.

Subjects.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

A.

II:

Cl.

•7

1.2

1.6

1-7

.8

.8
1-3
1.4

2

•5

I.O
I.O

^ ___ ly

—•4
.8

•5

Average.

1-3

I.I

• 7

•05

LAW OF ATTRACTION.

'59

1.3 cm. Similar relations appear throughout, when the influ-
ence in the case of different primary stimuli with secondary
stimuli at the same distances is observed. The effect of the
secondary stimulus increases not only in proportion to its own
intensity, but also in proportion to the intensity, or length, of
the primary stimulus.

THE EFFECT OF SECONDARY STIMULI WHEN THE PRIMARY

STIMULUS is VARIED.

This relation to which reference was made in the preceding
paragraph, was also brought out by a series of experiments
especially designed for the purpose. A series of eight cards
were prepared as follows : The primary stimuli were 8.0, 9.0,
10.0, ii. o, 12.0, 13.0, 14.0 and 15.0 cm. long; the second-
ary stimuli were in each case 2.0 cm., in length and the dis-
tance between primary and secondary stimuli was in each case
9.0 cm. The only variable factor, therefore in the conditions
was the length of the primary stimulus.

TABLE IX.

INFLUENCE OF Two SECONDARY STIMULI, EACH 2.0 CM. LONG UPON A PRI-
MARY STIMULUS 26.0 CM. LONG AT DISTANCES 14.5, 15.0,
15.5 AND 17.0 CM.

Distances.

14.5 cm.

15.0 cm.

15.5 cm.

17.0 cm.

Subjects.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

Influence of
Secondary Stimuli
in cm.

A.
Kg.
Ch.
Cl.

1.8
i-7
1-5
1-5

1.8
1.6
i-3
i-3

.6
1.2
I.I

•5

• 7
.1
.8
•4

Average.

1.6

1-5

•9

•5

The method of conducting the experiments was in general
similar to that already detailed. The cards were presented in
the order given above for the first five tests of each series and
in the reverse order for the last five tests of each series — ten
tests constituting a series. The experiments were conducted
by Misses Newton, McConnell and Pauline Smith, three
students in the department of psychology. The results are
shown in Table X.

i6o

HA YWOOD J. PEARCE.

TABLE X.

INFLUENCE OF Two SECONDARY STIMULI, 2.0 CM. LONG, UPON PRIMARY

STIMULI OF VARIED LENGTHS, BUT SAME RELATIVE

POSITION AND DISTANCE 9.0 CM.

Lengths of Primary
Stimuli.

8.0 cm.

9.0 cm.

10.0 cm.

n.o cm.

12 .0 cm.

13.0 cm.

14.0 cm.

15.0 cm.

Subjects.

No. of
Experi-
ments.

Influence

Influence

Influence

Influence

Influence

Influence

Influence

Influence

Me.
N.
K.

IOO
IOO
IOO

.24
.06
.28

.38
.17
•31

.40
.11
.06

•37
.24
.04

•51
.46
•43

•52
•03
•25

.61
•53
•36

•94
.28

.22

Average.

.15

.29

.19

.22

•47

.27

•50

.48

For purposes of comparison, I present also Table XL, which
records the results of experiments conducted under exactly
similar conditions by Misses Canning and Blalock. In this
group, however, the lengths of the several primary stimuli were

TABLE XI.

INFLUENCE OF Two SECONDARY STIMULI, 2.0 CM. LONG, UPON PRIMARY

STIMULI OF VARIED LENGTHS, BUT SAME RELATIVE

POSITION AND DISTANCE 14.0 CM.

Lengths of Primary
Stimuli.

iS.o cm.

19.0 cm.

20.0 cm.

21.0 cm.

22.0 cm.

23.0 cm.

24.0 cm.

25.0 cm.

Subjects.

No. of
Experi-
ments.

Influence

Influence

Influence

Influence

Influence

Influence

Influence

Influence

Bl.
Cu.

IOO
IOO

.00
.03

.07
.12

.OI
.12

.04
.16

.21
.63

•29
.36

.64

•35

•32
.78

Average.

.015

.10

.07

.IO

.42

•33

•50

•55

18.0, 19.0, 20. o, 21. o, 22. o, 23.0, 24.0 and 25.0 cm., and the
distance between primary and secondary stimuli was 14.0 cm.
Referring to the summary of results in both tables, it will
be observed that in general the influence of secondary stimuli
at the same distances increases as the length of the primary
stimulus increases. Irregularities, however, in the rate of
increase and actual exceptions to the rule are particularly
noticeable. The fact that relatively speaking the same incon-
sistencies appear in both tables would seem to indicate that the
order in which the cards were presented was a factor which

LAW OF ATTRACTION.

161

affected the result and that the natural ebb and flow of attentive
processes was involved. In addition, it should be observed that
linear stimuli of the lengths here given cannot be distinguished
from one another with any considerable degree of certainty,
unless they differ by more than one cm. in length.

We may eliminate these inconsistencies in the two tables
by taking an average of three different lengths of primary stim-
ulus. For example : The results in the case of primary stim-
uli 8.0, 9.0, and 10.0 cm. were 0.15, 0.29, and 0.19 cm.,
respectively. The mean of 8, 9, and 10 is 9 ; the average of
15, 29, and 19, is 21. Therefore the corrected result for primary
stimulus, 9.0 cm., would show an influence of secondary stimu-
lus equal to 0.21 cm. Similarly the corrected result for primary
stimulus 10.0 cm., shows an influence of 0.23 cm. When
Tables X. and XL, are corrected according to the method just
outlined the average appears as shown Table XII. Here we
see a consistent increase in influence of secondary stimulus
corresponding to increase in length of primary stimulus.

TABLE XII.

THE RESULTS SHOWN IN TABLES X. AND XI. WHEN REARRANGED AS
DESCRIBED IN THE TEXT.

Lengths of Primary
Stimuli.

9.0 cm.

10.0 cm.

n.o cm.

13.0 cm.

13.0 cm.

14.0 cm.

15.0 cm.

Average.

Table X.

.21

•23

.29

•32

.41

.42

Lengths of Primary Stimuli.

19.0 cm.

20.0 cm.

21.0 cm.

23.0 cm.

33.0 cm.

14.0 cm.

Average.

Table XI.

.06

.09

•19

.28

•42

.46

A closer examination of the two summaries of results
shown in Table XII. reveals another fact which has entered as
a disturbing element into my experiments and which I have
not succeeded in satisfactorily isolating. I refer to the fact
that the stretch of open space between the primary and secon-
dary stimuli enters as an element in determining the influence
of the secondary stimuli. When variations in the size of this
open space are small its influence upon the result may perhaps
be disregarded. But in cases of the sort now under discussion,
these open spaces play a considerable part. For example, in

1 62 HAY WOOD J. PEARCE.

Table XII. we see that when length of primary stimulus is in-
creased from 9.0 cm. to 14.0 cm. the influence is increased
from 0.21 cm. to 0.42 cm. Now if we accept the hypothesis
that this influence varies directly as the product of the intensi-
ties of the two sensations,1 and inversely as the square of the
distance between them, then the proportion of increase ought
to be readily determined by use of the well known formula for
the law of gravity. Using this formula (f= C \_(m x m'} -f- -O2])
as a basis of calculation, it will be found that the relation be-
tween the influence in the case of the 9.0 cm. and 14.0 cm.
primary stimuli, will be as 0.22 is to 0.35. The relation as
determined empirically was 0.21 to 0.42. The entire series as
determined empirically is 0.21, 0.23, 0.29, 0.32, 0.41, 0.42.
The entire series as determined by calculation based upon the
formula is 0.22, 0.24, 0.27, 0.30, 0.32, 0.35.

It will be observed that in the latter half of the series there
is an increase in the influence of the secondary stimuli, which
is in excess of that which is warranted by our hypothesis as
represented by the formula in question. It should also be noted
that this unexpected increase in influence is coincident with a
gradual lessening of the open spaces which separate the pri-
mary from the secondary stimuli. In the case of the 9.0 cm.
primary stimulus this open space was 3.5 cm., whereas in the
case of the 14.0 cm. primary stimulus this space was only i.o
cm. This would seem to indicate that close proximity of the
ends of the primary and secondary stimuli, increases the effect
of the secondary stimulus.

The indication just referred to is further emphasized by
reference to the other half of Table XII. Here we have pri-
mary stimuli increasing in length from 19.0 cm. to 24.0 cm.
The series of figures showing the influence of secondary stim-
ulus as determined empirically is 0.06, 0.09, 0.19, 0.28, 0.42,
0.46. A corresponding series calculated upon the basis of the
gravity formula would be 0.19, 0.20, 0.21, 0.22, 0.23, 0.24.
Here too there is a corresponding decrease in the size of the
space which separates the two stimuli.

It is further evident from the foregoing that the disturbing
effect of too close proximity of the ends of the two stimuli is in

1 Intensity of sensation is equivalent to 'sensation mass.'

LA W OF A TTRA C TION. 1 63

proportion to the length of the primary stimuli. This relation
is manifest in a series of experiments recorded in Table XIII.
The experiments were conducted by myself, the subjects being
students of psychology. Three subjects were used and results
for three series of ten judgments each are shown. In the first
column are recorded the judgments of the length of a single
line (4.0 cm.) without secondary stimuli ; in the second column
are shown judgments of the length of the same line when sec-
ondary stimuli (2.0 cm.) have been introduced at distance 3.5
cm. The distance between the end of the primary stimulus
and the end of the secondary stimulus was only 0.5 cm. The
third column shows judgments of a single line 6.0 cm. in length
and the fourth column shows judgments of the same line when
secondary stimuli have been added at a distance of 4.5 cm.
The distance between end points is again only 0.5 cm. And
so with each primary stimulus, viz., 8.0 cm., 10.0 cm. 12.0
cm., and 14.0 cm., the distance between primary and secondary
stimuli is respectively 5.5 cm., 6.5 cm,, 7.5 cm. and 8.5 cm.,
but the distance between end points of primary and secondary
stimuli is in every case only 0.5. cm.

Referring now to the average of results for all subjects, there
is shown a marked increase in influence for each primary stim-
ulus, despite the fact that the secondary stimulus was further re-
moved and probably less intense. It is at once manifest that
the formula under consideration cannot be used to determine
the relative effect of secondary stimuli in the case of such a
series as that which is represented in this table, unless another
element can be introduced into the formula. It is not yet clear
what this element should be. We can only say that close
proximity of the ends of the two stimuli increases the effect of
the secondary stimuli and that this increase in effect is itself
increased in proportion to length of primary stimulus.

INFLUENCE OF SECONDARY STIMULI OF VARYING LENGTH
UPON A PRIMARY STIMULUS OF A CONSTANT LENGTH

AND AT A CONSTANT DISTANCE.

For showing this relation I have not conducted a separate
series of experiments, but have rearranged the results shown in
Tables I., II. and III. and represent them in Table XIV.

164

HAYWOOD J. PEARCE.

3

"— '

»— I n

X

5

H 2

g P

•< O

B

1

I

CO

fe
o

g
§

5

•iinrans

tO%O C* *O C< *n O O ON

p»

8

AjBpuooas

ini<t<*'^-^J-'a--^rin<<a-

•t

M

§

a

0

q

in

00

•sn ntnns

ON P4 to M 00 M VO t^vO

co *>• •* >n o >noo t-» t--.

«

s^ius

tOtOtOtO-^tOfOfOtO

tO

•i nrans

M 10 •* ri-oo »nvo r>- •«*•

^

£

XaBpnooag

to N' pi pi ci P« N p5 pi

«

§

t

q

U1

•sninrans

1000 vo oo *=}-co t^ to in

&

a{8o|8

MMMMWMMMM

HI

•nnmijs

^^vo^join^voo

in

£

Aaupiioosg

o* o o" o" o o o o" o

o

a

u

a

0

o

•sninnnjs

»O ^" Is* **>• **) O O VO *O
CO Is* lOvD M t>«. t^* ONVO

f-

aiSms

O\ O^ O^ ^ O ON O> ON O^

M

ON

•Iinmns

O C* t^ ^vO CO M- ON O

00
CO

V?

1

B

cococooooooocooooo

00

q

CO

m

•sninrans

TJ- Tj- w lOCO to ON O O

.0

aiSujs

l>^^^^^^^t-

^

•Iinnms

PI PI 1/5 invo r^ in o vo
in" too PI P< to-troN

in

~

B

t

vo vo vo vo vo vo vo vo in

vo

0
-O

•mnmns

co MVO ^PI ^»nt>-M

ON ON l^CO ON30 00 00 VO

£

'l3niS

10 m m m m in m m m

in

•nnwns

VO ON *3"VO M M ^" T^- 1/5
Tf O O ON O O P« "*00

M

j

a

a

^-^^to^Tj-^-Trto

"*

q

in
to

•snmraps

•^- ONVO P) OO t-» lOCO to

O CO 00 00 00 CO ONOO t^

§8

Tt to to to to to to tOW

to

tn
tJ

V

to

MPttOMPttOMPJtO

Lengths of
Primary Stimuli.

Distances.

Subjects.

M P ^

M ^'
<J

Average.

Infl. of Sec. Stimuli.

LAW OF ATTRACTION. 165

Unfortunately, the results which are comparable were not
obtained in the same series of experiments, but all the results
for constant secondary stimulus (2.0 cm.) were obtained first
and then results for variable secondary stimulus were obtained
in a subsequent series of experiments with the same subjects.
Inasmuch as the magnitude of an illusion decreases with prac-
tice on the part of the subject ', we find that in the second
series of experiments the influence of the same secondary
stimulus under similar conditions is less than in the first series.
This is seen by comparing the instances in which the length of
the secondary stimulus was the same. For example, when
primary stimulus was 18.0 cm., distance n.o cm. and length
of secondary stimulus 2.0 cm. in both series, the influence in
Series i was 1.41 cm. and in Series 2 the influence was 1.32
cm., showing a decrease in influence due to practice of 0.09
cm. ; similarly, when primary stimulus was 17.0 cm. and
secondary stimuli 2.0 cm., influence in Series i was 1.18 cm.
and in Series 2 it was 1.08 cm., showing a decrease in influence
of 0.80 cm. ; further, when primary stimulus is 16.0 cm., and
secondary stimulus 2.0 cm., influence in first series was 1.41
and in second series 1.20 cm., showing a decrease of 0.21
cm. In comparing the results for the two series therefore we
must either subtract these values from the first or add them to
the second or, perhaps more accurately, subtract one half from
the first and add one half to the second.

Comparisons of individual results of the two series are not
satisfactory because of irregularities and we must resort to a
comparison of averages in order to discover any consistent
relations.

Taking first the results for primary stimulus 18.0 cm. we
find the mean of all the distances used is n.8 cm. ; the length
of the secondary stimulus used throughout the first series is 2.0
cm. ; the mean length of secondary stimulus in the second series
is 3.04 cm., the average influence in first series is 1.04 cm. and
in second series is 1.21 cm., or if corrected as above suggested
the influence in first series is 0.99 cm., and in second series is
1.26 cm. There is thus apparent a more or less direct ratio
between the length of the secondary stimulus and its influence.

1 Cf. Judd, Genetic Psychology for Teachers, p. 26.

1 66 HAY WOOD J. PEARCE.

The results shown in Table XIV. lend themselves, however,
to a more comprehensive treatment and enable us to apply
directly the formula implied by our hypothesis. By this hypoth-
esis, f— C(mx m')--r-Dz in whichy is the force of attraction exist-
ing between primary and secondary stimuli, m is the mass or
intensity of the primary stimulus, mr is the mass or intensity of
the secondary stimulus, D is the distance between primary
and secondary stimuli, and C is a constant which must be em-
pirically determined.

From the results of Table XIV. this constant appears to be
0.339, determined as follows : In case of primary stimulus 18.0
cm. (mxinr)-*-D = 0.258 and the influence as shown above was
0.99 cm. Hence we have 0.99 = C x 0.258, or C=(o.99)-r-
0.258=0.383. Determining C for the six possible instances,
I found the average to be 0.339 with a mean variation of 0.038.

Using the constant thus determined, it will be found by
making proper substitutions that the formula given is an ap-
proximate expression for each of the results obtained by experi-
ment, when the conditions are comparable with the foregoing
and that the consolidation of individual results increases the
perfection of such an approximation.

It should be expressly remarked, however, that the formula
with constant above given, cannot be applied indiscriminately
to all results in which widely varying distances involving vary-
ing intensities of secondary stimuli are included ; nor can it be
applied successfully to cases in which the ends of primary and
secondary stimuli are less far removed from one another than

i.o cm.

MEAN VARIATIONS.

It will be noted that the mean variation does not appear in
the tables. This is because the results shown in the tables always
represent a calculated effect and not a judgment. This effect
was determined by subtracting one series of judgments from
another. A mean variation parallel to the results shown in the
table would have no definite meaning. As regards the judg-
ments made by the subjects I may make the following general
statements : For all judgments the mean variation ranged from
0.2 to 0.8 cm. As a rule the mean variation is somewhat larger

LAW OF ATTRACTION.

.67

TABLE XIV.

COMPARISON OF INFLUENCE OF SECONDARY STIMULI OF DIFFERENT LENGTHS UPON PRIMARY STIMULI OF THE SAME
LENGTH AND AT THE SAME DISTANCE.

Primary Stimulus 18.0 cm.

•ma of
Xjvpaooas

JO 3011JI11JIII

Average Distance ii.S cm.
Average Length of Secondary
Stimulus in Series i = 2.00 cm.
Average Length of Secondary
Stimulus in Series a =3-04 cm.
In Series I (mX**') -t-&* = 0.258
" " i Average Influence
of Secondary Stimuli = 0.99 cm.
In Series 2 (mX**') -*- ^* = 0.393
" " 2 Average Influence
of Secondary Stimuli =1.26 cm.

•mo in

Xj»pnoo3s
jo qjJ8na*i

SSSSSSS'&S?

« M « « « « « «0« t/»

•[[minis

Ul 33UU]SI([

•O O •/» O O

O- H •» M • W • »*•

-M.S

Primary Stimulus 17.0 cm.

•111.-) Ill

jinmiVs
XjBpnoo*s
jo sousuy ui

TJ- 04 i-i O ON ON »OOO W »O

Average Distance 11.3 cm.
Average Length of Secondary
Stimulus in Series i = 2.00 cm.
Average Length of Secondary
Stimulus in Series 2 = 2.87 cm.
In Series i ( mX*»') -+• & =0.266
" " i Average Influence
of Secondary Stimuli =0.84 cm.
In Series 2 (mX**') -*• & =0.382
" " 2 Average Influence
of Secondary Stimuli =0.97 cm.

M M w M d d d d d d

'TO3 III

8R888S8S.88,

•tin urns

A\IBpaOD3S

plIB XjBHlUJ
U33M)3q 'HID

q <o q 10 w»

•s^S

Primary Stimulus 16.0 cm.

•nio tii

** t*J w OOO '-'OvO ^ «O
t^ ^ ^ W »H O OO OO ^ *O

Average Distance 10.8 cm.
Average Length of Secondary
Stimulus in Series i = 2.00 cm.
Average Length of Secondary
Stimulus in Series 2 = 2.31 cm.
In Series i (mX**') -*-/?* =0.253
" " i Average Influence
of Secondary Stimuli = i.oi cm.
In Series 2 (mX**f) -*-D* =0.314
" " 2 Average Influence
of Secondary Stimuli = i.io cm.

•mo in
Jinraijs

8R888S88-8R

•tinniriS
XjBpnoo3S

PUB AjBUllJcI

IH M M *H

•»H»S

MNMC,M«MW~«

168

HAY WOOD J. PEARCE.

when secondary stimuli are introduced and the variation is
larger when the primary stimulus is increased in length.

TACTUAL ILLUSIONS.

In order to compare the visual and tactual illusions and to
show the law of attraction as applied to the latter, I reproduce
from the article ' Ueber den Einfluss von Nebenreizen ' l to which
reference has been made, the results of some experiments with
a tactual illusion similar to the Muller-Lyer visual illusion.
The line was produced by pressure of a thin strip of brass upon
the skin of the forearm. The projections or arms were pro-
duced by pressure of short brass rods drawn to a point. An
apparatus was so constructed that the pressure from line and all
points could be given at the same time. For a more detailed
description of the method and nature of these experiments, the
original article must be consulted.

So far as these results are comparable with results of visual
experiments previously detailed, they appear in Tables XV. and

TABLE XV.

TACTUAI, II^USION. FIGURE SIMILAR TO THE MUI<I,ER-LYER FIGURE WITH
PROJECTING ARMS EXTENDING OUTWARD.

in

a

a

-3

•o

Subject Ke.

Subject W.

Subject M.

Average.

"3

«« a

s

o a

o'-S

V

u

V

tl$w

|b§

O 53 ^

u .

ga

^

o .

a a

u O

^

ga

^

o .

ga

£

X

c HJ a

4) OJ2

~ u

3 °

3 °

3 °

3 a""

E^g***

K«

(a a

O'1*

ai

c a
c '^

si

'H-S

S

q3 d
I1"

s

1

£

1

5*

M

M

M

M

6.0

2.Q

20°

2.9

0.9

2-5

0.5

3-4

04

2.9

0.6

0.

6.0

5-0

20

3-1

I.O

2.9

0.7

4-3

0.4

34

0.7

o.<

7-0

2.9

2O

2.5

0.4

2.5

0.4

0.,

7-o

5-o

2O

34

I.O

34

I.O

o.<

8.0

2.9

2O

2.2

0.7

2.0

0.2

34

0-5

2.5

0.5

0.

8.0

5-0

20

2.4

0.6

2-5

0.4

4.0

0-3

0.4

O.(

6.0

2.9

30

2.8

0.5

1-7

0.4

2.7

0-3

2.3

0.4

O.(

6.0

30

2-3

I.O

1.8

0.2

2.8

o.7

2-3

0.6

o.<

7.0

2.9

30

2.2

0-3

2.2

0.3

7.0

5-0

30

1-5

0.3

1-5

0-3

0.(

8.0

2.9

3°

2.1

O.2

1.2

0.6

3.1

0.9

2.1

0.6

0.

8.0

5-o

30

2.1

0.7

1.8

o.5

34

0.7

2.4

0.6

o.<

XVI. The last column of each of these tables contains the
value of (m x in') -r- D2 arithmetically expressed. Here m equals
1 Archivf. d. Gesamte Psychologe, Vol. I., pp. 31-109.

LAW OF ATTRACTION.

169

TABLE XVI.

TACTUAL ILLUSION. FIGURE SIMILAR TO THE MULLER-LYER FIGURE WITH
PROJECTING ARMS EXTENDING TOWARD THE CENTER.

3

"3

Subject Ke.

Subject \V.

Subject M.

Avenge.

<ol

U.S

OS .

SJ?
EJd

a

&

s?5

Length
>ndary S
in cm

o c3

M B •

V *•

" V'-3
ft*«

fjfc

Dfluence
in cm.

S

fluence i
cm.

S

nfluence
in cm.

jj

8 .
JJ

X

JP

fi

1

a

10.0

2.9

2.S°

,.9

0-7

1.4

0.2

I.O

0.6

•4

0.5

0.71

IO.O

5-0

25

2.2

o-3

1-5

0.3

1.6

0-3

.8

0.3

0.89

12.0

2.9

25

1.8

0.4

1-9

0.7

1-3

0.4

•7

0-5

0.62

12.0

5-o

25

2-5

0.2

1-5

o.o

1.6

0.4

•9

0.2

0.83

10.0

2-9

45

1.8

o-3

0.9

0.3

0.5

0-3

.1

o-3

0.71

10.0

5-o

45

1.6

0.3

I.I

0.5

0.6

0.2

.1

o. T

0.89

12.0

2.9

45

1.6

0.5

1.2

O.2

0.6

0-3

.1

o»3

0.62

I2.O

5-0

45

2.O

0.3

i-3

0.2

0.5

0.2

• 3

0.2

0.83

the length in centimeters of the line or primary stimulus ; m!
equals the distance of the end point of the projecting arm from
the end of the line or primary stimulus. The assumption that
this last is the secondary stimulus is somewhat questionable.
Its justification, so far as there is any, is based upon the follow-
ing facts : (i) The introspective evidence of the subjects showed
that the end points of the line were most prominent in conscious-
ness, and consciousness of stimulation of the skin between the
two end points of the line was very vague and sometimes alto-
gether absent. Consequently the judgment really concerned a
distance between two points (corresponding to the line) influ-
enced by a consciousness of a distance between these two points
and four other points. I have therefore considered these dis-
tances the secondary stimulus rather than the actual points
stimulated.

D in the formula above given is the distance from center of
primary to center of secondary stimulus, measured along the
line which consciousness must inevitably follow. In other
words D is here one half the primary plus one half the second-
ary stimulus.

A comparison of the average E (which here represents the
average influence of the secondary stimulus or, more properly,
the effect of the force of attraction between the primary and
secondary stimuli) with the numerical equivalent of

170 HAY WOOD J. PEARCE.

shows a fairly consistent proportional relation. The accuracy
of the proportion is increased when averages of all comparable
groups is taken.

It must be observed that comparisons of instances in which
the angles are different cannot be made. The intensity of the
secondary stimulus is decreased as its angle increases. We
cannot compare satisfactorily the results of the two tables for a
similar reason.

The figures which are used to represent the intensity of the
secondary stimulus are at best only relative, not absolute, and
hence the reason that we cannot at present complete the formula
and assign a definite numerical value to C.

If, however, comparisons are made of instances in which
the conditions upon which the intensity of m' depends are con-
stant, the accuracy of the proportion existing between the attrac-
tive forces and the values of (m x mf) -r- Z>2 is very striking. For
example, taking data from the first two lines of Table XV., we
have the proportion 2.9 : 3.4 : : 0.90: 0.99 or 3.060 : : 2.871,
there being a difference in the proportion of 0.189 cm. But
inasmuch as there were four secondary stimuli in the experi-
ments recorded, the actual difference in proportion for a single
secondary stimulus would be only 0.047 cm.

The inaccuracy of the proportion in the second half of each
table is largely increased. Here the results are from experi-
ments in which the angle of the secondary stimulus was quite
large. When this angle was large the magnitude of the illusion
was considerably diminished. As a consequence the observer
was more liable to be misled by other influences than the imme-
diate objects of perception. Moreover the difficulties of accur-
ately determining the numerical value of the illusion were in-
creased for the experimenter. Hence the values given in the
latter half of each table are less trustworthy than the corre-
sponding values in the first half.

INDIVIDUAL DIFFERENCES.

It is a notable fact that some individuals are more susceptible
than others to an illusion of the kind under discussion. Binet
has remarked1 that young children are more susceptible than

lL'Annee Psychologique, 1894, ' ^'illusion d'optique de Mueller-Lyer, ' p. 330.

LA W OF A TTRA C TION. 1 7 1

older persons. This fact is also very manifest in the results
which are here reported. For example, referring to Tables I.,
II. and III. a very casual examination is sufficient to reveal the
fact that the illusion values for the first four subjects are less
than the corresponding values for the remaining six subjects.
The first four subjects were adults, while the remaining six
were children varying in age from twelve to fourteen years.

This difference may be accouuted for partly on physiological
and partly on psychological grounds. In the case of children,
the nervous organism is not so firmly * fixed ' ; alterations among
its parts may be more easily effected. Attraction between the
elements of the organism, therefore, has a greater effect.

On the other hand, psychologically speaking, the judgment
of the other person is more evenly balanced, which is perhaps
equivalent to saying that experience furnishes to the older per-
son a larger supply of data upon which a judgment may be
based. A high degree of susceptibility to illusion, therefore,
may indicate on the one hand, especially in children, a nervous
organism which is plastic and impressible and therefore highly
educable, and on the other hand, especially in adults, a weak-
ness of judgment.

In the article ' Ueber den Einfluss von Nebenreizen ' to which
reference has already been made, I reported experiments by
myself upon several groups of children taken from different
classes of two elementary schools in Wiirzburg. The object
of these experiments was to determine quantitatively the effect
of a secondary stimulus upon the localization of a point stimu-
lated upon the skin of the fore arm. It was found that a fairly
consistent parallel existed between the amount of influence ex-
erted by the secondary stimulus and the degree of mental ability
attributed to the pupil by his teacher. This parallel was more
striking when groups of dull children were compared with
groups of bright children. The children used in these experi-
ments varied in age from six to fourteen years. It is very
questionable if the same relation would hold for older indi-
viduals.

The individual variations in the case of subjects of the same
age are marked also in the case of the visual illusion, but the

172 HAYWOOD J. PEARCE.

number of subjects for which results are reported is so small
that comparison would be valueless. Future experiment must
determine whether or not a relation, such as I have indicated,

exists.

GEOMETRIC-OPTICAL ILLUSIONS.

The literature of this subject is peculiarly rich and not un-
profitable. I shall attempt to touch briefly upon such salient
features only as are directly related to the phenomena which
have been under my observation.

Heymans has shown l that the Miiller-Lyer illusion (a) in-
creases with the length of the projecting arms, is then (<5) sta-
tionary, and finally (c) decreases as the length of the arms
increases.

These facts which seem to me fatal to most of the theories
which have been advanced to explain the illusion, are perfectly
in accord with the law of attraction as developed in the fore-
going pages. For, increase in length of projecting arm means,
(a) increase in intensity of the secondary stimulus and (#) in-
crease in the distance of the secondary from the primary stim-
ulus. In the former case, we have increase in influence and in
the latter case decrease in influence of the secondary stimulus.
If we begin to increase the length of the arms when they are
very short, each increment in length corresponds to a relatively
large increase in intensity, but as a result of the operation of
Weber's law, there comes a time when a very large increase in
length of arm (or secondary stimulus) results in only a rela-
tively small increase in intensity. On the other hand, each
increase in square of the distance has diminished the influence
of the secondary stimulus in proportion.

In the beginning therefore, (a) the increase in influence due
to increase in intensity is greater than the decrease in influence
due to increase in distance, later (#) the effect of the two factors
is equal, the one counterbalancing the other, and finally (c) the
decrease in influence due to increase in distance is greater than
the increase in influence due to increase in intensity.

Heymans further shows in the same connection (p. 227),
that there is a consistent proportional relation between the size

1 Zeitschrift fur Psychologic und Physiologie, Vol. IX., p. 236.

LA W OF A TTRA C TION. 1 73

of the illusion and the cosine of the angle formed by the pro-
jecting arm and the central line (Schenkelwinkel). When the
angle increases the illusion becomes less pronounced. This
fact harmonizes with our law of attraction, inasmuch as it is to
be expected that an attractive force will have greater effect
when acting in a straight line than when acting at an angle
upon a given object.

The application of the law of attraction to the other geo-
metric-optical illusions with any degree of accuracy is difficult,
if at all possible. In general, we may observe, however, that
displacements take place in the direction of greater « sensation
masses.' In the Poggendorf figure, for example, the points
where the diagonal joins the parallels are drawn, the one
upward and the other downward, by the relatively large sensa-
tion masses represented by the sides of the two angles formed.
The same may be said of the Zoellner figure.

In all such cases, where there is a displacement of a line
from its true objective position, it will be found that certain
points which mark the direction of said line are acted upon un-
equally by neighboring * sensation masses.'

The most difficult factor to determine in all of these phe-
nomena is the value of a particular * sensation mass,' or, as
previously termed, the intensity of a particular visual stimulus.
Spatial measurement is a very inadequate expression of this
intensity, as we had reason to observe in our discussion of the
results of Tables IV., V. and VI. There are evidently several
elements which go to determine the intensity of a given visual
stimulus. The first of these is undoubtedly spatial size. The
second is the proportionate part of all active sensory processes
which the sensation in question represents. When an experi-
ence is already crowded with sensory elements, the addition of
a new element has comparatively little sensory value. A third
element is position in the visual field — the same stimulus being
more intense upon the fovea centralis than on the periphery.
A fourth element is the amount of central reinforcement which
may be given the stimulus. Mach says,1 for example : * Der

1 Pfluger's Archiv, Vol. 60, p. 509. Also Zeitschrift fur Psychologic und
Physiologic, Vol. 16, p. 298.

1 74 HAYWOOD J. PEARCE.

blosse Wille rechts zu blicken gibt den Netzhautbildern an be-
stimmten Netzhautstellen einen gresseren Rechtswerth.' An
illustration of both the third and fourth elements just mentioned
is furnished by a phenomenon which I have observed in mak-
ing some experiments with the illusion to which Professor Loeb
first called attention. One of the illustrations which Professor
Loeb offered was as follows : If one places two pieces of money
on a table so that they seem equally far removed to one's right
and then places a third piece further towards the right so that
the three pieces form a right angle triangle it will be found that
the relative position of the first two has been so altered that the
lower one which is on a horizontal line with the third, now
appears further to the left than the upper one. In my own ex-
perience the phenomenon to which Professor Loeb calls attention
does not always appear and in fact the reverse phenomenon
sometimes appears, i. £., the lower one of the two vertically
arranged pieces appears further toward the right. On giving
the matter closer attention I found that the change in result was
brought about by a difference in the direction of attention. If,
for example, the attention is directed to the two lower pieces
the third is attracted by both and the phenomenon mentioned by
Professor Loeb may be observed. On the other hand if atten-
tion is directed toward the one above and the one to the extreme
right, or the two forming the hypoteneuse of the triangle, it will
be found that the reverse phenomenon takes place, viz., the
third is again attracted by the two to which attention is being
given, the angle opposite becomes obtuse and the upper of the
two pieces which were arranged vertically now appears to lie
too far to the left. Here attention, or central reinforcement,
and bringing of the two images nearer the fovea, both operate
to increase the intensity, or sensation mass, of the two sensa-
tions, diminishing in a corresponding degree the intensity of the
third sensation. The consequence is that the two stronger
attract the third with a greater force than it attracts them and
it is therefore displaced from its true relative position.

Finally, a fifth element in determining the intensity of a
given sensation is the duration of the stimulus — the intensity
of a stimulus diminishing as it grows older. This last element

LA W OF A TTRA C TION. 1 75

has perhaps not figured in the case of the illusions which have
previously been discussed. But in the case of the so called
illusion of reversible perspective it probably plays a large part.
Such illusions are usually brought about by staring at a figure.
The result of the staring is to diminish gradually the intensity
of the sensations occasioned by the points which determine the
form of the figure. The intensity of the neighboring points
is relatively less affected. Consequently, these neighboring
points finally have a larger * sensation mass ' or intensity than
the others, and when this happens, they determine the form of
the figure according to their own disposition. After this has
happened a few times, one can so reinforce the intensity of
these latter points from within that he is able to control the
phenomenon at will.

THEORIES.

Heymans and Wundt1 both agree that the Miiller-Lyer
illusion is due to an almost physical impulse (fast physische
Zwang) to follow the direction of the projecting arms with eye
movements. One might argue in support of this theory that the
law of attraction governs in fixing the strength of the impulse
to eye movement. The question still remains unsettled, how-
ever, as to whether the judgment is a by-product of this impulse
to movement, or whether the impulse to movement is itself a
product coordinate with the judgment.

Wundt, in criticism of Heymans' contrast theory, calls
attention 2 to the fact that the illusion takes place when either
figure is compared with a straight line and no contrast of eye
movement is possible. This criticism is justified by my experi-
ments which were concerned with only one type of figure.

It is manifest that such theories 3 as the confluence theory of
Mueller- Lyer, the Auerbach indirect vision theory, the Brentano
pseudoscopic angle theory, the Thiery perspective theory, the
Einthoven dispersion image theory and all others which are
based upon phenomena growing out of the extension of arms
at an angle, are shown to be inadequate by the fact that the
illusion is present when no such angles appear in the figure.

^Physiological Psychology, Vol. II., p. 149.

* Die geometrisch-optischen Tauschungen, p. 47.

'Titchener, Experimental Psychology, pp. 321-328.

176 HAYWOOD J. PEARCE.

GENERAL CONCLUSIONS.

The law of attraction as represented in the present paper is
an attempt to state in definite form a principle which has been
more or less prominent in the theories of several writers upon
this subject.

Jastrow,1 for example, has attempted to explain optical il-
lusion in general on the principle that all objects are judged
relatively to their environment. Our judgment of a thing is
modified by the other things which surround it.

A great variety of facts which illustrate the principle may
be drawn from every day experience as well as from experi-
mental laboratories. For example, Lipps 2 calls attention to the
fact that cows appear to be larger when they are in narrow,
low stalls than they do when outside.

Professor Baldwin in!an article upon the ' Effect of Size-con-
trast upon Judgments of Position in the retinal Field,'3 reports
that a point, in the field of vision, midway between two figures
of unequal size, as two squares or two circles, will be attracted
towards the larger figure. Further, the tendency to error in-
creases with the relative increase of the side of the larger
figure and the tendency is about twice as great when the figures
are arranged vertically as when they are arranged horizontally.

In an article entitled ' Normal Motor Suggestibility,'4 1 have
reported a series of experiments showing that the localization of
a point stimulated upon the skin of the forearm is influenced by
the stimulation of a second point, either above or below the one
to be localized. It was also shown that the localization of a
visual image in the peripheral field was similarly affected by
the appearance in the same field of a second visual image, and,
similarly, the localization of a sound was affected by a second
sound.

The tendency to fuse together of two or more sensations
which are simultaneously experienced has been frequently re-
marked and experimental psychology has shown conclusively

1 American Journal of Psychology, Vol. IV., p. 381.

2 Raumaeslhetik, p. 65.

'PSYCHOLOGICAL REVIEW, Vol. II., p. 244. Cf. also the further figures
given in his Fragments in Philosophy and Science, pp. 275 ff.
* PSYCHOLOGICAL REVIEW, Vol. IX., pp. 329-356.

LA W OF A TTRA C TION. 1 77

that one of the chief defects of the older introspective psy-
chology was its failure and inability to recognize in experience
the elementary sensations which composed it.

All these and many other similar facts seem to point to a
general law of relativity, which may be stated somewhat as
follows : Every sensation is influenced by every other sensation
which may be -present in any complex experience. The nature
of this influence seems to be a direct interaction of one upon the
other, the resulting effect of this interaction being determined
by the nature of the interacting sensations.

All of the sensations with which I have dealt experimentally
are such as to make up * extensive ideas ' ; * they were either
sensations defining position or form and magnitude, and the
nature of the interaction seems to have been an attractive force,
which I was able to measure. This attractive force is governed
in its action by the same general law which governs the action
of the attractive forces in nature, with which we are already
familiar and which has been given mathematical expression in
the well-known formula f= C (m x m') -r- D2. In its applica-
tion to the phenomena which have been under our observation,
y, in the formula, equals the force exerted by two sensations,
the one upon the other, m is the intensity or sensation mass
represented by a primary stimulus, m' is the intensity or sensa-
tion mass represented by a secondary stimulus, and D is the
distance between the primary and secondary stimuli, measured
from center to center. The constant C must be determined
empirically, and is not the same value in the case of visual and
tactual sensations.

The apparently physical nature of the law leads to the sug-
gestion that this attractive force operates between the nervous
elements, electro-chemical in nature, which mediate sensation.

On the other hand, one might be justified in admiring that
universality of the law, manifest in its consistent operation in
two such widely separated spheres as the material and the
spiritual.

The time is not opportune, I think for a discussion of the
vexed question as to the nature of mind which is involved in
the two possible theories here suggested.

'Titchener, An Outline of Psychology, p. 154.

178 HAY WOOD J. PEARCE.

SUMMARY.

1. When a subject is required to judge the length of a single
line, by comparing it with a second line which is variable in
length, the single line is always underestimated.

2. Using the same method of comparison, if the subject is
required to judge the length of the same line, now accompanied
by shorter lines which represent extensions of the line of direc-
tion of the original line but which are separated from it by
open spaces, the original line is judged to be longer than it was
when unaccompanied by the shorter lines, and, generally, it is
judged to be longer than objective measurement shows it to be.

3. When the results for several subjects are consolidated it
is found that the influence attributable to the addition of the
shorter lines, or secondary stimuli, is (a) increased when the
length of the secondary stimuli is increased, also (b) that this
influence is increased when the length of the line to be judged
is increased, and finally (c) an increase in the distance of the
short lines from the central lines, or primary stimulus, measured
from center to center, is followed by a decrease in the influence
of the short lines, or secondary stimuli.

These general relations obtain in the case of individuals as
well as for groups of individuals, but the individual variation is
somewhat large, and comparisons of individual results are not
thoroughly satisfactory.

4. A fourth and a disturbing element in determining the
amount of influence of the secondary stimuli was the space
between the ends of the primary and the secondary stimuli.
When the distance between the ends of the primary and the
secondary stimuli was decreased, the amount of influence of the
secondary stimuli was correspondingly increased. The exact
proportionate relation was not determined.

5. When the disturbing factor just noticed could be dis-
regarded by reason of a favorable arrangement of conditions, it
was found that the well known formula, expressing the law of
attraction in the material universe, can be applied to the results
of the experiments here described.

6. The results of certain experiments in judging the length
of lines stimulated by pressure upon the skin of the forearm,
also yield to a similar statement.

THE RELATION BETWEEN THE VASO-MOTOR
WAVES AND REACTION TIMES.

BY WM. R. WRIGHT,
University of Michigan.

The experiments herein recorded are a series of reaction
experiments conducted with the view of ascertaining any pos-
sible relation existing between the vaso-motor wave and the
reaction time of the subject ; or, stated in the form of a question,
does the reaction time of the subject vary in length in accord-
ance with the rise and the fall of his vaso-motor, or ' Traube-
Hering,' wave?

The subject was placed in a room separate from the record-
ing apparatus so that all distractions of sight and of sound were
reduced to the minimum. He was so seated facing a small
table that both of his arms rested easily upon the top of the
table. Within the palm of the subject's left hand was fastened
a Hallion and Comte plethysmograph, while with the right hand
he operated a telegraphic key. To the subject was attached
also a Sumner pneumograph, the records of which were taken
with the view that they might be of special value in the study
of another problem in the future. Although no use was made
of these records in this series of experiments, it was found that
the markings of the pneumograph could be recorded along
with the other records without interfering in any way with the
subject's attention to the particular task assigned him.

On the table in front of the subject was a telegraphic
sounder. This was screened from his view, and furnished the
auditory stimulus to which he reacted by pressing the tele-
graphic key mentioned above.

All the recording apparatus was placed in the experimenter's
room, and was connected by air-tight rubber tubing and insu-
lated wires with the apparatus in the subject's room. The
records were taken on two kymographs, one of the horizontal

179

I So WM. R. WRIGHT.

type with a traveling carrier, and the other a Zimmermann, of
the vertical pattern. The motive power for the revolving drums
was furnished by an electric motor, the horizontal drum, 50 cm.
in circumference, being so regulated in regard to speed that it
made one revolution in 7.8 seconds. By means of graded
pulleys the rate of speed of the vertical drum was so adjusted
that the surface of the drum moved at a rate equaling the rate
of the longitudinal movement of the markers connected with
the horizontal drum ; but it was found that complete reliance
could not be placed upon this adjustment alone, as the least slip
of one of the belts made a perceptible change in the rate of the
speed of the drums.

The vertical drum received the markings of a Lombard-
Pillsbury piston-recorder which was connected by a rubber tube
with the plethysmograph, the markings of a Marey tambour
connected by a rubber tube with the pneumo graph, and also the
records of an electric marker so connected with the reaction
time-marker of the horizontal kymograph that the beginning of
each reaction was written on the vertical drum. Care was
taken to keep the three writing points in the same vertical line
upon the drum.

On the carrier of the horizontal kymograph were fastened
two electric markers, one, connected electrically with a vibrat-
ing tuning fork, marked fiftieths of a second, and the other by
its deflections marked the reaction period of the subject. At
first a tuning fork of 200 double vibrations was used, but this
necessitated such rapid revolutions of the drum that the re-
sponses came in close rhythmical succession, and the subject
responded when he expected them and not to the signal.

On the pulley of the horizontal drum was fastened a metal
attachment that automatically closed and kept closed an electric
circuit during one half of a revolution of the drum. The elec-
tric marker of the vertical drum, the reaction marker pf the
horizontal drum, the telegraphic sounder and the telegraphic
key were so wired together and connected with the automatic
circuit closer that the closing of the circuit gave the subject his
signal and recorded on both drums the beginning of each re-
action ; and the pressing of the telegraphic key by the subject

PSYCHOLOGICAL REVIEW, VOL. XI., 1904.

PLATE II.

VASO-MOTOR WAVES AND REACTION TIMES. l8l

released the reaction marker on the horizontal drum and marked
the close of the reaction. For the early experiments the elec-
tric current was furnished by storage batteries ; but, as these so
often proved unsatisfactory, use was finally made of a small
dyna-motor which gave a steady reliable current for the tuning
fork and the markers.

The only instructions given the subject were that he should
keep one position without moving his left hand, and that he
should press the telegraphic key with his right hand each time
he heard the signal.

Midway between two signals, the release of the telegraphic
sounder could be faintly heard, and this became an equivalent
for the experimenter's usual ' now.' The length of the time of
the revolution of the horizontal drum giving the signals for the
reactions was such that the subject was fully able to recover
himself before the warning ' now,' and thus there was little
fluctuation in the degree of his attention throughout a sitting.
Thirteen and occasionally fourteen revolutions of the horizontal
drum formed one series of experiments ; and, after considerable
experience, the experimenter was able to secure three series
within an hour.

As the sheets of the kymographs were filled, each was fixed
by the usual bath ; and, for convenience in reading the records,
the sheet containing the vaso-motor waves was pasted across the
reaction sheet in such a manner that, for an ocular demonstra-
tion, the joining of the points marking the close of the reactions
formed a series of curves under the vaso-motor curves (see
Plate II.). The readings were taken by measuring in fiftieths of
a second the length of each reaction. By means of lines drawn
perpendicular to the line connecting the points marking the be-
ginning of the reactions, the exact positions of the reactions in
time with reference to the vaso-motor waves were found. The
lengths of the reactions, or the reactions in seconds, were then
grouped into four groups, as to whether the reaction occurred
at the base of the vaso-motor wave, on the rise of the wave, at
the crest of the wave, or during the fall of the wave. The re-
action times of each group were then averaged according to the
number of experiments in each group. A correction of .016

1 82 WM. R. WRIGHT.

second was made to cover the latent period of the reaction
marker.

Five persons served as subjects, Dr. Pillsbury (P.), Mr.
Hayden (H.), Mr. Freund (F.), Miss Lee (L.) and Mr. Wright
(W.). With the last named, Mr. Hayden conducted the ex-
periments.

Experiments with P. were conducted between nine and ten
o'clock, a. m., and the table below shows the results obtained.

Number of Average Reaction

Experiments. Time in Seconds.

Base of vaso-motor wave, 113 .170

Rise of vaso-motor wave, 65 .186

Crest of vaso-motor wave, 90 .194

Fall of vaso-motor wave, 82 .187

H.'s reactions were taken at ten o'clock, a. m., two o'clock,
p. m., and four o'clock, p. m. on different days. His records,
given below, are considered first as forming one series irrespec-
tive of time ; then each hour is represented as making an inde-
pended series.

Number of Average Reaction

Experiments. Time in, Seconds.

Base of vaso-motor wave, 53 .249

Rise of vaso-motor wave, 63 .261

Crest of vaso-motor wave, 42 .270

Fall of vaso-motor wave, 56 .262

EXPERIMENTS CONDUCTED AT TEN O'CLOCK, A. M.

Number of
Experiments.

Average Reaction
Time in Seconds.

Base of vaso-motor wave, 14

.241

Rise of vaso-motor wave, 24

.262

Crest of vaso-motor wave, 1 1

.283

Fall of vaso-motor wave, 15

•254

EXPERIMENTS CONDUCTED AT Two

O'CLOCK, P. M.

Number of
Experiments.

Average Reaction
Time in Seconds.

Base of vaso-motor wave, 7

.249

Rise of vaso-motor wave, 10

.252

Crest of vaso-motor wave, 6

.248

Fall of vaso-motor wave, 17

.261

EXPERIMENTS CONDUCTED AT FOUR

O'CLOCK, P. M

Number of
Experiments.

Average Reaction
Time in Seconds.

Base of vaso-motor wave, 32

.249

Rise of vaso-motor wave, 29

.264

Crest of vaso-motor wave, 25

.269

Fall of vaso-motor wave, 24

.266

VASO-MOTOR WAVES AND REACTION TIMES. 183

F.'s records, taken at 8 o'clock, a. m., are as follows:

Base of vaso-motor wave,
Riie of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of

Experiments.

9

7
II
8

Average Reaction
Time in :

•334
•342

•389
.299

L.'s record, taken at 2 o'clock, p. m., are as follows :

Base of vaso-motor wave,
Rise of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of
Experiments.

13
10

19
I?

Average Reaction
Time In Seconds.

•273
.282

•293

.284

W.'s hours corresponded with H.'s and are similarly reported.
ENTIRE SERIES OF EXPERIMENTS.

Base of vaso-motor wave,
Rise of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of
Experiments.

112

58

IIO

71

Average Reaction
Time In Seconds.

.187
.189
.201
.191

TEN O'CLOCK SERIES OF EXPERIMENTS.

Base of vaso-motor wave,
Rise of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of
Experiments.

40
27
45
23

Average Reaction
Time in Seconds.

.189
.191
.208
.201

Two O'CLOCK SERIES.

Base of vaso-motor wave,
Rise of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of
Experiments.

Average Reaction
Time in Seconds.

47

.190

15

.199

30

.201

24

.186

FOUR O'CLOCK SERIES.

Base of vaso-motor wave,
Rise of vaso-motor wave,
Crest of vaso-motor wave,
Fall of vaso-motor wave,

Number of
Experiments.

25

16

35
24

Average Reaction
Time In Seconds.

.182

.175
.190
.186

184 WM. R. WRIGHT.

It is to be noted that the reactions followed one another in
order at a uniform rate throughout a series of experiments re-
gardless of the position of the vaso-motor wave, and that the
subject at all times was ignorant of the relation existing between
his reaction time and its relative position with reference to his
vaso-motor wave, hence the number of experiments occurring
in the different groups, /. £., base, rise, crest and fall of wave,
differed widely.

A single sheet of records showed little constancy in its re-
sults, and a slight variation may be noted in the breaking up
of the whole number of both H.'s and W.'s experiments into
the hour series ; F.'s and L.'s experiments are too few to be of
any special significance ; yet even in these cases there is a
decided tendency toward the results plainly seen in connection
with the whole number of experiments of P., H. and W. The
records of the three last named persons show clearly a differ-
ence in the times of the reactions that occur at the time of con-
stricted vaso-motors, at the time of dilated vaso-motors or at
points midway.

To review P.'s reactions, we find his reactions the shortest
(.170 sec.) when the blood supply in his hand is at its lowest,
and that his reactions are the longest (.194 sec.) when there is
a full supply of blood. Between these two points, the reactions
are slightly quicker if they occur at the time of the dilating
(.186 sec.) than if they occur at the time of the contracting
(.187 sec.) of the vaso-motors.

The results of H.'s and W.'s experiments, taken in their en-
tirety as one series for each, agree with the facts already noted
in connection with P.'s reaction times; e. g., H.'s are .249,
.261, .270 and .262 seconds, and W.'s are .187, .189, .201
and .191 seconds. These same records when broken up into
hour series still show, with one or two slight variations, similar
relations. The only real discrepancy in H.'s results is to be
seen in the time of his reactions occurring at the crest of the
wave, but with this it will be noted that the number of experi-
ments (6) is too small to offset general results. With W.'s
hour series but two variations occur, and neither one of these
materially changes the general trend of the curve of reaction
times established by the greater number of experiments.

VASO-MOTOR WAVES AND REACTION TIMES. 185

In reply to our query stated in the beginning of our report,
we would repeat that the results of the above experiments show
that the subject's reactions form a curve, which, in shape, agrees
close with the curve of his vaso-motor, or * Traube-Hering,'
wave.

L. M. Patrizi in 1896 (see L'Annte Psychologique^ Vol. 3,
1897, p. 359)1 conducted a series of experiments similar to the
above with the exception that he had the opportunity of taking
the plethysmographic record from the brain of his subject and
thus write the curve of cerebral volume. His reactions are
recorded in only two groups, minimum and maximum volume,
or base and crest of vaso-motor wave. In all he secured 244
reactions, 128 at the crest and 116 at the base of the wave, and
his general average of reaction times shows .3325 seconds for
the crest and .345 seconds for the base of the wave — just the
opposite of those found in our experiments when the record was
taken from the finger.

Notwithstanding the difference found by M. Patrizi in his
experiments, he concludes that it is too slight to establish any
relation whatever between blood supply and reaction time ;
whereas our results, conducted upon more subjects and in con-
nection with the blood supply of the hand, do show differences
great enough to warrant the assumption that the rate of reaction
is related to the * Traube-Hering ' wave.

Were Patrizi more sanguine as to the correctness of his re-
sults, it would be interesting to discuss the question of the rela-
tion of direction between ' Traube-Hering' of brain and finger;
but, as it is, no conclusions on this point can be drawn.

1 Original article, Archivi di Psichiatria, 1896. We have seen only the
summary.

ON THE HOROPTER.1

BY 'GEORGE T. STEVENS, M.D., PH.D.,

New York.

Of all the subjects in physiological optics none has been
thrown into greater confusion by conflicting views of different
investigators and none has been surrounded by greater mystery
than that of the horopter. Helmholtz, after devoting about
ninety pages of his monumental work on physiological optics to
the horopter, pages replete with experiments and with abstruse
mathematical formulae, evolved a theory which no other investi-
gator could verify even of the few who claimed to be able to
understand it. With all this erudite labor and with all the
enthusiastic interest of the great philosopher he worked out a
single horopter of the infinite number which may exist and even
that one, being based on false premises, was absolutely faulty
for well adjusted eyes and entirely impracticable for any eyes.

It is, therefore, when all the divergent opinions are con-
sidered, not altogether without an appearance of justice that so
astute a man as Giraud-Teulon should have characterized the
horopter as a * transcendental fancy.'

" When," he says, " all the labor of determining the surface
curve (fulfilling the geodesical condition of the horopter) was
ended it was discovered that this surface assumed the form of a
torus. * * * It was not noticed that a table with four legs, a
chair placed before us, were seen singly although they certainly
had none of the attributes of a torus."3

Nevertheless the subject of the horopter or to put it better,
of horopters, is one of great practical importance. We may
emphasize the expression and say that it is one of preeminent
importance.

1 Read before the New York Branch of the American Psychological Asso-
ciation and the Section of Anthropology and Psychology of the New York
Academy of Sciences, at Yale University, New Haven, October 20, 1903.

8 The Function of Vision, translated by Owen.

186

ON THE HOROPTBR. 187

A horopter may be defined as consisting collectively of all
the points in space whose images, with a given adjustment of
the eyes, fall upon corresponding points of the two retinas.

Notwithstanding the view I have expressed of the notable
rank which should be accorded to this subject, the general de-
finition as just given is almost the only point concerning the
phenomena of horopters on which investigators, those who have
conceded a horopter, have agreed.

By some it has been described as a line, by others as a sur-
face and by Helmholtz especially as a most complex and quite
incomprehensible combination of curves, planes and straight
lines.

Without entering upon the merits of Helmholtz's proposi-
tions that the horopter is * a line of double curvature produced
by the intersection of surfaces of the second degree (hyperboloid
to a nappe, cone or cylinder) ' that * it is a straight line and a
curved plane of the second degree,' etc., we may for a moment,
without accepting the doctrine, consider the position of the
horopter according to this philosopher when the plane of regard
is directed to the horizon.

" In a single case only," says Helmholtz, " is the horopter a
surface, it is when the point of regard is situated in the hori-
zontal and median planes and at an infinite distance. The
plane of the horopter is then parallel to the plane of regard.
* * * In the case of normal eyes thus directed toward the
horizon the horopter coincides approximately with the ground
on which the observer walks."

If we consider this proposition with care it will appear that
if it were correct its accuracy would involve much ocular incon-
venience. We do not look at the horizon when we walk. One
who would hold the head erect and direct the eyes to the horizon
would stumble often in his march. But, according to the propo-
sition, if the eyes should be directed to the ground at a few feet
in advance of the pedestrian he would bury his horopter beneath
the soil and all the objects in his pathway would appear, so far
as a horopter is concerned, confused and indistinct.

I have taken so much of your time with an introduction in
order that we may at the outset form an idea of the present state

1 88 GEORGE T. STEVENS.

of the doctrine. Recurring to our definition, if a horopter is
the collection of the points in space whose images, with a given
adjustment of the eyes, fall upon corresponding points of the
two retinas, it follows that horopters succeed each other in end-
less variety and with amazing rapidity. With every glance of
the eyes, with the passing of the line of regard from one part
of the page of a book to another, in fact, with every change of
the head, of the body or of the eyes themselves and with every
degree of convergence a new horopter is developed. A horopter
will be formed when the two eyes are so adjusted as to enable
the image of the point fixed to be located exactly at the maculas
of the two retinas.

The innate impulsion to form a practically complete horopter
with any given fixation is so imperious that only insurmountable
obstacles will serve as a restraint.

Two tenets or conceptions constitute the essential foundation
for the doctrine of the horopter. They are, the theory of the
position and direction of the meridians of the retinas and the
theory of corresponding points.

In respect to both tenets Helmholtz and most modern
searchers in this field have adopted views which have resulted
in the confusion in which the subject has been involved.

Before we can proceed to the phenomena of the horopter
then it is essential to obtain a correct idea of these two funda-
mental theories.

We speak of vertical and of horizontal meridians of the
retina. They are, like the meridians of the globe, imaginary
lines yet they have distinct relation to sight impressions. For
example, let us suppose a horizontal meridian passing through
the macula, the eye being directed straight forward and the
head being in the primary position. The eye fixes a given
point the image of which is impressed at the macula. Now if
another point at one side of this point of fixation is situated on
a higher plane than the point of fixation, its image will be im-
pressed at one side of the macula and below the horizontal
meridian. It is unnecessary to consider in detail this doctrine
but we may at once assume an understanding of the general
principle. Helmholtz, Volkmann, Hering and other investi-

ON THE HOROPTER. 189

gators came to the more or less uniform conclusion that the
horizontal meridians were all parallel with the external horizon
but that the vertical meridians were only apparently vertical,
and that they leaned out above and approached each other
below. Helmholtz's experiments led him to the belief that the
vertical meridians of each eye leaned out about i^°. A
number of investigators immediately found that their vertical
meridians in each and every instance leaned out exactly i^°.
It remains for a society of psychologists to determine how it
happened.

My own researches led me to devise the clinoscope which
has now become one of the most important and essential of in-
struments in practical examinations of the eyes. One of the
first things which the clinoscope did was to demonstrate that
these leanings are natural defects — personal peculiarities —
and that they vary with different individuals from one to a dozen
degrees ; that it is rare to find two persons in succession who
record the same leaning. These leanings I have called decli-
nations.

Abundant experience in the correction of these defects of
declination have demonstrated beyond all reasonable doubt that
the proper position for a vertical meridian is the vertical position.

That Helmholtz had what I have called a plus declination
for each eye I am convinced. There is much reason however
to believe that it was considerably in excess of iX°-

Thus, Helmholtz included in his most elaborate mathe-
matical calculations his individual defects which he assumed
were physiological features common to mankind.

This was one of the fundamental errors.

The second foundation tenet is the doctrine of corresponding
points of the retinas. We may quote Helmholtz's proposition
as the accepted view of the doctrine. " The points which , in
the retinal horizons are at equal distances from the point of
fixation are corresponding -points" He states the proposition
similarly for the vertical meridians. In fixation with the two
eyes the image of the precise point of fixation is impressed upon
each retina exactly at the macula or fovea, and, according to
the above proposition another point outside the point of fixation

190 GEORGE T. STEVENS.

will be impressed upon corresponding retinal meridians which
in the case of each eye will be equally removed from the macula.
In respect to corresponding points in the field of vision we may
quote again. " Corresponding points in the two visual fields
are those which are at equal distances and equal in direction
from the corresponding horizontal and apparently vertical
meridians"

While this proposition is not altogether clear it is evident
from the context that according to it a series of points equally
distant in the field of view and from which proceed lines of
direction toward equally distant points in the retinas are corres-
ponding points.

These propositions can not both be true except under cir-
cumstances entirely at variance with Helmholtz's illustrative
experiments. These experiments are made, not with curved
surfaces, hollow spheres, but with plane surfaces like the usual
stereoscopic cards or the flat page of a book.

Accepting the experimental illustrations as the only practical
tests, the two propositions are inconsistent.

Let us first suppose the distances between corresponding
points on the horizontal meridian of the retina are exactly equal.
Place a sheet of paper exactly in front of the eyes on which
are several points in a straight horizontal line corresponding to
the plane of regard (Fig. i). Let the eyes be fixed on the
central point. Then, according to the first proposition these
points are not corresponding, for straight lines drawn from
them through the two nodal points to the retinas will not form
equal angles and will not meet equally removed points of the
two retinas.

Suppose our points to be one half an inch distant from each
other and fifteen inches in front of the eyes while the two eyes
fix the central point, A. Then will the line of incidence passing
from the point at the right of the central position, B, form with
the line of incidence passing from the point of fixation through
the nodal point to the macula of the left eye an angle of i° 54'
5" while the angle formed by the line of incidence from this
same secondary point will constitute with the line passing
from the point of fixation to the right macula i° 53' 26"

ON THE HOROPTBR. 191

Passing to the next succeeding point of the series, the in-
cidental line from the new point will form with the line of inci-

B C

FIG. i. Let R and L be the nodal points of the two eyes and A the point of
fixation. The points B, C, etc., are outside the point of fixation. Suppose

.#0=1.25 in., OA = is in.
= ? 45' 49"

f <AK£=i° 53' 26" f

10S4'S" \<ALC=4°39'S*"'

The points corresponding to the incidence of the lines CR and CL are not thus
equally removed from the maculae.

dence from the fixation point to the left macula an angle of 3°
46' i". The incident line to the right eye will form with the

I92 GEORGE T. STEVENS.

original line an angle of 4° 39' 58". It will thus appear that
in this example, passing from the macula toward the periphery
of the temporal side of the retina the angle increases while on
the side of the retina medial to the macula it decreases.

Since these incidental lines cross at the nodal points it is
evident that they must extend to unequal distances in the vault
of the retina.

Miiller, recognizing this, believed the horopter to consist of
a circle passing through the nodal points of the eyes and the
point of fixation and of a vertical line. This in fact amounts
to no horopter. Time does not permit an examination of all
these theories and it is sufficient to say that no theory based on
equal distances for corresponding points of the retinas can serve
as a satisfactory basis for a doctrine of the horopter.

These two tenets on which the whole structure has been
erected being rejected the doctrine is to be abandoned or new
basic theories must be found.

As already remarked, the clinoscope and practical work
based upon its revelations have demonstrated that beyond all
question the vertical and horizontal meridians of the retinas are,
in typical cases, precisely what their names imply, exactly hori-
zontal and exactly vertical. We may then substitute this fact for
Helmholtz's theory and it will serve as our first basic principle.

In respect to corresponding points it is unnecessary to say
that there is no such anatomical symmetry as to demand that
equal extents on the retinal surfaces should represent equal ex-
tents in the field of vision. It is not the fact. It is however,
true that there is an innate sense of the goniometrical value of
motor impulses directed to the muscles of the eyes, and that the
distances between retinal corresponding points need not be sym-
metrical for the conception of this muscular sense but that never-
theless they bear certain mathematical relations to each other.

We may define corresponding points of the retinas then as,
those joints in the retinas which answer to proportional degrees
of rotations of the eyes about their centers of 'rotation, and which,
from given joints in the plane of the point of fixation receive
incident rays which must pass through the nodal -points.

They represent therefore the relation between the muscular
and the retinal senses.

ON THE HOROPTBR. 193

The definition is perhaps less easy to the average compre-
hension than it is to divide the retinas into squares of millimeters
and point off so many to the temporal side of one and so many to
the medial side of the other retina and call these corresponding
points. There are various combined physical and psychical
functions which are not to be measured by a pocket rule.

The actual movements of the eyes about the rotation centers
are not always essential to an estimation of the relative positions
of objects in space. In the absence of the objective movement
there is the subjective conception of the impulse required to in-
duce a given movement. We are all familiar with the experi-
ment of Dove in which the observer, looking into a dark box
until the eyes are supposed to have assumed parallel directions
sees an electric spark within the box and it is seen singly. Of
course the impression is made at the temporal side of each
macula and there should be by rigid rule, when impressions are
at these non-corresponding points, an impression of two sparks
seen heteronymously. The rule in this case is not tenable.

If there were actually two sparks there would be two im-
pressions on each retina whereas as a matter of fact there is but
one. The consciousness of a single image for each retina and
of its position external to the macula leads to the conclusion that
a convergence of the eyes would be required to locate the image
at the macula and the extent of the required motion would indi-
cate the angle of convergence and therefore the distance of the
spark. Of course there are other elements in this complex psy-
chical phenomenon but that mentioned is enough to suggest the
course of the psychical process. It is such processes of uncon-
scious conclusions that bring many points within the field of
vision into a subjective horopter.

Before proceeding directly to examine the principles of the
horopter it is necessary to recall some of the changes in the
directions of the meridians as the eyes pass from one point of
fixation to another. When the point of fixation is at infinite
distance and in the median plane all horizontal meridians are
horizontal and all vertical meridians are vertical. So also if in
the plane of the horizon the point of fixation is brought nearer,
the meridians maintain their original relations and these rela-

194 GEORGE T. STEVENS.

tions will also continue if the two eyes are directed upward or
downward provided the visual lines remain parallel. But if the
point of fixation is at such distance as to demand convergence
of the lines of regard and if it is above or below the horizon
(the head being supposed to be in the primary position) all hori-
zontal and all vertical lines assume new directions. The eyes
rotate on their antero-posterior axes. This form of rotation is
known as torsion. These torsional rotations are governed by
fixed laws and the general principle is known as the law of
Listing.

Should the visual lines of the two eyes converge at the same
time that the plane of regard is depressed the horizontal meri-
dians of each eye will tilt downward toward the temporal side
and upward toward the medial side. The vertical meridians
will also tilt with the upper part outward and the lower part in-
ward. The tilting is in every case in proportion to the depres-
sion and the lateral direction of the line of vision.

Accepting the two basic principles as they have been stated
and with an understanding of the laws of -torsion we are in posi-
tion to examine the phenomena of the horopter, eliminating the
mathematical intricacies of Helmholtz and substituting only
simple calculations in plane trigonometry. Time will not per-
mit us to inquire in detail into its form in many positions,
three will suffice to illustrate the principles and the details of
only one of these need be given.

First the observer directs the gaze toward the horizon in the
median plane at infinite distance, the head being in the primary
position. A horopter is formed at the distance of the point of
fixation and it will be a plane surface at right angles to the
plane of regard. Objects within or beyond the distance of the
point of fixation will not be in the objective horopter but may
be in what we may term a subjective horopter. They may be
impressed on the two retinas and they will appear, as in the
case of the spark in the Dove experiment, as single, the prin-
ciples controlling the psychical phenomena in that experiment
being here modified to meet these different conditions.

Second, if the gaze is directed somewhat downward and to
a point a few feet in advance, as in the case of a person walking,

ON THE HOROPTBR. 195

the horopter will still be very nearly at right angles to the plane
of regard, tipping forward slightly since, although there is de-
pression (a negative ascensional angle) of the plane of regard,
the convergence (the lateral angle) is so slight as to induce small
torsional action and the principle of objective and subjective
horopter may be applied as in the first case.

There is also, at the lower part of the field of view a bend-
ing in of the horopter so that more of the pathway is in the
horopter than would be were it through its whole extent a plane.

Coming to the third case we may proceed in more mathe-
matical detail.

Let us suppose the case in which the eyes are directed to the
page of a book in the ordinary position for reading.

Assume that the gaze is directed so that the point of fixation
is in the median plane, and that the plane of regard is depressed
35°. Assume also that the distance between the nodal points
of the two eyes is 2^ inches and that the convergence of the
eyes (the lateral angle) is for each eye 5°. We have from these
data to determine the distance of the horopter and its form and
position relative to the place of regard.

To determine the distance of the point of fixation (which will
be in the center of the horopteric field) we have the base, 2#
inches and the lateral angles 5°. Taking one half the base

14.287

FIG. 2. Angle A = 5°, | = Cot A, b = 14. 287.

and one lateral angle we have a base of i^ inches, a right
angle and an angle of 5° to find the perpendicular or distance
from the base line to the page of the book which is readily
found to be 14.28 inches (Fig. 2).

The distance being ascertained by the formula = cot A,

I96 GEORGE T. STEVENS.

a being the base, 1.25 inches, A the angle opposite the base
and b the distance sought. At this distance from the base line
the image of the point of fixation will be exactly at the macula
of each eye.

According to the law of torsions by this depression of the
gaze and the convergence the meridians will have left the hori-
zontal and vertical positions. Referring to the table of torsions
found in Helmholtz's work1 we find that for the ascensional
angle of 35° and lateral angle of 5° the tilting of the horizontal
(and of the vertical) meridians is i° 35'. These conditions be-
ing given what will be the relation of a straight line passing
horizontally through the point of fixation across the page to the
horizontal meridians of the retinas now tilted i° 35' from the
actual horizon. A series of points in a straight line thus pass-
ing through the point of fixation must impress themselves along
the horizontal meridian of each eye otherwise the points will
appear confused or double. But how can this series of points
in a horizontal line be impressed upon the meridians which are
tilted up toward the nasal side each i° 35'.

It is a most interesting fact that the images of these points
will in fact be thus impressed exactly along these tilted meri-
dians of the retina and it is precisely because these meridians
of the retina are thus tilted that it is possible for the impressions
to be made along the proper meridians.

Too much space would be occupied were we to enter upon
a mathematical demonstration of this statement but a little con-
sideration by one familiar with the relation of lines and angles will
show that in principle the statement is correct. A demonstration
however would show that beyond a certain degree (10° to 20°)
in the plane of regard a straight line actually appears to curve.

We come next to the more complicated question in respect
to the position of a line running from the top to the bottom of
the page. Will this line be at right angles to the plane of regard
as the horizontal one is parallel with it or will it lean more or
less toward or from the plane of regard?

We may select points above and below the point of fixation
and determine their distance from the base line and thus obtain
the angle of the surface of the book to the plane of regard.

1 Optique Physiologique, p. 607.

ON THE HOROPTBR. 197

Take, first, a pbint 5° above and one 5° below the point of
fixation. The distance of the point of fixation from the base
line connecting the nodal points has already been determined
at 14.28 inches. In that case there was a lateral angle of 5°
for each eye. Now, since the vertical meridian of the retina
tilts out as it rises above the macula this lateral angle will
increase as the image is impressed above the macula and it will
decrease in proportion to the extent that the impression is made
below the macula. Before we can proceed, therefore, it is
necessary to find the exact amount of increase and decrease for
the selected points 5° above and 5° below the point of fixation,
since our angle of convergence will increase in proportion to
the extent to which the vertical meridian leans out from its
original position exactly at the selected dis-
tance and decrease in proportion as the me-
ridian leans in below the macula at the selected
distance.

We may find the extent of removal by the
formula :

a = 5 x. 02764 = 0°, 1382 = 8'i7>£". (Fig. 3.)
7 = tan A ; tan A = .02764,

In which b is the selected distance above or

below the macula, a the required increase (or FlG> 3-

decrease) in the lateral angle and A the angle Angled =i°— 35°,

of i ° 35'. |=Tan^,

This gives .1382 of a degree which is to be Tan A — .02764,

added to our lateral angle (angle of conver- *=5»

gence) when we can proceed as in the first case =0° 8' wr
to find the distance from the base line to the

selected point below the point of fixation (Fig. 4), - = cot A, in

which b is the distance sought, a the base line, 1.25 inches,[^4
the angle opposite #, 5°. 138. From this we find that b = 13.904
inches.

To obtain the distance of the point above the point of fixa-
tion we must subtract the o°.i38 (o°8'i7^") from 5° when

198

GEORGE T. STEVENS.

by the same formula we find the distance to be 14.6976 inches
(Fig. SO

We have now the distances

5° above the point of fixation 14.6976

At the point of fixation 14.287

»S^ 5° below the point of fixation 13.905

Forming from these distances two triangles by joining the
three lines at their extremities we have a line joining them and

FIG. 4.
Angled = 5° 8' 54"-

forming bases which represent a vertical line in the horopter at
the level of the page of the book (Fig. 6).

The acute angle at this surface of the book for the upper

FIG. 5.
Angle A = 4° 51' 7^".

triangle of these two is, 69° 38'. That of the lower triangle
is 70° 48' 50."

We have thus, in the space of 10° up and down the page, a
curve of about i ° In other words the horopter in this direction
is approximately a plane surface. If the calculation is carried
to 10° each way, as in Fig. 6, equal to a space to about five
inches on the page of the book the result is nearly the same

ON THE HOROPTER.

«99

but the curve is somewhat greater as we approach the periphery
of the field of vision.

This gives us the position of the page in relation to the plane

FIG. 6.

of regard in which the horopter is most completely formed and
we find that the page is tilted about 15° beyond the right angle
with the plane of regard, or at about 105°. We have found only

FIG. 7.

the direction of the horizontal and vertical meridians of the
horopter but any other meridian may be found in a similar
manner.

200 GEORGE T. STEVENS.

An interesting and very simple experiment for those who
are able to unite stereoscopic figures by convergence without
the aid of a stereoscope beautifully confirms the above cal-
culation.

Draw two vertical lines parallel and at a distance of two
and one half inches from each other on a card board (Fig. 7).

Hold the card board so that in fixing the center of the lines
the gaze is directed downward 35°. Hold the card board
twenty-eight inches from the eyes.

One who is expert with such exercises will be able to unite
the two lines at the distance of fourteen inches from the eyes.

If, instead of permitting a perfect union of the lines in the
stereoscopic image they are held at about one eighth of an inch
asunder it will be easy to find at what angle the board must be
held to render the two stereoscopic images exactly parallel.

In my own case I find by numerous experiments and careful
measurements that the board must be tilted forward as nearly
as can be ascertained exactly 15°.

I have i° of declination of the right eye which would have
little influence on the experiment.

Thus mathematical and experimental research lead to prac-
tically the same result in locating this horopter. By the formula
given we may locate any horopter in the median plane. In
other planes the formula will be modified.

Without discussing the application of these principles to
space perception, a field of much interest and in which many
empirically known facts in art and in architecture may be
analytically tested, only brief time remains to allude to the
more practical application of the horopter.

All the discussion which has preceded has been based upon
the assumption that the adjustments of the eyes are typical in
the sense of being the most favorable to the function of com-
bining the images of the two eyes in a horopter.

In real life anomalous conditions of adjustments, conditions
which interpose difficulties in forming perfect horopters, are of
extreme frequency.

These anomalous conditions may act as slight hindrances or
they may prevent any but an imperfect horopter from being
formed.

ON THE HOROPTER. 2OI

Let us consider some of these.

It has'been seen that with a given depression of the plane
of regard and a given convergence a horopter is formed in
a position which can be predicated when these two elements
and the length of the base line between the nodal points are
known. The depression of the plane of regard is controlled
by an impulse which is not accidental or ephemeral, but which
is automatic and uniform for different persons for the same de-
pression under like circumstances.

Suppose a person whose eyes are so adjusted that with the
minimum of impulse to the governing muscles they are directed
8° or 10° of arc above the plane of best adjustment. Among
people of New England ancestry this is almost a characteristic
as it is with some other groups of people. It is not a disease, it
is the normal development from a certain form of cranium.

Suppose again that this person takes a book in hand to read.
He holds it in the position and at the distance which we have
assumed for our third horopter. Is it not plain that this person
must not only depress the plane of regard the 35° assumed, but
that he must also induce an additional depression of 8° or 10°
as the case may be?

This extra depression at once automatically induces a
greater tilting of the meridians. No horopter can then be
formed. To remedy the difficulty in a measure the person may
throw the head forward 10°, but in so doing there is some dis-
turbance of the equilibrium of the muscles, hence even with this
concession the horopter, which will be better than before, may
still be somewhat imperfect.

In several papers I have shown that as a matter of fact
people with this adjustment of the eyes do throw the head for-
ward and the bending of the neck is, other things being equal,
in proportion to the excess of the normal upward direction of
the eyes. We will return to this presently.

A second condition which may interfere with the formation
of a horopter in the appropriate position is in direct contrast to
this. The eyes may be adjusted so that the plane of vision is
normally directed low.

Suppose one whose eyes are 10° too low. By the same

202 GEORGE T. STEVENS.

reasoning as before we see that because the dynamic depression
of the gaze would be less than in the typical adjustment the tilt-
ing of the retinal meridians would be insufficient for the horopter
and such a person must force the chin high in the air in order
to be obliged to depress the gaze sufficiently to induce the
necessary torsion.

I have written of these conditions and writers have inter-
preted the difficulty as a strain on the muscles of depression or
elevation.

This is an entire misapprehension. It does not follow that
there is any considerable strain on the muscles of adjustment
but the head must be placed in position in which the automatic
torsions shall in some measure correspond to the direction of
the gaze.

A third form of hindrance to the constitution of the horopter
is found in the condition which I have called declination.

This consists of an anomalous leaning of the meridians of
one or both eyes. It is a very common defect and results in
great perplexity to the adjusting muscles. Its practical im-
portance is greater than those conditions already mentioned. It
may induce, like the two conditions named, a throwing forward
or a tipping backward of the head, depending on the direction,
symmetry or degree of the declination defects in the two eyes.
All that has been said about the forward and backward holding
of the head in the other conditions may apply to these cases and
in some instances the unnatural pose of the head and body
from this cause are extreme.

What I have to add might perhaps better be addressed to a
company of physicians than to psychologists yet in order to
comprehend the importance of a subject we must know some-
thing of its practical application.

Recall the case of the person whose eyes are adjusted for
too high a plane. The head is thrown forward as part of the
automatic process of adjustment. The larynx is partly closed,
the chest is sunken. Air passes less freely to the lungs than it
would were the head held erect. It is among this class of peo-
ple that consumption commits its ravages. There are few, if
any consumptives who do not have a high adjustment of the

ON THE HOROPTER. 103

eyes or a form of declination which induces a corresponding
head position.

Then there is the person whose eyes are adjusted for too
low a plane and whose head is thrown back.

It is with this class of persons and with those whose decli-
nations induce a similar pose that the occipital neuralgias, pains
in back of the head and neck and in the lumbar region are
found. The number of such persons is enormous and the suf-
fering from this cause infinite.

From declinations which do not induce false carriage of the
head arise perplexities in adjusting for the horopter which re-
sult in headaches, dyspepsias and a long array of nervous ills.

A subject whose importance cannot well be overestimated
has been presented in this brief outline in the hope that not-
withstanding the necessarily incomplete nature of the presenta-
tion, some interest may be awakened among men whose special
training peculiarly fits them for more elaborate investigations in
this most difficult yet notably practical field of inquiry.

THE LOGICAL AND PSYCHOLOGICAL DISTINCTION
BETWEEN THE TRUE AND THE REAL.

It was Mrs. Carlyle (was it not?) who said that ' mixing things is
the Great Bad.' To the writer it seems that there is a peculiarly in-
jurious variety of the ' Great Bad ' in much of our recent psychological
logic. It is because that sort of philosophy which the writer for over
fifteen years has been calling ' dynamic ' and which now seems to
have come to its own under the name ' functional ' — it is because, we
say, that this kind of dynamic philosophy and functional psychology
is peculiarly adapted to correct this ' mixing of things ' that the writer
offers a few words upon the distinction between the ' real ' and the
' true.'

This sounds like a question of definition and a matter for logic to
dispose of, but we submit that it is also a question of psychology, and
that psychology has already made a distinction (also a matter of defi-
nition, to be sure, the facts having been understood from time imme-
morial) peculiarly adapted to explain the logical distinction here
required.

It is remarkable that recent writers seem not to have been aware
of the ambiguity arising from the identification of the real with the
true. The present writer has elsewhere defined reality as ' affirmation
of attribute ' and this dynamic statement may usefully be contrasted to
Lotze's descriptive definition that ' reality consists in standing in rela-
tion.' Upon critical analysis the two statements come to the same
thing, but our present method in both metaphysic and psychology re-
quires the dynamic form. Nothing can be real apart from a realizer.
As Hoeffding says, * The real is what we apprehend as real — which,
in spite of all effort to the contrary, we must ultimately leave as it is
— which we cannot but recognize,' though he at once goes on to con-
fuse this real with what is true.

It may be assumed that all will agree with our definition of simple
reality as a statement of metaphysical reality. Dewey says : ' The
copula gives the statement of being, asserts the reality.' But he, too,
goes on to discuss truth as relational. In our own extended discussion
we endeavor to point out the union of subjective and objective in this
204

DISCUSS/ON. 205

identification of essence and attribute, which is only possible in an
active percipient.

The logical abstraction of ' pure being ' as the activity of the sub-
ject apart from the content (meaning — *. e.t attribute) is possible,
but it involves, as Hegel abundantly showed, the loss of reality. Pure
being and non-being were in this sense the same, both being all one
to the subject who demands the act of asserting or identifying as well
as the mode asserted.

Professor Baldwin has made, as we intimated above, the important
distinction between psychic and psychological, and both Professor
Bawden and the writer have shown that the psychic cannot become
the subject of scientific analysis. Nevertheless it does afford the
foundation on which science (the psychological) must rest. The
predicate of reality pertains and can pertain only to the psychic. We
do not construct reality but simply perceive (affirm) it. This ultimate
fact in experience is reality. The opposite to real is not false, it is
non-existent or unreal.

If it be objected that this limitation does violence to common usage
it must be replied that any necessary logical distinction may do the
same. The distinction between psychic and psychological traverses
ordinary usage from end to end but if it expresses a true distinction it
is well worth while to reconstruct terminology. In fact, it may well
be that any further great advance in psychology must wait for a
wholesale reformation of terminology.

The point is that we must have a word for this primary feeling-
cognition which we have called reality. Reality is not something we
say about experience but a quale of experience itself. We ascribe
truth to relations of things or events among themselves, or ultimately
as parts of a universe of things and events. Any reality would be no
less real if it existed alone. If we must use Lotze's definition of
reality as a « standing in relation ' we should say reality grows out of
a relation to the subject alone, but this is a metaphysical after-thought.

When the naked fact of experience comes to be thought about or,
in Baldwin's language, becomes psychological, we begin to develop
relations which are true or false in so far as they do or do not cohere
in an organized whole. The whole duty of science is so to cause the
facts thought of to cohere in an organization. This is the sphere of
truth.

There is a sense, however, in which reality escapes from the limi-
tation of the psychic and sits enthroned over all thinking. In last
analysis elements of our thinking have to be verified by reference to

206 DISTINCTION BETWEEN TRUE AND REAL.

real experience. Sometimes we get a long way from such experience
in abstract thinking. We keep building one set of relations upon
another, trying with all our powers, meanwhile, to keep these rela-
tions true among themselves, much as one might work out the orbit of
a comet, but at last the test is whether things in experience stand back
of the true relations — whether the comet can be really found in this
orbit.

When Hoffding says: "The evidence of reality is given, then
* * * in the firm connection of percepts. We can never be so strongly
convinced of the reality of single things and occurrences, as of con-
nected series of things and occurrences," he has confused reality and
truth. Compared with the earlier statement quoted above, the incon-
gruity appears grotesquely. He said that, ' in spite of all effort ' we
' can't help ' recognizing reality, and that there can be no question of
any other than this subjective criterion, and now he proposes to add
to this once-for-all reality greater reality by multiplying relations.
But this is just the difference between truth and reality. Reality,
once realized, can by no possibility be improved upon or made more
real, while, on the contrary, truth grows more certain the more nearly
all known relations are found to cohere with the given relation.

It is not meant by this, of course, that the truth increases with the
number of instances, as in the common logical fallacy, but truth be-
comes more convincing the greater the scope of interaction discovered.
The truth that all Felida? are carnivorous is not greatly increased by
observing one cat repeatedly to eat meat nor by seeing that one kind
of cat always eats meat, but the finding that a different species of
animal combines feline dentition with a carnivorous habit adds greatly
to the evidence by proving that certain combinations are non-essential
and throwing into prominence the organic or genetic relations.

Bosanquet seems to state the law of reality in the definition :
4 Logic treats of the mental construction of reality,' ' the world which
surround him is there only as an idea, t. e., only in relation to some-
thing else, the consciousness which is himself.' But immediately and,
indeed as in duty bound (his subject being logic) he proceeds to dis-
cuss the true. For him the objective world is * what we are con-
strained to think in order to make our consciousness consistent with
itself.' In other words, reality consists in consistence of relations,
which is precisely truth. Logic might be defined as the science of
truth.

Perhaps the discrimination of reality from truth may even help in
the much discussed problem of the subjective and objective. For ex-

DISCUSSION. 207

ample, when we discover (by a round-about means) that a presenta-
tion has been made to consciousness we also get directly (subjectively)
an affirmation of attribute. This is an ultimate of experience. It
does not make an external world. ' Light is,' and that is all there is
about it. But when I, psychologically, accumulate a lot of data and
construct the concept of substance, this is a matter of relation. The
brightness, heat, weight, etc. are made to cohere in the substance,
' candle,' a thing projected out of self and, by implication at least,
contrasted to self, as an object. All these relations of activities are
true to the extent that they cohere in one system or organism. When
the question arises in our metaphysics as to the truth of the objective
world as a whole, as it will when we become aware of the subjec-
tivity of all knowledge, there is but one answer — the one already
used. The objective world is true because it is in one organism with
the subjective mind. Just as our partial judgments are true or false as
they prove to be founded on relations in one whole, organically, not to
say causally, connected, so the larger judgment 'there is a true exter-
nal world apart from the mere act of perceiving it ' is true only if the
percipient or perceiving force be organically part of the same universe.
No other criterion is possible.

The feeling of reality comes from the immediateness of the ele-
ments of experience.1 It defies analysis and requires no definition and
yet is implicit in all practical life. The judgment of truth, on the
other hand, is a fluctuating evaluation based on relations which are
known rather than felt. The weight of evidence forces me to believe
what is true, I require no evidence to cause an experience to be real
nor will any amount of evidence lessen its reality.

The old illustration of the inability of the blind to realize visual
data though they may weave about them all sorts of relations, of the
truth of which they are fully convinced, may not be realizable by the
non-blind. To this end let us take another example.

A friend of mine who is expert in both physics and physiology,
informs me seriously and in detail that he has discovered that by using
the radium waves £, and passing them through a set of refracting ap-
pliances, he is able to produce a series of irritants which, when ap-
plied to the sensitive nerve plexus in the hollow of the human foot,
give rise to sensations unlike any other. They possess a great keenness
and penetrating force and seem to vibrate throughout the organism by
a process of excessive irradiation. Each of these sensations has the

1 Cf. Baldwin's explicit treatment of ' Reality-feeling ' in distinction from
' Belief in his Feeling and Will, Chap. VII.

208 DISTINCTION BETWEEN TRUE AND REAL.

peculiarity of localization in certain parts of the body. One ' wave-
length ' causes irritation at the root of the tongue and marked increase
of blood supply. Each is also accompanied by its own emotional
response, so that one kind of stimulus predisposes to religious fervor
and exalted egoism and the other causes morose and turbulent pas-
sions. One even produces a violent desire for something of which no
concept can be formed. Now I may believe all these as true state-
ments of fact but they do not nor can they produce in me any sense
of reality such as five minutes of actual experience might produce.

The writer believes that a consistent limitation of these words to
the spheres respectively indicated will lighten the burden of the stu-
dent of metaphysics as well as of psychology.

The loose use of the words real and true in psychology coupled
with clear consciousness of the distinctions involved is encountered in
James' Psychology. " The sense that anything we think of is unreal
can only come when the thing is contradicted by some other thing of
which we think. Any object which remains uncontradicted is ipse
facto believed and posited as absolute reality" But the only thing
that can never be so contradicted is immediate experience. A subse-
quent experience may explain, it can never annul it. The only things
that can so be contradicted are judgments of relations. The presenta-
tion ' rain-bow ' is real, but the judgment ' rain-bow now in the sky '
can be proven untrue.

If the word ' real ' be considered to have too strongly intrenched
itself in the wide sphere in which it has been used so carelessly, surely a
new word is required for the primary affectation of consciousness called
' sense of reality ' and ' reality-feeling.' The further characterization
* a sort of feeling more allied to the emotions than anything else ' may,
perhaps better apply to the recognition of truth. The reason for this
relation to the feelings will be found in the nature of feelings. The
writer in his inhibition-irradiation theory of pleasure-pain (which has
theory') has attempted to derive all emotional acts, physiologically
considered, from resistance, obstructions, depletion, or other inter-
ference with the flow of nervous impulses, so that there is irradiation
or inhibition respectively. If this derivation be correct it will follow
that all acts of identification must share in this peculiarity. The new
concept meets a barrier at the threshold of recognition which is finally
thrown down and the wave of thought finds outlet in a path of least
resistance, it is identified with previous acts. This release affords the
recently received a psychological restatement by Fite1 as ' resistance

1 PSYCHOLOGICAL REVIEW, X., 6.

DISCUSSION. 209

condition for pleasure. Identification in one form or another, is back
of nearly all intellectual pleasures. Discovery of a true relation is
accompanied by pleasure, failure to identify is painful.

It is not without interest in this connection to observe how easily
and satisfactorily the dynamic (functional) psychology disposes of the
confusion expressed in the classical discussion between nominalism,
realism, and conceptualism. So long as precepts, recepts, Anschaun-
gen, concepts, and the like, are conceived as possessions or contents
of the mind this discussion is inevitable, but when we become fully
aware that these are names for acts or parts of processes the difficulty
disappears.

When a mode is perceived there is a simple psychic act, even
though the stimulus be of the most varied character. Here we have
to draw a line as important as any in psychology. We, from the out-
side as observers, say that a stimulus has been perceived, but what
we actually did was to affirm a mode (quality, attribute). Subsequent
(psychological) activities consist in combinations of this material into
relations. The act of perceiving does not posit any relation (unless
the implicit relation to the subject be so considered, and this is thought
back into the psychic and is a matter of metaphysic and not of psy-
chology). Psychological work is all apperceptive ; its processes are
all synthetic (even its analyses). What Romanes calls a ' recept ' is a
thinking together of percepts. This unifying work of consciousness
is a function of its unity which, as an equilibrium, is organically nec-
essary. All organization must unify.

Here, for example, is a roll of paper passing through a ruling
machine armed with many pens. I load one pen with blue ink and
from that time forward a blue trace moves along the paper along with
the red, green, and black traces. I may |shift the adjustment here
and there and these traces are brought into various relations, forming
patterns, etc. The initial inking is the perceiving. This process adds
to the activities in the mind a new one which may be shifted, com-
bined, and modified in various ways but never thereafter will the
mind-process-group be the same as it would have otherwise been.
The psychic equilibrium has been changed. The relations between
the several percepts is infinite but some of these are employed instead
of others in our constructive thought. Out of activities, all of which
cohere in an organism, our selection of part and our conception or
thinking together is more or less an act of violence and must always
so remain. In so far as a teleological nexus is formed the thinking
together is true, in so far as the union is a purely arbitrary one or
non-teleological, it is false.

210 THE PERIOD OF CONVERSION.

For example, in our classification I have a concept, ' gopher-genus
Geomys.' Another naturalist has another idea of the limits of the
genus. Our concepts may be equally true but this truth consists in
both cases in the recognition of a Ideological bond. He perhaps
includes more of the segment of evolution or career of the ' gopher
movement ' in nature than I do. The difference is nominal, the agree-
ment is conceptual. We may not say in either case that Geomys is a
real thing but it does in both these cases represent a true concept.
Let the generic limits of Geomys once be set and agreed to, I then
place in the genus an animal proving to belong to another line of
descent, the reference is false. It is a question of relation.

It is wrong to say that a concept is only a name. It does exist in
nature as the subjective expression of the truest thing we know and
the most important. It is a ' genetic ' verity. It is a career — a doing,
in relation with all doing. It is a teleological verity.

But, it may be said, we are only holding a mirror up to nature and
see the trajectory of a flying bird, for example, momentarily depicted
thereon, or we are but exposing a sensitive plate in a telescope and
get only a bright trace thereon. But these illustrations do not go far
enough. In order that the plate may receive the star trace correctly,
the mechanism of the telescope must follow the path of the star.
There must be coordination. So our concept is a conceiving or fol-
lowing of the trajectory of nature. The proof of correctness is exactly
the interaction. Our conceptualism has, therefore, a link to realism
in that only upon the assumption that we are part of the organism
from which the stimulus comes could these correspondences become
intelligible. When we no longer find the trace on our photographic
plate we adjust the movement. The feeling of reality and the con-
viction of truth have their justification in the monistic construction of
organism. C. L. HERRICK.

SOCORRO, NEW MEXICO.

THE PERIOD OF CONVERSION.

The recent scientific study of religious experience has led to many
interesting and important results, not the least of which is the relation
between conversion and the period of most rapid growth. But the
difficulties attending the study of such phenomena are evident here as
elsewhere. Among these difficulties may be mentioned that of deal-
ing with a very complex group of mental phenomena without any
adequate method of controlling their conditions. The usual method

DISCUSSION. an

employed is the * questionnaire,' supplemented more or lew by hyp-
notic and other experiments. The chief reliance must be placed upon
answers given at a distance, to questions which may or may not have
been perfectly understood, by persons more or less accustomed to
scientific observations and dependent upon their memories of event*
which transpired in the past. Several obvious sources of error lie in
this method. First, the unreliability of memory and second, the un-
reliability of the observer's judgment as to the meaning of the question
and as to its proper answer. But a third objection and this time a
purely psychological one, arises from the complexity of the phenomena
under investigation. Professor Wundt used to say in private conver-
sation (but whether he has ever put it in print the writer is not able to
say), that one can seldom be sure as to the meaning of answers secured
by the questionnaire method, because of the complexity and consequent
variability of the factors entering into the observer's judgment. De-
scriptive data of a more or less exact character are the most that can
be hoped for.

But this matter of investigating religious experience is beset by
two further difficulties which threaten its scientific value. To be
thoroughly scientific, experiments and observations must be of such a
nature as to be repeated by others ; while in these matters, both the sig-
nificance of most of the questions asked and the interpretation of the
answers are alike beyond the control of other investigators. Conse-
quently, the valid results of work in this field are fewer in number
than might be expected considering the number of investigators and
the extent of their efforts.

These difficulties are illustrated in the investigations as to the
period of most frequent conversions. Different writers do not seem to
take the term conversion in the same sense. To one author it means
the change in human character by which any set of religious ideas and
aims become the center of a man's life; to another it is that change in
man's character by which Christian ideas and aims become central in
his life. Now, for the psychologist, Christianity is a set of aims and
motor attitudes belonging to religion as its genus; Christianity is a
species of religion. Men may become religious without being Chris-
tians although they can not become Christians without becoming
religious.

With this distinction in mind let us look briefly at this question of
the period of conversion. By religion in general let us understand a
natural consciousness of relations to a Totality of Existence of which
we are a part and upon which we depend, together with the beliefs,

212 THE PERIOD OF CONVERSION.

types of action and institutions which have grown out of this con-
sciousness. Lower animals and young children are not religious be-
cause they have no notion of themselves as subjects of experience,
They do not think of themselves as personally identical, as capable of
right and wrong conduct and of laying plans for the future. A man
may think of himself as an animal and regulate his conduct accord-
ingly; he may think of himself as a member of a certain church,
family, club or political party ; and in each case his thought of himself
and his belief about himself are the most important factors of control
in his conduct and character. If it were possible for me to determine
your thought of yourself I could afford to let your conduct take care
of itself ; if I had absolute control over your idea of yourself, I could
let your religion take care of itself. This reflective thought of self is a
man's recognition of his peculiar place and function in the totality of
life out of which he has arisen, to which he is organically related and
upon which he depends. He may conceive himself, as Tolstoi says, as
a spirit passing through a series of existences so related to each other
that his conduct in one existence determines his position in the next.
He may conceive himself to belong to a people chosen by God to re-
ceive protection and blessing, upon condition of perfect obedience to
God's commands. He may think of himself as one of the thoughts
and purposes of an Absolute Being who is perfectly rational and per-
fectly, good. However he may conceive the Total Existence of which
he is, in some sense, a part, and however he may define his position
in it, this conception and definition together with the emotions and
conduct growing out of them, constitute his religion, his faith. Con-
version, in the naturalistic sense of the term, is that inner change by
which some such conception and definition, constituting a set of religi-
ous ideas and aims, become central in a man's personality.

Now I suppose it is conversion in this sense which has been found
to be one of the regular phenomena of adolescence, usually occurring
somewhere between the ages of twelve and twenty-five, the year of
greatest frequency being the seventeenth. Accordingly, the period of
most frequent conversions seems to come just after the periods of
greatest brain-weight and of greatest increase in body-weight. Pro-
fessor Starbuck's conclusion is that the periods of conversion and the
periods of most rapid bodily growth tend to coincide.

I suppose the most important result of these investigations is the
conclusion that conversion, in some broad sense, is the normal experi-
ence of every man, marking the transition from his childhood to his
youth, or that from youth to his manhood. It is the step by which an

D/SCUSSJON. 313

adequate sense of selfhood is approached and made possible. Con-
version as a natural phenomenon is a deepening and broadening of
one's ideas of himself which is at the same time a deepening and
broadening of one's ideas of others; it marks the entrance into a new
life based upon a profounder view of the kind of being one is, of the
class of beings one belongs to. Conversion is an affair of the social
consciousness ; by it a youth comes to feel that he belongs to a noble
company and to a divine order of things; by it he enters at last upon
real life in a real world.

It has been objected that this period for the greatest frequency of
conversions comes too early and that conversions frequently occur far
on in life. Certain it is that a large part of the work of the church
has been directed in vain to the converting of adult men, if the period
mentioned above is the only period of possible conversions. Now with
reference to this objection, aside from the doubt which attaches to the
conclusions of Starbuck and Coe as a result of the methodological diffi-
culties mentioned above and the relatively small number of cases ex-
amined, I should say that the word conversion is here used in our two
different senses. As used in these investigations it seems to me to
mean the natural coming to himself which every normal man, be he
Hindoo or Hebrew, Latin or Greek, at some time or other experiences.
To those who object to the results of these investigations it means the
accepting of Christ and Christian truth as the central religious facts of
the universe. Just what view of life and what attitude toward the
world a youth is converted to will depend upon the training and sur-
rounding influences of his life ; that is, it will depend upon imitation
and suggestion. He may by conversion become a Hindoo, a Moham-
medan, or a Christian ; he may be converted, like Tolstoi, to a simple
faith which he understands to be the universal essence of all great re-
ligions ; he may be converted to some country and make patriotism his
religion ; he may be converted even to some calling so that henceforth he
defines his relation to the world as that of a miner or a teacher of En-
glish Literature, or a preacher of the gospel of Christ. I believe there
are men in whose lives patriotism or devotion to some calling have,
for a time at least, nearly all the essentials of a religion. That an in-
dividual should undergo conversion is insured by the laws of mental
growth and the conditions of social intercourse; but that he shall be-
come a Christian by conversion must be insured by teaching, preach-
ing and living Christianity. Hence the significance of the church, of
family worship and of all those noble agencies through which Chris-
tian truth is taught and Christian attitudes are trained.

214 THE PERIOD OF CONVERSION.

That Starbuck and Coe must be nearly right as to the time of
conversion in this general sense of the term, I think there can be but
little doubt. For consider what the adolescent period is. It seems
necessary to keep saying that puberty is not the whole of it, nor even
its most important part. At about the tenth year girls begin to grow
more rapidly than they have ever grown since infancy. Something like
a year later, boys start, outstripping the girls about the fifteenth year and
ultimately attaining the larger stature of the two. They first shoot up
like iron-weeds, then broaden out and then fill up. At about fourteen
in girls and fifteen in boys the brain weighs more than at any other
time in life. During this period new organs develop, new instincts
and acquired reactions show themselves, new centers in the sympa-
thetic and central nervous systems develop and begin functioning.
Moreover, growth is never proportional throughout the body. One
organ after another and one nerve-center after another starts into
activity and then subsides ; and with these spasmodic developments,
the youth's interests flash up and die away. Now he will be a great
poet and artist, now a great orator and statesman, now a great ad-
venturer and desperado or a great naturalist or a great inventor. And
all the time he is living under an enormous blood-pressure and the
demands of his growing organs are draining the energies of his central
nervous system. If a youth does not discover himself at this time of
disappointment and growth and trial, at what time in life is he more
likely to do so? As a matter of fact the mental characteristics of the
period are just what we should expect — a deepening awareness of
self such that all experiences, especially in girls, come to have an in-
tensely, sometimes a morbidly personal reference ; an insatiable crav-
ing for sympathy and comprehension from those who are older;
an extravagant passion for self-sacrifice, and a certain fickleness
and fancifulness of interests, ambitions and tendencies. That re-
flective self-awareness develops especially at this time, is shown
by the fact that when this craving for sympathy and this impul-
sive self-sacrifice are not satisfied by wholesome family and social
relations, certain morbid types of self-consciousness are apt to result.
Such morbid types are manifest in the desires to enter monasteries
and nunneries, to become trained nurses and heroes of tragedy, and to
commit suicide rather than endure the awfulness of living. Here also
belong the feelings of many youths and maidens that they are .different
from other people, that they are hopelessly bad and alien, that no one
understands them, and that God (if there be a God) has somehow
left them out of His great plan for His world. The more one studies

DISCUSSION. JI5

%

the mental states of adolescent boys and girls and the more one com*
prebends their fierce doubts, their titanic yearnings and their tremen-
dous burdens of anxiety and fear, the more one is convinced that this
is the natural time for the great awakening. At this time, and par-
ticularly just after the period of greatest bodily growth when the
energies of the central nervous system are no longer drained to supply
the demands of developing organs, one should come to realize one's
place in the experience of God and in the institutions of His world.

But the term conversion as used by some recent writers means
that change in man's religion by which Christ comes to be its center.
If the foregoing view is true, every normal human being should un-
dergo conversion some time within or near the adolescent period;
every mature man is in this sense of the word religious; but to be-
come religious in this sense is obviously not the same thing as to be-
come a Christian. Two classes present themselves, viz., those who
in their early conversion become Christians and those who in their
early conversion become religious without becoming Christians. In
the latter class I do not see why conversion to Christianity may not
occur at any time after the adolescent period. In this second sense of
the term, conversion may occur many times in the course of a life;
and it is the almost universal experience of ministers and teachers that
conversion does occur at any age from the twelfth year until death.

It is in the belief that the term conversion is used in these two dif-
ferent senses indiscriminately by different writers of the recent scien-
tific movement in the study of religious phenomena, that the two
ought to be kept distinct especially in dealing with the question as to
the periods of conversion, and that their confusion has resulted from
the complexity of the problem, the limited number of cases examined,
and the necessary inadequacies of the method used, that this article is
written. To the writer it seems idle to condemn in toto all efforts to
attain exact knowledge as to the psychology of religious experiences,
and we see no reason why recent efforts in that direction should not
be welcomed and encouraged. But it seems to be a field in which re-
sults must finally be reached by deduction from general psychological
hypotheses which are yet to be inductively established in other depart-
ments of the science. Meanwhile such deductive procedure is aided
by such facts as Leuba, Starbuck, Coe and James are seeking accu-
rate, systematic and exhaustive accounts of.

As to the time of conversion, two separate lines of investigation
suggest themselves as extensions of the work already so faithfully done
in a field where, both from the point of view of the science of psy-

2l6 THE PERIOD OF CONVERSION.

chology, from that of the practical worker in education, and from that
of the conscientious individual seeking an intelligent control of his
judgments in religious matters, light is so sorely needed. First, an
elaborate series of investigations carried out in different lands among
persons of different religious beliefs for the purpose of comparing the
religious experiences of people in different countries, climates, races
and civilizations. Secondly, a series of investigations carried out by
teachers and ministers of different persuasions in Christian countries
for the purpose of determining the times, the conditions and the
nature of conversions to Christianity, and to other types of religious
conviction. The difficulties of such investigations, from a scientific
point of view, are enormous and results can only be contributory to a
future consummation devoutly to be wished. The American investi-
gator would be obliged to work through missionaries and college
teachers in remote lands, and this circumstance, in addition to the in-
herent difficulties of our methods, necessitate the utmost judicial care

in sifting results.

G. A. TAWNEY.

THE GENETIC PROGRESSION OF PSYCHIC OBJECTS.

The recent relatively novel attempts in the literature to approach
the logical processes from the genetic point of view, have made it
clear that a good deal of close psychology is still needed in this field.
What has impressed the present writer is the lack of an actual tracing
out of the series of determinations of objects at the successive stages of
cognitive development, and the motives in each such progression from
one 4 psychic object ' to the next. This is the topic to which I am
here applying the title of this short paper. The term ' progression '
is one which I have used in a somewhat technical sense elsewhere l ; it
denotes a real genetic movement from one mode or stage of develop-
ment to another.

In a series of university lectures, which are to be revised for
chapters in a forthcoming work on the genetic treatment of the logical
operations, I have worked out a tentative scheme of the sort ; • and as
its points of emphasis are not altogether those hitherto familiar, I ven-
ture to present it here for preliminary criticism ; hoping as well that it
may incite to a renewed discussion of the general topic.

1 PSYCHOLOGICAL REVIEW, May, 1903. I use the term ' object ' in the sense
of whatever consciousness means or intends — that is whatever can be in any
way, shape, or manner psychically set up, presented, or aimed at. Cf. the
•writer's Diet, of Philosophy, sub verbo.

DISCUSSION. 317

In a preliminary demarcation of the field, we may ask two broad
questions : first, what are the conditions determining the construction
of objects at any given stage of mental development; and second, what
are the psychic characters of the objects thus determined at any stage.
Of course, the treatment of * any stage ' means the treatment of * every
stage,' and that involves the determination of the entire continuous move-
ment of the cognitive function, with the ranging of all the objective
determinations or specifications of psychic objects in a genetic series.

In the process of bandying this question about — making it every
sort of psychic object! — the following types of enquiry have come to
more or less definite shape. If we take the traditional outstanding
distinctions of sorts of objects, such as sense objects, objects of memory,
of thought, etc., as starting point, we may work out the more evident
characters of such objects, range them in their apparent genetic order,
and call them, as so arranged, the series of ' objective modes.'1 We
may then endeavor to work out the factors of determination for these
modes in succession from the simpler to the more complex, in so doing
recognizing any finer distinctions which appear, and rearranging the
genetic order as we may find ourselves led to do so. This compels us
— or has done so in my own case — to trace out certain relatively inde-
pendent strands of genetic change, the transformations which certain
great phases of psychic process undergo, along with the changes in the
objects proper. These accompanying series, in so far as they are
essential aspects of what we may call the ' object psychosis,' are indeed
necessary to a full statement of the objective progressions. I find it
at least interesting therefore — not to make dogmatic statements as to its
possible value in each case for the main problem itself — to distinguish
in the actual results to which I have been led, the following phases of
consciousness,2 traced in each case along with the objects, through
a series of modes in turn: (i) the controlling conditions of the deter-
mination (that is, the ' control ' of the object, a problem recently made
much of in the writings of Professor Dewey), (2) the motive to the

1 The term mode may be applied, I think, both to the sort of function whose
progressions we are tracing out, and also to each characteristic stage in these pro-
gressions themselves, e. g,, the thought mode is a stage in the development of
the cognitive mode.

2 This description of these series was drawn up in answer to a question
raised by one of my students after the table (Table I., herewith) in which the
results were spread out, was presented on the blackboard ; I say this to avoid
the suggestion that the lines of enquiry were worked out under any prearranged
scheme. On the contrary, the different modal series, as they may be called,
resulted directly from the attempt to analyze and trace out the objective deter-
minations in order.

2l8 GENETIC PROGRESSION OF PSYCHIC OBJECTS.

determinations each in turn (the problem of ' interest,' which I find of
extreme importance in the late,r discussions as to the determination of
'truth,' as 'practical,' or 'theoretical,' or both), (3) the function in-
volved in each determination (the sort of attention in which the actual
interest finds its vehicle), (4) the meaning of the object, over and above
its actual objective marks (here the question of ' logical meaning ' is
of course uppermost, and with it what I call the problem of ' indi-
viduation,' or range of application — in logic, ' quantity ' — and the
question also of ' real reference,' or the psychic meaning of ' reality ') .*

With so much statement of the problem and the way of approach-
ing it, the accompanying Table I. is I think fairly intelligible. The re-
marks upon it which follow are meant to explain the more unfamiliar
features of the results which it exhibits, including the definition of cer-
tain new terms which it has been found necessary to use.1

As to the successive objective determinations themselves ( i to 8 in
col. i.), they are largely explained by what is found in other columns
of the same level, in each case. The principal innovations in the
series consist in the essential progression from Memory objects (2) to
Judged or Logical objects (7) through the stages represented by Fancy
(3, in which arises the dualism of ' inner-outer ') , Play (4, character-
ized by the constructions of ' semblance ' or ' make-believe ' — col. iii.
— and 'experimental control' — col. vii. — ) with the important tran-
sition, through the rise of psychic control and quasi-logical classifica-
tion (col. iii.) to the dualism of ' mind-body,' called the ' Substantive
mode' (5). In fact, I am prepared to insist that what is here called
the ' Semblant ' mode is an absolutely necessary term in the rise of the
great dualisms which make the logical consciousness possible.5

Another point to be remarked is that the dualism of ' Self — not-
self (6) is made germinal to Judgment (7), and that the former
carries with it as also preliminary, and so available to say the least, the

1 Of course many other questions might be asked about the objective con-
sciousness, as e. g.t what its emotional coloring, its conative accompaniments,
etc., but these might just as well be asked in the reverse form — in tracing out
the progressions of feeling or conation. Here we are dealing with what is found
to be necessary in (i) the determination and (2) the characterization of the ob-
ject qua object.

1 As to the new terms, they are of minor importance, of course, and need not
claim to be ' fittest ' ; the suggestions of others would be most welcome on this
as on other features of the matter.

* Of course this and the other essential features of the progressions, so far
as in any degree novel, are argued in detail in the full treatment, which is to
appear in a volume from the press of Messrs. Swan, Sonnenschein & Co.,
London.

DISCUSSION. 119

« social-personal ' distinction. This means, it would seem, that a strain
of social worth appears in all determinations of judgment. Further,
as to judgment, it is found to be the criterion of the Logical, properly
speaking (col. ii.), although the progressions up to it, through the
preliminary dualisms (3, 5, 6), illustrate strikingly the fact of conti-
nuity. The modes 3 to 6 are in a very real sense * quasi-logical.' '

Note also that 'Moral objects' (8) are 'extra-logical,' except
when made matter of theoretical interest (col. v.), and that './Es-
thetic objects ' (9) are hyper-logical, in the sense of having both prac-
tical and theoretical Individuation (col. Hi.), and also as involving a
higher form of interest and control.

Indeed, still speaking of the aesthetic, I may add that another of
the points most in need of clearing up, and hence earnestly worked at
here, is the relation of the two forms of ' Semblant ' objects, those of
Play and Art (col. iv., 4 and 9), considered in respect to their psychic
meanings, to the other forms of objective construction. It is my con-
viction that in both of these — and it is part of the fact that any pos-
sible psychic object may be determined as one of them — we have the
genetic resolution of the dualisms and pluralisms of the various cogni-
tive modes as such. So I find it necessary to use terms which lack the
partial connotations of those employed for such modes. The aesthetic
is ' hyper-logical ' 2 (as explained above) ; it is ' contemplative,' or
as regards its end, ' pan-telic,' having both practical and theoretical
interest; and it has the further extraordinary character that it is
under what I venture to describe as ' syn-nomic ' control : that is, it is
a form of determination in which both the psychic and also the psy-
chically-recognized-as-foreign conditions of determination are satis-
fied. There is here a higher psychic immediacy in which all the du-
alisms of the mental life, at the stage reached, may on occasion merge
in an immediate contemplative value of real presence; the dualisms
of 'theoretical and practical,' 'mind and body,' 'inner and outer,'
' freedom and necessity,' all merge to the vanishing point in the
aesthetic.

It only remains to be added, in the consideration of the objective
progressions as such, that the dualism of Self-notself is described as one
of ' Content,' inasmuch as it arises only when the dualism of Mind-
body, gives place to that distinction within the psychic sphere in which

1 The progressions in the development of the Logical mode itself are matter
for later statement, as are also those of the Self-social mode (cf. this REVIEW,
May, 1903, pp. 226 ff.).

2 Meta-logical, suggested to me and otherwise apt, has been preempted for
the sense given to it by Schopenhauer {Fourfold Root, 5 33).

220 GENETIC PROGRESSION OF PSYCHIC OBJECTS.

part of the content is set off as ' self ' over against the rest of the ob-
jective content or ' not-self.'

The progressions of the ' Individuation ' mode (iii.) are at once
most difficult to trace and most important in relation to the questions
of logical value (ii.) and real reference (iv.). The considerations in-
volved are so detailed that the catchwords given in the column iii. must
suffice in this place. Yet attention may be drawn to the position that
the recognition of 'class' is the term of transition from the Play mode
to the Substantive mode, and that the recognition of ' general meaning
or intent ' is preliminary to the 'general concept' which alone is logical.
Psychically there is reason also for maintaining that all individuation
is a function of recognition.

The Reality progressions (iv.) are in familiar terms ; though I may
remark that by ' practical judgment,' I mean the sort of practical use
of means to ends recently described by Hobhouse {Mind in Evolution)
as probably occurring sometimes among animals. Genetically it seems
to be closely associated — as regards the psychic elements involved —
with the experimental treatment of objects so conspicuous in the Play
mode (see col. vii., 4).

In the Interest progressions (col. v.), one should note the overlap-
ping of the practical and theoretical interests, and the arrangement of
the psychic stages with reference to the ends of the interests, namely,
as * a-telic,' (e. g., without psychic end), ' auto-telic,' J practical
(?-telic),2 theoretical (?-telic),J and 'pan-telic' (inclusive of all sorts
of ends). It may also be noted that the rise of theoretical interest
is put in the transition from the Substantive to the Content mode —
the interest which motives the distinction between self and not-self
being both practical and theoretical.3

The progressions in the Attention mode are sufficiently explained
by the explanations of the genetic formula for the attention given in
the chapter of an earlier work where that formula is proposed (Ment.

1 Cf. the Did. o/Philos., sub verbo.

2 Suggestions of proper compounds in these two cases are in order ; possibly
' pragma-telic ' and ' noo-telic ' would do. It is advisable to confine the term
'practical' to the objective psychological point of view, and to use 'pragma-
telic ' for that psychic ; for pragmatelic interest is not at all coextensive with
practical interest. To make the same distinction general, as between 'telic '
(psychic) and ' teleological ' function, would aid in banishing the utter con-
fusion which prevails in the use of the latter term. The teleological is the end-
attaining, to an observer ; the ' telic ' is the end-seeking in psychic process.

'Theoretical in the germinal sense of being experimental — a necessary
phase of theoretical interest, as I believe.

DISCUSSION. aai

Dcvel., chap. X., § 3).' The arrows are explained by the terms
theoretical and practical with which they are associated in col. v.

The Control modes are necessarily described as * mixed ' in respect
to the contrast of psychic and psychological (or objective), inasmuch
as the only possible variations in the description of the sorts of con-
trol are those characteristic of the two contrasted points of view. For
instance, control is ' heteronomic ' to the psychic, when it is described
as external (e. g., biological, organic, etc.); it is 'a-nomic' when it
is or appears to be lawless from both points of view. Later on in the
progressions we have terms in use for both sorts of control : 4 belief '
over against ' truth ' (and * fact '), ' duty ' over against ' right,' ' aesthetic
quality ' over against ' beauty ' (this last being |a joint sort of control
covered by the term ' syn-nomic '). A further point of interest to the
writer is that suggested by the double brackets of different lengths be-
tween 5 and 7 (col. vii.) ; namely, the point that the theoretical form
of control ('fact' and 'truth') extends from 5 to 7 — over a certain
range of objects — while there are also other control-forms extending
not only over the same range, but beyond it in both directions.

It may well be that such a schematic presentation as this has no
value or suggestiveness ; and I should not be surprised to hear this
opinion expressed. But the tentative character of the results, and the
absence of the detailed grounds which are to my mind reasonably
strong, may be just the needed stimulus to some one to treat the topic
more fruitfully.

J. MARK BALDWIN.

JOHNS HOPKINS UNIVERSITY.

NOTES.

I TRUST that the paper of Professor Max Meyer, in the RE-
VIEW for March, on ' Attributes of Sensation,' may stimulate reflec-
tion, if not discussion, on that subject. Personally, I prefer the term
'element' for what Dr. Meyer calls 'attribute,' but that is mainly a
verbal difference between us. The positive contribution of the paper
seems to me to be the teaching that the existence of a psychologically
simple ' tone-quality,' which varies with the pitch and yet is intro-
spectively distinct from it, discredits independent variability as a prin-
ciple of distinguishing * attributes ' of sensation.

1 The genetic formula Attention = A -f a -f a, in which A stands for the gross
muscular and other sensational processes of attention, a the added contractions,
etc., of recognizing a class (e. g., visual objects), and a the finer adjustments of
individual recognition.

222 NOTES AND CORRESPONDENCE.

The paper is marred by the uncritical adoption of duration as at-
tribute of sensation in the sense in which qualities, intensities, and
extensities are said to be attributes. But duration as content of con-
sciousness is a complex, not an elemental, experience ; and duration
viewed as attribute is predicated of all events, physical as well as
psychical, and is therefore not an attribute at all, in the psychological
sense. Even a psychologist who does not admit the preceding state-
ments, ought at least to recognize that the traditional treatment of
duration has been challenged.1

MARY WHITON CALKINS.

WELLESLEY COLLEGE.

On account of the accumulation of material the REVIEW will issue
a double number (July-September) on July i . The PSYCHOLOGICAL
BULLETIN of June 15 will also be a double number (June-July) de-
voted to Mental Pathology, of which Dr. A. Meyer, of the New York
Pathological Institute, will be the ' effective editor.' It may be added
also that no less than four Monograph Supplements are in our hands
for immediate publication. The annual bibliography, THE PSYCHO-
LOGICAL INDEX, is to appear in a few days. It shows a marked
falling off from the last year in the number of titles listed.

THE EDITORS.

1Cf. on the duration problem, a paper by Professor M. F. Washburn, this
REVIEW, July, 1903, and a shorter discussion by the writer, ibid., vol. VI., 1899,
p. 506.

N. S. VOL. XL No. 4-5 July-September, 1904

THE PSYCHOLOGICAL REVIEW.

AN EXPERIMENTAL STUDY OF THE PHYSIOLOG-
ICAL ACCOMPANIMENTS OF FEELING.

BY L. PEARL BOGGS, PH.D.,
Urbana, III.

It had practically been taken for granted since the time of
Kant that all feelings might be classified as pleasantness or un-
pleasantness until Wundt propounded his theory of the three
directions of feeling. He names them pleasantness — unpleas-
antness, excitement — repose, and strain — relaxation. This
psychologist believed that he had not only the evidence of his
own introspection and that of others in favor of this view, but
also the results of certain psycho-physiological experiments
which showed that each of the six sorts of feeling had as its
accompaniment certain changes in the vaso-motor system.
These experiments were chiefly those of Mentz ' and Lehman,1
the former having made use of the sphygmograph and the latter
the plethysmograph for recording the radial artery.

Not long since Dr. Max Brahn published an article s setting
forth the conclusions reached from a long series of experiments
made for the purpose of testing the tridimensional theory of
feeling, and he finds six forms of pulse changes, each one of
which is the persistent accompaniment of some one of the six
kinds of feeling. It was also with the purpose of a test of this
theory that the following experiments were undertaken and I
find that in the main the results are a corroboration of those of
Dr. Brahn.

14 Die Wirknng akustischer Siunesreize auf Pula u. Athmung,' Pkil.St.t
Vol. XI.

tffauptgesetzed. Gefiihlslcben, Leipzig, 1892.

•Brahn, ' Ex. Beitrage zur Gefiihlslehre, ' Phil. St., Vol. XVIII.

223

224 L. PEARL BOGGS.

It were perhaps useless to go into a lengthy account of the
tridimensional theory. It is to be found in Wundt's * later psy-
chological works and Brahn has given a short resume.2 It is
very probable that the differences in the one-dimensional and
tridimensional theories arise from inaccurate introspection which
does not clearly distinguish between sensation and feeling ;
from loose concepts as to the meaning of feeling ; and from
vague theories as to the origin of feelings and their physiological
accompaniments. Wundt calls feeling the subjective element
of psychical life and says that in seeking for the physiological
accompaniments ' it follows from the subjective nature of the
feelings, that we should not expect to find them among the proc-
esses produced in the organism directly by external agents, as
the sensations are, but rather in reactions which arise indirectly
from these first processes.' 3 Again Wundt does not allow that
there is such a thing as a concrete state called pleasantness or
excitement, but believes that these are only general names
under which a multitude of feelings are brought which have a
certain general resemblance. But putting theoretical questions
aside we shall pass on to the account of the experiments.

INSTRUMENTS AND METHODS OF PROCEDURE.
Preliminary experiments for this study were carried on at
the University of Cornell during the fall of 1902 but nearly all
of the results which are embodied in this report were obtained
from experiments made at the University of Illinois in the spring
semester and the summer of 1903. I used a Marey's sphygmo-
graph for part of the experiments and for others a smaller and
simpler one, which was very useful when the pulse lay very near
the surface of the arm, as it did with some subjects. The record
was always taken from the radial artery, usually the right one,
with the arm in a supine position and resting comfortably on a
high cushioned stool. For registering the pulse changes air
transmission by means of a thick walled rubber tubing and
Marey's tambours was used, the writing lever being laid lightly

1 Especially Outlines of Psych., 1897, p. 82, ff., Phil. St., Vol. XV.

2 Op. cit.t pp. 127-132. For a criticism see Titchener, Zeit. f. Psy. und
Phys., Vol. XIX.

3 Wundt, Outlines of Psych., p. 86.

AN EXPERIMENTAL STUDY OF FEELING. 125

on the smoked paper of a regularly revolving drum. The
kymograph was rendered as noiseless as possible by placing a
heavy felt mat under it. An electro-magnetic time marker con-
nected with a second pendulum recorded the time while another
electro-magnetic marker was used for recording the beginning
and the close of the experiment, interruptions, or special signals
by means of a double contact key which was inserted in the
circuit and was sometimes in the hand of the operator or an as-
sistant, and sometimes in that of the subject, as the circumstances
of the experiment required. Later on I also used the pneumo-
graph, but more by way of control for the pulse records. In
nearly all cases the thoracial breathing was taken, and while the
instrument was not very sensitive it seemed to be accurate within
its limits. The breathing curve was registered in the same way
as the pulse.

Much has been said for and against the sphygmograph as
an instrument adapted to precise and accurate work. The
criticism most often made is that it gives only the frequency of
the heart pulsations since the plethysmographic effects are so
great that the strength of the pulse cannot be accurately meas-
ured ; that is to say, the volume of the arm changing, the pres-
sure of the spring adjusted to the throbbing artery is changed,
and therefore the excursion of the same is changed. In reply
to this objection Brahn has spoken at some length,1 citing von
Frey as authority.2 The conclusion is that such an error is pres-
ent only where the change in volume is very great, for example,
where the arm is tightly bound or in Valsalva's experiment,
but that in ordinary experiments, especially of short duration,
the error is neglible. Several curves were taken simultaneously
with the plethysmograph on the left and the sphymograph on
the right arm during the course of my experiments, and after
careful scrutinizing of the two records I can find no traces of
such an error. However, I do find the sphygmographic curve
much plainer and more regular with more pronounced and uni-
form characteristics. Of course the sphymograph requires the
most carefnl adjustment to begin with and I found several per-

1 Brahn, Phil. St., Vol. XVIII., pp. 143-144-

2 Von Frey, Die Untersuchung des Pulses, Berlin, 1892.

226 L. PEARL BOGGS.

sons who were unsuitable as subjects on account of the pulse
lying too near the surface of the arm, or too deep, and each in-
dividual required a different amount of pressure to insure the most
accurate results.

The experiments were conducted in a quiet room between
nine and eleven o'clock in the morning and two and four in
the afternoon, the subject sitting usually three quarters of an
hour. My subjects were nearly always in a fresh condition but
for the sake of observing the physiological effects and the varia-
tion in the results when one was tired or slightly indisposed, the
subjects were occasionally taken in a slightly abnormal condi-
tion. One subject slept for twenty minutes during which time I
took a number of records. Another subject who had just been
smoking before he came in showed such a weak and irregular
pulse that nothing much was to be gotten out of it for psychol-
ogical purposes.

The introspections were given conscientiously and carefully
and whenever there was reason to suspect anything wrong the
whole experiment was thrown out. I did not especially train
my subjects in introspection as Brahn did, but occasionally
asked them to compare or contrast certain feelings. I some-
times cautioned them about distinguishing clearly between feel-
ings and sensations. Occasionally an assistant was present and
usually two persons were required to conduct the experiment if
I was the subject. For this service so kindly performed I wish
to thank Mrs. Alice Parks and Professor Colvin, as well as for
their participation as subjects. My other subjects were unprac-
ticed observers, or comparatively so, and were kept totally
ignorant of the real purpose of the experiment in order that their
introspection should not be influenced by any desire to make
the proper introspection.

MEASUREMENT .

The measuring of the curves is a matter of great impor-
tance. It did not seem to me that the highest accuracy was to
be obtained by measuring each pulse length as Brahn does.
In the first place the variations are so small that the mechanical
error in measuring may be considerable; and secondly, the

AN EXPERIMENTAL STUDY OF FEELING.

variations of single pulse curves within a respiratory oscillation
are so great that errors are likely to arise if due notice is not
taken of this fact. Mentz adopted the plan of taking the mean
averages of the shortest and longest curves of each oscillation.
I modified this plan, simply measuring each oscillation and
taking the mean average of the whole number of pulses, be-
ginning with whichever pulse occurred first in the reaction as
the starting point for all measurements for that particular in-
stance. Lehman's method of following the natural groupings
according to pulse lengths seems most unscientific, while some
one has hit upon the ingenious plan of measuring the first half
of the reaction and the second half separately and comparing
the mean averages of the curves. The instrument used in
measuring was a vernier recording tenths of millimeters.

Different rates of rapidity were used for the revolving drum
but for paper 480 mm. long 93 and 80 seconds time was most
frequently used. The measurement of the breathing I shall
only give in words as also that of the height of the pulse.

No attempt will be made to reproduce the curves since it is
impracticable to reproduce all and a few isolated curves would
have little value since each curve should be read in comparison
with the normal curves and other curves of the subject.

A Brief Account of the Results of Other Investigators.

As to the results of the experimental work previously car-
ried on in this field a few words should be said, and first in
regard to the parallel physical processes of attention.

Involuntary attention : Mentz ' finds that acoustical stimuli
received with involuntary attention caused a slowing of the
pulse and usually of breathing. Lehman2 says that an intense
and sudden sensation causing attention usually does not affect
the breathing ; the first pulses are usually short but afterwards
become longer and so the general characteristic of the curve is a
lengthened pulse. Mosso's3 statement that the sensations from

1 Op. tit.

* KSrperliche Ausserungen psychischfr Zustdndc, Leipzig, 1899.
*Diagnostik des Pulses, Leipzig, 1879. Die Tfinptratur des Gehints,
Leipzig, 1893.

228 L. PEARL BOGGS.

the various sense organs cause a slowing of the pulse while the
breathing changes show no satisfactorily uniform result finds
support in the two later experiments.

Zoneff and Meuman l find that there is no difference be-
tween involuntary and voluntary attention but say that a con-
centration of the attention causes a slowing of the pulse and
an obstruction in the breathing. In respect to sense and intel-
lectual attention the breathing is more obstructed in the former.

Voluntary attention, according to Mentz, is accompanied by
a quickened pulse and usually more rapid breathing. Lehman's
results shows a quickened pulse. If it is a case of long-con-
tinued intellectual work the breathing is usually more rapid and
shallow.

Brahn discusses strain as the feeling side of attention, and
prefers the expressions prepared and unprepared attention to
that of voluntary and involuntary. Prepared attention gives
the best form of strain while unprepared attention is likely to
be ushered in by a feeling of excitement, which may be
changed to that of strain only to be replaced again by excite-
ment. Naturally some stimuli to which we attend do not cause
a strain but pleasantness, unpleasantness or quietness accord-
ing to their nature. This state of strain is accompanied by a
quickened pulse and pronounced or low dicrotic while the op-
posite feeling of relaxation is accompanied by a slowed pulse
and high dicrotic. In regard to the fluctuations of attention
Neuman and Zoneff say that according to the degree of concen-
tration of attention the breathing and pulse is obstructed, whereas
so soon as the concentration is diminished it becomes faster.
This statement is to be criticised later on.

As regards pleasantness and unpleasantness there is practi-
cally universal agreement. Mentz finds in the case of the
former a slowing of both pulse and breathing and for the latter
the reverse. While agreeing with this, Lehman says in addi-
tion that pleasantness is accompanied by a heightened pulse
curve and unpleasantness by a lower.

Zoneff and Meuman find pleasantness accompanied by a
less rapid pulse, but with more rapid and shallow breathing,
while the reverse is true for unpleasantness.

i Phil. St., Vol. XVIII.

AN EXPERIMENTAL STUDY OP PEELING.

Brahn agrees with Lehman is regard to pleasantness and
unpleasantness. Excitement is accompanied by a heightening
of the pulse without the lengthening, while repose is accom-
panied by a decrease in the height without change in length.
Brahn follows Wundt's suggestions about the choice of stimuli,
those for the different sense organs being accompanied by dif-
ferent feelings, such as taste stimuli by pleasantness and un-
pleasantness, color by excitement or repose. Also high and
low tones were likely to produce one of the last named pair.
This author believes that in the light of his knowledge of the
feelings which certain stimuli arouse and their physiological
accompaniment he is able to find many traces of feeling in the
works of other investigators.

He finds that FeVe" * asserts that all feelings of depression
show a diminution in the volume of the extremities while all feel-
ings of excitement and pleasure show a volume increase. Strain
and relaxation are also mentioned. In regard to Mentz, J he
says : ' We need only to follow the meaning of sthenic and
asthenic and of voluntary and involuntary attention without
prejudice and take the matter in an elementary way in order to
find the three directions of feeling.' Also in Dumas', 3 Binet-
Courtier's, 4 and Binet-Vaschide's 5 works he finds hints of the
tridimensional theory, and in Lehman's6 latest work he believes
that if the author had analyzed more thoroughly and adopted the
tri-dimensional theory many of the results would coincide with
his own. I shall speak of these works later in comparison with
my own results.

1 FeVe', Sensation et mouvement, Paris, 1887.

2 Mentz, op. cit.

8 ' Recherches experimentales sur la joie et la tristesse,' Revue philos.t 1896.

* ' La circulation capillaire dans ses rapports avec respiration et les phenom-
enes psychiques, ' L'Annee psychol., II.

1 Influence de la vie emotionelle sur le coeur, la respiration et la circulation
capillaire,' L'Annee psychol., III.

5 ' Influence du travail intellectuel, des emotions, du travail physique §nr la
pression du sang,' L'Annee psychol., III.

• Op. cit.

230 L. PEARL BOGGS.

STRAIN AND RELAXATION.

Strain and relaxation are the feeling side of attention. The
feeling seems to be strongest when the subject is attending to
stimuli about 8 seconds apart, the intervening time between the
signals being free from any sensations or ideas, and all con-
sciousness being intent on the recurring signal. However at-
tention is sometimes accompanied by what the subject frequently
designates as concentration. Here the feelings of strain and
relaxation are not so pronounced, although my own introspec-
tion and that of others find them present. This is the case
where one attends to a continuous succession of stimuli or ideas
without an appreciable interval between. It is interesting to
note that the physiological accompaniment of the two are some-
what similar. We find voluntary attention is accompanied by
the purest forms of strain and relaxation, involuntary attention
being often accompanied by excitement. In the following ex-
periments we are to think of attention as being the former.

Experiment I. — P. P. 7/i6/'c>3. This subject has on this
occasion a marked respiratory rythm in the sphygmographic
curve of about u pulses and his breathing was unusually slow
and deep during a state of rest. The stimulus was the clicks
of a double contact key which by means of the magnetic time
marker were recorded on the smoked paper. The breathing at
once became more shallow and about twice as rapid while the
respiratory rhythm is much less pronounced, occurring every
four or five pulses. The pulse is somewhat lower with a lower
dicrotic during the period of strain while the pulse is higher with
a higher dicrotic during the period of relaxation. The pulse
rate is more rapid during strain than relaxation.

Normal. Strain. Relaxation. Strain. Relaxation. Strain.
No. of pulses, 4-5 4-4 4-4 4~4 4~5 5~5

Av. length, 5.2 5.1 5.1 4.8 5.1 5.1 4.8 4.8 5.5 5.1 4.8 4.9

At this point there seems to be a change. We have at re-
laxation 4.9 and 4.6 followed by 5.3, 5.1, 5.

Experiment 2. — P. B. 8/5/'o3. Stimulus was clicks as
before 5 seconds apart. The breathing was shallow and rapid
while the respiratory oscillation almost entirely disappeared.
Dicrotic changes as before.

AN EXPERIMENTAL STUDY OF FEELING. 3$l

Strain. ReUxation. Strain. Relaxation. Strain.
No. of Pulses, 3-456 56

Av. length, 5.5 5.1 5.3 4.9 5.3 5

Here as in the former experiment a change is apparent. We
find next 4.9 4.9 4.6 4.9, which according to the subjects' intro-
spection is a sort of shifting or adjusting of the attention.

Experiment 3. — P. P. 7/14. The stimulus was again clicks
about five seconds apart and the subject felt strain and re-
laxation.

Strain. Relaxation. Strain. Relaxation.
No. of pulses, 6 6-2 4-4 4-4

Av. length, 4.9 5-35-5 5-14-9 5 5-i

The dicrotic changes are as before.

Experiment 4.. — A. P. 7/20. Stimulus is clicks eight sec-
onds apart. The dicrotic changes are the same, low in strain,
high in relaxation.

Normal. Strain. Relaxation. Strain. Relax. Strain. Relax.
No. of pulses, 5-5 5-5 5-5 5-5 5-5 5-5 5-5

Av. length, 4.7 4.7 4.7 4.7 4.9 4.8 4.8 4-9 4-8 4-8 4-7 4-7 4-8 4-9

This subject shows very little change in pulse length in
any reaction.

Experiment j. — A. P. 7/16. Stimulus was clicks five sec-
onds apart, which came too quickly for this subject, so that she
was somewhat excited all the time.

Experiment 6, following a few minutes after with clicks ten
seconds apart caused a feeling of depression although the strain
and relaxation were also felt.

Experiment 7. — A few minutes after this clicks eight sec-
onds apart seemed to give pure feeling of strain and relaxation,
that is, the taps came just when expected, neither too soon nor
too late.

Experiment 5.

Normal. Strain. Relaxation. Strain. Relax. Strain. Relax.

No. of pulses, 5-5 767676

Av. length, 4.5 4.5 4.5 4.3 4.5 4-5 4-5 4-5
Experiment 6.

No. of pulses, 5-5 6-6 6-7 6-7 6-7 6-7 6-7

Av. length, 4.5 4.5 4.5 4.5 4.5 4.4 4.4 4-4 4-4 4-4 4-3 4-3 4-3 4-3
Experiment 7.

No. of pulses, 5-5 5-5 5-5 5~5 5~5 5~5

Av. length, 4.3 4.3 4.4 4.6 4.4 4.4 4-4 4-6 4-4 4-4 4-4 4-3

232 L. PEARL BOGGS.

In all three of these the curve shows the dicrotic changes,
the lack of variation in length is no doubt due to excitement in
the first curve and depression in the second. The shortened
pulse during the first period of relaxation of first curve is due to
unpleasantness according to the subject's introspection. In
the second curve we find the greater frequency of the pulse
which so often accompanies prolonged attention.

In the third curve there is already present a low dicrotic and
during the first period there is a stronger pulse with now a high,
now a low dicrotic. The next shows a uniformly high dicrotic,
while the last shows a lower dicrotic. The whole curve is very
low and becomes almost threadlike, showing, I believe, the
effects of fatigue.

Experiments. — P. B. 7/5 /'o3. Stimulus: clicks eight
seconds apart.

Normal. Strain. Relaxation. Strain. Relaxation. Strain.

No. of pulses, 5 5-5 4-5 5-5 5~5 5~5

Av. length, 5 4.4 4.8 5 4.8 5 4.9 5 5.1 5 4-9

The breathing is rapid and shallow and the dicrotic during
the first period of strain is low, increasing gradually in height
until the middle of the next period, but for the remainder the
dicrotic changes are not so marked. My introspection here
was that the taps were too weak and far apart and so I became
confused without any feeling of strain and relaxation.

In a word, we find in agreement with Brahn that strain is a
feeling whose physiological accompaniment is a quickened
pulse with a lower dicrotic while its opposite feeling, relaxation,
is accompanied by a slower pulse with higher dicrotic. In re-
gard to the breathing, of which Brahn does not take account,
it is more rapid, regular, and shallow than in a state of rest.
The respiratory rhythm disappears to a very large extent. This
feeling, like every other feeling cannot exist in strength for any
great length of time but gives way to a sort of confused state
caused by the shifting of attention (Ex. i, 2, 8). The physio-
logical processes also lose their pronounced characteristics dur-
ing this period. I find a difference in subjects in regard to the
time of the appearance of strain and relaxation as it depends on
the quickness of the subject in responding to stimuli.

AN EXPERIMENTAL STUDY OF PEELING.

There are a number of experiments which were undertaken
in order to ascertain what effect the state of attention during
psychical activity had upon the vaso-motor system. My own
introspection and that of my subjects give a feeling of concen-
tration as the accompaniment of attentive, psychic action, which
is largely the feeling of strain, flash-like period of relaxation
preventing its becoming too strong. Between the curves of at-
tention with the feeling of concentration and strain and relax-
ation there is a great resemblance.

Experiments p. — S. C. 5/127*03. The subject looked at a
figure of tumbling blocks. Each time the figure took a different
form the subject pressed the electric contact key and recorded the
fluctuation on the drum. The breathing was shallow, rapid
and regular ; the apex of the breathing curve usually coincided
with the time of fluctuation. Sometimes the fluctuation did not
occur at every breath but at the second breath. Normal aver-
age length of pulse is 5 mm.

Fluctuations, i 2 34-56 7 8-9 10 n

No. of pulses, 555566647

Av. length, 4.3 4.4 4-i 4-5 4-2 4-3 4-6 4-6 4-5

Fluctuations, 12 13 14-15 16 17-18 19 20-21 22

No. of pulses, 75 55 55 5

Av. length, 4.5 4-i 4-2 4-3 4-3 4-4 4 4-7

The dicrotic changes correspond roughly to changes in
length which occur with the fluctuations. It is lower on the
whole than in the normal.

Experiment 10. — S. C. The subject looked at a very
simple figure this time and the fluctuations were more regular,
corresponding as before to the breathing apex. The concentra-
tion was not so great and the breathing is correspondingly slower,
especially at first. The dicrotic is high during the first twenty
pulses, after which it becomes lower and fluctuates with varia-
tions in length of the pulse curves.

Fluctuations,

1-2

3-4

5-6

7-8

9-10

11-12

No. of pulses,

4

4

4

4

4

4

Av. length,

4.8

5

4.8

4-7

4-7

4.6

Fluctuations,

13-14

15-16

17-18

19-20

21-22

No. of pulses,

4

6

8

7

7

M

Av. length,

4.6

4.8

4-6

4.8

4-6

234 L. PEARL BOGGS.

Fluctuations, 23-24 25-26 27-28 29-30 3*~32

No. of pulses, 88678

Av. length, 4.5 4.4 4.4 4.3 4.4

The average length of these pulses is greater than in the
preceding.

Experiment //. — E. C. 5/22. The subject attended this
time to the counting of a clock's ticking. He said he was
obliged to concentrate his mind on it. The breathing is shallow
and rapid. The respiratory rhythm tends to disappear although
it is usually quite pronounced and is moreover a three pulse
rythm. His normal pulse is regular as to height but irregular
as to length, its average perhaps being 6.6.

No. of pulses, 3 throughout.

Av. length, 6.5 6.3 6.3 6.5 6.1 6 5.9 6.3 6 6.3

The dicrotic is lower excepting at 6.5, 6.1, with a slight rise at
the last 6.3.

Experiment 12. — E. C. 5/22. After a short pause the sub-
ject listens to a music-box medley to which he attends without
much effort and some pleasure. The breathing is deeper and
slower and the rhythm in pulse more pronounced.

No. of pulses, 3 throughout

Av. length, 5 5 5 6.3 6 6.8 6.8 6.4 6 6.6 6,

Before. Stimulus.

During the four periods before the stimulus and the first
after, the dicrotic is low. After that it is higher but fluctuates
somewhat. The average length of pulse is greater than during
the preceding experiment. The subject appeared to be on a
strain before the stimulus began.

Experiment ij. — A. P. 5/2i/'o3. The subject is reading
an interesting essay. Breathing characterized by long pause
after expiration, fairly regular, and of medium depth.

No. of pulses, 5 throughout.

Av. length, 4.5 4.3 4.3 4.2 4-2 4.1 4-2 4-3 4-3 4-2

Before. Reading.

4.4 4.1 4.4 4.3 4.1 4-i

Reading continued.

The dicrotic is lower generally but shows variations correspond-
ing to the length of the pulses.

AN EXPERIMENTAL STUDY OF FEELING. 235

Experiment 14. — A few minutes afterwards another tracing
is taken, the subject having read on without interruption. The
breathing is more rapid and shallow. In both the respiratory
rhythm almost disappears. The dicrotic changes are as before.

No. of pulses, 5 throughout.

Av. length, 4.3 4.3 4.2 4.2 4.4 4.3 4.4 4.2 4.3 4.2
Av. length, 4.5 4.4 4.8 4.6 4.6

At this point a violet odor was given to the subject. It was
pleasant but subject could not concentrate her mind on her read-
ing again. The pulse shows a lengthening.

Experiment 15. — S. C. 5/12. Subject is reading an inter-
esting selection.

No. of pulses, 5 throughout.

Av. length, 5 4.8 4,8 5.2 5

The dicrotic is rather high throughout, a little higher at 5.2
than elsewhere.

A little later another tracing is taken while the reading is
still going on. The pulse is more rapid as is also the breathing
while the dicrotic is lower.

No. of pulses, 5 throughout.
Av. length, 4.3 4.4 5 5 4.8 5 4.7 4.7 5

Several times I have mentioned the disappearance of the
respiratory rhythm in the pulse during attention. This was very
marked with two subjects whose respiratory rhythm was usually
so pronounced that the tracings were once discarded so far as
the question of emotional accompaniment was concerned. How-
ever with these subjects any sort of rhythmic action, bodily or
psychical, tended to destroy the respiratory rhythm in a remark-
able degree.

Experiment 16. — C. W. 4/24. The subject was a very
athletic young woman, generally not responding emotionally to
stimuli. A slow in and out movement of the free arm caused a
greater rapidity of pulse with a little increase in height, the
rhythm almost disappears, while the dicrotic is high in the out-
ward and low in the inward movement curve.

Normal. Out. In. Out la. Oat. In.
No. of pulses, II 12 II 13 14 10 l8

Av. length, 4.3 3.6 3.5 3.1 3-5 3-3 3-5

236 L. PEARL SOGGS.

Another experiment with the same subject in which she opened
and closed the hand resulted in practically the same tracing.

Experiment ij. — E. H. Another subject showed still
greater variation of pulse within an oscillation, sometimes the
highest and longest pulse being almost twice the dimensions of
the smallest. The arm movement caused the rhythm to disap-
pear to a great extent as did a ticking metronome which ' both-
ered ' the subject. During a Beethoven selection on an organ
which the subject pronounced ' sleepy,' ' monotonous,' the
breathing is rapid and shallow with great regularity of the pulse.

Before. Music. After.

No. of pulses, 10-10 10 10 10 10 10 10

Av. length, 4 3.8 3.6 3.5 3.6 3.5 3.5 4-4

The dicrotic was somewhat lower during reaction.

The most marked and regular respiratory rhythm was ob-
served in the tracing of a subject who peacefully slept for twenty
minutes while records were being taken. It was impossible to
arouse the subject sufficiently to take part in the experiment
with attention and in the tracings taken during a waking condi-
tion the rhythm persists strongly.

The results show a striking resemblance to the results of the
previous series of experiments. Rapid, regular, shallow breath-
ing, a more rapid pulse on the whole and dicrotic changes in
the pulse corresponding to variations in length. Also the
rhythm parallel with respiratory movements tends to disappear.

The results agree with those of most other experimenters as
regards the greater frequency during attention, Binet and Cour-
tier1 having also called attention to the fact that the dicrotic is
later and lower during psychical activity.

Meuman and Zoneff, however, find a slowing of the pulse
during both sense and intellectual attention. I should be in-
clined to attribute this result to the choice of stimuli used.
Nearly all require precise muscular accommodation of the sense
organ. For example, a card containing five points was brought
near until the subject was able to count them ; another time a
ticking watch is brought within hearing ; again it is to determine
the threshold for the sense of touch according to Weber. All

1 ' Circulation Capillaire,' L'Annte fsychol., Vol. III.

AN EXPERIMENTAL STUDY OF FEELING. *yj

of these require or at least call forth an involuntary holding of
the breath in order to further the sense accommodation.

The same is true in the case of the fluctuations of attention.
I found myself * holding my breath ' when I tried to keep a wire
moving along a certain line. A criticism might also be made
upon the choice of stimuli in intellectual attention. Problems
in mental arithmetic are likely to arouse excitement, confusion,
pleasantness, or unpleasantness, according to the ease and suc-
cess with which they are solved. Then again their conclusions
do not seem to be justified by the results. In three out of ten
experiments the pulse is more rapid. In five the pulse is more
rapid during the first ten seconds and in another equal. In one
experiment the rate is only given for the whole reaction so that
there are only three cases where the pulse is first slower and in
one of these the second ten minutes show a greater frequency.
On the whole their results are not so much at variance with
those of others as it seemed at first glance that they might be.

PLEASANTNESS AND UNPLEASANTNESS.

In regard to pleasantness and unpleasantness, little new has
been found regarding their physiological changes in the vaso-
motor system. Pleasure is almost always accompanied by a
slower pulse and unpleasantness by a quicker pulse, the height
usually increasing in the former and decreasing in the latter.
The following are a few examples :

Experiment 18. — A. P. 4/9/'o3- Subject was amused
at a high organ note.

No. of pulses, 555 curve slightly.
Av. length, 3.8 4.4 4 higher at 4.4.

Experiment 19. — S. C. 5/4/'o3- White rose odor was

pleasant.

No. of pulses, 555
Av. length, 4.5 5 4-8

Experiment 20. — L. B. 5/12. Clove oil was pleasant.

No. of pulses, 5555

Av. length, 4.2 4-2 4-4 4-4 4-2.

238 L. PEARL BOGGS.

Experiment 21. — S. C. 5/i2/'o3. A purple red colored
glass held before the subject in front of a non-transparent win-
dow was pleasing.

No. of pulses, 555

Av. length, 4.8 5.3 5

Higher.

Experiment 22. — A. P. 5/7/'o3. Turpentine odor called
up a number of pleasant associations.

No. of pulses, 55555
Av. length, 3.5 3.8 3.8 3.6 5.5

A very interesting fact about this subject is, that out of a
large number of tracings taken when the subject gave the intro-
spection of pleasant after an odor the pulse was seldom slower.
By chance I discovered that the subject had a habitual dislike
of all kinds of odors, which was stronger than the momentary
pleasantness. This may account for the regular pulse varia-
tion being absent during these reactions. The feeling of un-
pleasantness is one which is comparatively easy to bring about
but I shall give only a few of the reactions.

Experiment 23. — S. C. 4/2O/'o3. The stimulus was a dis-
agreeable odor.

Reaction.

No. of pulses, 6 6 6666 6

Av. length, 3.5 3.5 3.2 3.1 3.1 3.2 3.6

I<ower.

Experiment 24.. — S. C. 5/4/03. The stimulus was a blue
glass plate twice exposed but the subject was unable to focus it
and hence had an unpleasant feeling.

Reaction.

No. of pulses, 66666666
Av. length, 4.6 4.7 4-5 4-2 4- a 4-5 4-2 4-2.

Occasionally a very disagreeable feeling caused a slowing
of the pulse.

Experiment 25. — S. C. 4/i3/"o3. The stimulus was a
painful pressure.

Reaction.

No. of pulses, 64666
Av. length, 4.8 5-2 5-4 4 4-3

Higher.

AN EXPERIMENTAL STUDY OF PEELING. 239

Experiment 26. — A. P. S/28/'o3. The stimulus wa at
few drops of lemon juice which was very disagreeable —
« terrible.'

No. of pulses, 555
Av. length, 4.9 4.4 4

Lower.

Experiment 27. — On another occasion the stimulus being
again lemon juice the reaction was much the same.

No. of pulses, 5555555
Av. length, 4.5 4.5 3.9 3.6 3.8 4.2 3.7

Lower. Higher.

In cases where the feeling of pleasantness or unpleasantness
is mingled with other feelings the pulse changes show varia-
tions from the above as we shall see later.

EXCITEMENT AND REPOSE.

Whether one is to consider excitement and repose as feeling
is a matter of most accurate definition and careful introspection.
But taking feeling in the sense of a reaction to sensations and
ideas, we can perhaps call that consciousness of increased mus-
cular or psychical power with a tendency to action a feeling to
be designated by the terms excitement, or liveliness. The con-
sciousness of weakened or inhibited psychical or physical force
we may designate as a feeling of repose or depression. Either
of these is less often found alone than other feelings and it is
especially difficult to bring them about in the laboratory, Brahn
having succeeded in finding five cases of pure excitement in
two hundred curves. The stimuli best adapted for causing the
feelings are bright-colored transparent plates, high or low tones,
and the ticking of a metronome. Also certain odors bring
about this feeling but it is more likely to be mixed with some
other feeling. As I have said before, I measured the height of
the curves only in terms of higher and lower. The following
are examples of excitement.

Experiment 28. — L. B. 5/5. High chord of organ.
Excitement.

No. of pulses, 55555
Av. length, 4.6 4-6 4-6 4-4 4-6

240 L. PEARL BOGGS.

Experiment 29. — L. B. 5/12. The stimulus was rasping
files which first excited and then amused me.

No. of pulses, 55555
Av. length, 4.4 4-4 4-4 4-5 4-2

Experiment 30. — L. B. 5/12. A shrill Galton whistle set
my nerves on edge but I was obliged to smile to relieve the
tension. I was on a strain in anticipation of what I suspected
was coming.

No. of pulses, 5 5

Av. length, 3.7 3.8

Experiment 3 1 . — L. B. 5/18. Another day a high whistle
excited, ' keyed me up.' I afterwards felt a thrill of pain
behind the ear.

No. of pulses,
Av. length,

Higher.

Experiment 32. — L. B. 5/18. Ammonia caused a strong
state of excitement.

No. of pulses, 55 5 55

Av. length, 4.7 4.8 4.7 4.7 4.7

Experiment 33. — L, B. 5/26. A bright yellow glass
plate held between my eyes and the window caused a slight
feeling of cheerfulness.

No. of pulses,
Av. length,

Experiment 34.. — A. P. 4/16. A discord on the organ
proved exciting.

No. of pulses, 5555
Av. length, 3.9 3.9 3.9 3.9

Experiment 35. — A. P. 4/16. A red plate of glass
caused a strong excitement ; subject felt she must jump up from
the chair ; cold chills run up and down her spine. There was
no doubt some unpleasantness present at the close of the
experiment.

AN EXPERIMENTAL STUDY OP FEELING. 241

No. of pulses, 55555
AT. length, 4.4 4.2 4.2 4 4.4

Experiment j6. — A. P. 5/28. The same color proved
* something fierce ' and subject felt cold chills after it was taken
away.

No. of pulses, 55555555
Av. length, 4.4 4.6 4-4 4-7 4-7 4-4 4-6 4.8

Experiment 3 j. — A. P. 5/14. Red and yellow glass
placed side by side was said to be « awful, distracting.'

No. of pulses, 55555
Av. length, 4 3.9 3.9 3.8 3.8

Higher.

Our conclusion then is that excitement is accompanied by a
normal rate of the pulse but that the pulse is often stronger.
In the cases where higher is not written there was no appreci-
able change in height. A number of curves were also re-
corded during a state of unpleasant excitement and pleasant
excitement, while there are also cases of strain where excitement
is present. The most striking examples of the last named
combination are the reactions which took place during the tick-
ing of a metronome which varied in rapidity from 48 to 90
strokes per minute. The following are a few examples :

Experiment 38. — L. B. 4/28/'o3. The stimulus was a
quick metronome and the reaction was a feeling of strained ex-
citement. My usual expression was that it ' keyed me up.'

No. of pulses, 5 throughout.

Av. length, 4.6 4.4 4.8 4.5 4.5 4-5 4-7 4-7 4-7 4-6

The dicrotic was lower during the first three periods and
almost disappeared they were so high during the last two.

Experiment 39. — L. B. 5/26/'c>3. The stimulus was a
very quick metronome while the reaction was a ' keyed up feel-
ing.' I seemed to stop breathing and my heart seemed to beat
faster.

No. of pulses, 5 throughout.

Av. length, 4.8 4.8 5 5-3 5-3 4-8 4-8 4-8 4-9

242 L. PEARL BOGGS.

The height of the pulse and the dicrotic were both greater
during the first two periods.

Experiment 40. — A. P.1 4/9/'o3- The stimulus was a
slow metronome and the subject felt very excited, ' it was a
torture to sit still.' The subject was expecting a stimulus of this
sort.

Reaction.
No. of pulses, 5 throughout.

Av. length, 4.1 4 4 4.1 4 3-9 3-9 3-9 3-9 4

The curve was higher with a higher dicrotic during the
third period before the stimulus, then both apex and dicrotic
become lower during the next two periods but become higher
very gradually after the stimulus until the fourth period when
they become lower and vary slightly till the end when the
dicrotic becomes higher.

Experiment 41. — A. P. 4/16. The same stimulus as the
above caused practically the same reaction and same physio-
logical accompaniment.

No. of pulses, 5 throughout. Reaction.

Av. length, 4.1 4.1 4.1 3-9 4 3-8 3-7 3-7 3-7 3-8

Experiment 4.2. — S. C. 4/20. The stimulus was a met-
ronome at the rate of 48 strokes per minute. The subject
declared it to be exasperating and unpleasant.

No. of pulses, 6 throughout.
Av. length, 4.4 4-3 4-4 3-9 3-7 4 3-8 3-8 3-9

The pulse curve is at first high with high dicrotic. During
the next two periods both become lower while at the fourth
period the curve is very irregular, showing on the whole a
higher dicrotic. It gradually becomes lower until the end of
the stimulus when it tends to return to the normal height.

Other examples might be given but as we are concerned
more with simple than complex phenomena these must suffice
to show that a mixed feeling of strain and excitement has physi-
ological accompaniments differing from those of either of the
pure states. A few examples will also suffice for the other
mixed states.

AN EXPERIMENTAL STUDY OF PEELING. 143

Experiment 43. — A. P. $/2i/lO3. A crash on the organ
was surprising and unpleasant, distracting.

No. of pulses, 5 throughout

Av. length, 4.5 4.3 4.6 4.8 No variation in height.

Experiment 4.4.. — A. P. 5/28/'c»3. A strange odor caused
a strong feeling of surprise and unpleasantness.

No. of pulses, 5 throughout.

Av. length, 4.8 4.8 4.6 4.4 4.5

Shallower.

Experiment 4.5. — S. C. 5/i2/*O3. A high whistle was
irritating.

No. of pulses, 6 6

Av. length, 4.8 4.8 No change in height

Thus there seems to be no consistency in these results, certain
physiological effects predominating at one time, at another time
others, probably according to the difference in the composition
of the feeling.

Experiment 46. — L. B. 4/28/03 . A high tone caused a
feeling of excitement but was at the same time amusing.

No. of pulses, 555 5

Av. length, 5 5-2 5-3 5-3

Experiment 47. — S. C. 4/20/03. Anise oil odor was
pleasant and exciting.

No. of pulses, 66 66 666

Av. length, 3.7 3.6 3.8 3.9 3.7 4 3-8

A trifle higher.

Experiment 48 — S. C. 5/18/03 . Menthol proved a pleas-
ant and stimulating odor.

No. of pulses, 66 666

Av. length, 4.5 4-5 4-6 4-4 4-4

Experiment 49. — A. P. 5/14/03. A rich red plate of
glass gave the subject a pleasant exhilarated feeling, « toned

her up.'

No. of pulses, 5 55

Av. length, 4.2 4.2 4

^v~>

Higher.

244 L. PEARL BOGGS.

The results here are more consistent because excitement
and pleasantness have the heightened curve in common as a
physiological accompaniment.

Repose or depression as a pure feeling occurred a few times
and was usually not accompanied by any marked changes in
the pulse either in the height or frequency. The following are
examples.

Experiment 50. — L. B. 5/127*03. A glass plate sage
green in color was quieting.

No. of pulses, 55 55

Av. length, 5 4.9 4.9 4.8

Experiment 51. — L. B. 5/267*03. Bergamot odor was
heavy and depressing.

No. of pulses, 555555
Av. length, 4.8 4.8 4.9 4.6 4.6 4.8

The after effect of this odor was unpleasant.

Experiment 52. — A. P. 5/147*03. Sage green was de-
pressing.

No. of pulses, 5555
Av. length, 4.1 4.1 4.2 4.2

Experiments 53, 54., 55. — A. P. 5/7/'o3« Low organ
tones produced in[these three cases a solemn quiet feeling.

No. of pulses, 5 throughout

Reaction. Reaction. Reaction.

Av. length, 3.6 3.5 3.7 3.7 3.7 3.6 3.6 3.8 37 3.7 3.6 3.5 3.7 3.6

Experiment 56. — 4/30. A light blue glass plate caused a
feeling of depression.

No. of pulses, 5 5555

Av. length, 3.9 3.8 4-3 3-9, 3-9

Experiment 5J. — L. B. 5/5*03. A low organ tone was
felt as earnest, rich, a somewhat peculiar feeling but fitting in
here because it seemed to induce repose.

No. of pulses, 555 5

Av. length, 5 4.6 4.8 4.7

AN EXPERIMENTAL STUDY OF FEELING. 245

Experiment 58. — This was followed by the odor of castor
oil which was depressing.

No. of pulses, 5555
Av. length, 4.7 4.6 5 4.4

In all of the above cases there is very little variation in the
length of the pulse curves and little or none in the height.
There is not so far as I can see the uniformly lowered pulse
which Brahn finds, however there is certainly not an increase in
height in any of them. The failure to find the lowered pulse
may be due to the fact that the cases are not those of pure
repose, though this seems very unlikely. I am much more
inclined to think that the feeling of excitement and repose is
far more often accompanied by muscular than circulatory
changes. A number of experiments with the automatograph and
ergograph which I shall not describe here, as well as the intro-
spection of my subjects led me to this supposition. In nearly
all cases of pronounced excitement the subject speaks of a desire
to move, while in the repose the quiet, restful effect is spoken of.
Everyday observation teaches us that certain music has the
effect of stimulating to action while other melodies have a sooth-
ing, sleepy effect, etc. and one is very averse to movement.

Just as in the case of excitement so here in repose we find
mixed feelings ; repose and pleasantness and repose and un-
pleasantness. The following are examples :

Experiment 59. — A. P. 4/i6/'o3. A low tone produced
a feeling of unpleasant sadness.

No. of pulses, 555555
Av. length, 3.8 4 3.8 4 3-8 4

Experiment 60. — L. B. 4/28/'c>3. A low organ tone was
pleasant and quieting.

No. of pulses, 555
Av. length, 4.8 4.8 4.8

Experiment 61. — A. P. 5/2i/'c>3. A low organ tone

was pleasant and rich..

No. of pulses, 555
Av. length, 4.4 4-4 4-4

246 L. PEARL BOGGS.

Experiment 62. — S. C. 4/5 /'o3» Light blue glass was
quiet and pleasant.

No of pulses, 55555
Av. length, 3.5 3.4 3.7 3.6 3.6

Experiment 63. — S. C. 5/4/'c>3. The subject described
the reaction from the sight of a rich purple-colored glass as be-
ing very satisfying, rich, and the most esthetic of all feelings,
t. e.y not organic.

No. of pulses, 55555
Av. length, 4 4-2 3-9 3-9 4-i

In the above we find comparatively small changes though all
except the last show no increase in the frequency of the pulse.

Experiment 64.. — L. B. A very unpleasant depressed
feeling was caused by inhaling a sickening odor of onion es-
sence three times in succession. The curve is very much lower
in height than it usually is and at the same is more rapid. It
became stronger just at the time of the second and third appli-
cations.

No. of pulses, 5555555555 con.
Av. length. 4.7 4.8 4.7 4.7 4.6 4.6 4.6 4.5 4.5 4.4 4.4

Very Weak. Strong. Weaker.

No. of pulse (con.), 555
Av. length, 4.8 5.1 4.8

Strong. Weaker.

Experiment 65. — L. B. 5/26/*O3. The odor of berga-
mot was very depressing and unpleasant, the latter effect fol-
lowing the former.

No. of pulses, 5 555 5 55

Av. length, 4.8 4.8 4.8 4.9 4.6 4.6 4.8

IvOwer.

Experiment 66. 4/2O/'o3. The stimulus was an odor which
produced a depressed disagreeable feeling.

No. of pulses, 6 throughout.

Av. length, 3.5 ,3-5 3-2 3.1 3-1 3-2. 3-6

I,ower.

Experiment 67. — S. C, 6/i/'o^. Castor oil odor caused

AN EXPERIMENTAL STUD Y OF PEELING. 247

the subject to feel very depressed 'brain congealed, life not
worth living,' followed by unpleasantness.

No. of pulses, 6 throughout.

Av. length, 4 4 3-9 3-8 3-6 3-9 3-7 4«i

Lower.

The feeling of unpleasantness-repose causes in all cases a
decided decrease in the height of the pulse with a somewhat
greater frequency, and it is clearly noticeable that the feeling
of unpleasantness follows that of repose here as pleasantness and
unpleasantness have been seen elsewhere to follow that of ex-
citement.

SUMMARY.

1. The physiological changes in the vaso-motor system ac-
companying the states of strain and relaxation are as follows.
Strain is accompanied by a decrease in the length of the pulse
curve, while the dicrotic becomes lower. Relaxation is accom-
panied by an increase in the pulse length and a heightened
dicrotic.

2. The state of attention is accompanied by a feeling of
concentration partaking of the nature of strain and has parallel
physiological process closely resembling those of strain and re-
laxation but with less marked and regular characteristics.

3. The breathing in both the feeling of strain and relaxation
is more rapid, regular and shallow than in an indifferent state.
This is also true of attention accompanied by a feeling of con-
centration.

4. Pleasantness is accompanied by a longer and higher pulse
curve ; unpleasantness is accompanied by the reverse.

5. Excitement and repose are accompanied by less marked
changes in the vaso-motor system than the other feelings. There
is practically no change in the pulse frequency and while the
former is accompanied usually by a heightened pulse, the latter
is accompanied less frequently by a lower pulse according to
my results, but certainly never by a higher.

6. The excitement-strain feeling shows characteristics of
the curves of both, while the excitement-pleasure is pretty con-
stant in showing a lengthened and heightened curve. The

M8 L. PEARL BOGGS.

curves for excitement-unpleasantness have no constant charac-
teristics. Repose-pleasantness is accompanied by no constant
marked characteristics but unpleasantness-repose is accompanied
usually by a shorter and decidedly lower pulse curve.

7. Where the feelings are mixed, pleasantness and unpleas-
antness do not appear until after those of excitement and repose.
The time of the appearance of feeling of strain depends on the
time it takes for the concentration of the attention to the stimu-
lus. Sometimes it appears almost at once and again two or
three pulses later. When it appears in mixed feeling it comes
later than the others.

8. The rhythm in the pulse lengths which correspond roughly
to the act of breathing, i. <?., an inspiration and an expiration,
tends to disappear during states of attention and the more so the
stronger the feeling of strain and relaxation or of concentration. l

1 The MSS. of this article was received on February 8, 1904.— ED.

THE PSYCHOLOGY OF AESTHETIC REACTION TO
RECTANGULAR FORMS.

BY THOMAS H. HAINES.'AND ARTHUR ERNEST DAVIES.
Ohio State University .

Since the aesthetic value of the golden section was first
developed by Zeising, l all questions in the psychology of the
aesthetics of form have centered around this one, the psychology
of the golden section. And to-day it is as much an open ques-
tion, why we like the golden section, as it was when Fechner
wrote, 2 " If you ask me, I simply say, I do not know." Fech-
ner's researches showed that the golden section was not so nearly
the universal choice as had been claimed. His measurements
of many thousand paintings, and the results of choices, amongst
ten cards of different proportions, made by several hundred men
and women, show, however, that the preponderance of choices
is near the golden section. Witmer's3 investigations emphasize
the importance of the relation in both the rectangle and the
ellipse, and also in the division of lines, and in relations of
separate lines to one another. Angier's 4 results likewise indi-
cate its importance in the division of the horizontal line. But
the evidence is not sufficient for suggesting an exact law ; it is
not as strong, e. g., as it was in favor of Weber's law. In fact
Fechner says that his * results for every class of pictures show
that the relation of the larger to the smaller dimension is on the
average much less than that of the golden section.' (' Welche
Bilderklasse man ins Auge fassen mag, das Verhaltniss der
grosseren zur kleineren Dimension durchschnittlich viel kleiner
als das des goldenen Schnitter ist') (*. e., to make the golden
section, the large ought to be larger, or the small, smaller).

1 jEsthctische Forschungen, 1855.

2 Vorschule der Aesthetik, 1876.

3 ' Der Experimentellen Aesthctik einfacher raumlicher Formvcrhaltnisae,'
Philosophische Studien, 1894.

4 Harvard Psychological Studies, Vol. i, pp. 541 ff.

249

250 T. H. HAINES AND A. E. DA VIES.

So much for the facts. On the side of explanation we meet
a much greater variety. It is therefore the least satisfactory
aspect of the problem, and the one that calls most loudly for
serious constructive work.

1. To pass proposed explanations of this class of phenomena
briefly in review, we may mention, first the theory which pro-
poses the mathematical relations of the lines, as the ground of
preference. The rectangle whose sides are in the relation of
extreme and mean ratio, is pleasing because of the simplicity
of the process by which the variety here offered is grasped as a
unity. The smaller being in the same relation to the larger as the
larger is to the sum of the two, gives us the simplest possible
concept of an organism. It presents a pleasing variety, which
of itself, by virtue of the simplicity of its mathematical rela-
tions, leads the mind to the idea of unity. Hence Zeising saw
in the golden section, 'the complete embodiment of an ultimate
aesthetic principle — the combination of a complete diversity in
an harmonious unity.' This theory is closely analogous to the
one which explains musical harmony on the basis of an uncon-
scious counting of vibrations. Both are open to the same
objection. Psychology knows nothing of unconscious cere-
bration. But the mathematical explanation of the golden
section, fails for another reason ; considerable deviations from
the golden section, are as pleasant as it itself is, to most people.
Thus there is nothing analogous to the dissonance arising from
beats, in tone.

2. Various attempts have been made at a physiological ex-
planation of aesthetic problems. First suggested by Schiller,1
it has been developed by Spencer,2 Bain,3 and Allen.4 In gen-
eral it holds that the aesthetic gratification in any form of
activity, depends upon the condition of the whole nervous sys-
tem, or at least of all parts more immediately concerned in the
reaction. Surplus energy, consequently, is the essential phys-
iological condition of an aesthetic reaction. But a discharge
of the surplus energy of a given organ, e. g., of the retina is,

1 Letters upon the ^sthetical Education of Man, Letter 27.

2 Principles of Psy., Part IX., ch. 9.

3 Emotions and the Will, 1859. Mental and Moral Sciences, 1868.
* Physiological ^Esthetics, 1877.

PS YCHOLOG Y OF -4?S TUB TIC HE A C TION. 3$ »

in itself, insufficient to bring about an aesthetic feeling. The
arousal of an aesthetic feeling connected with the visual proc-
esses, depends upon the relative amounts of stored energy or
freedom to discharge, throughout the whole visual apparatus.
It involves, at the minimum, retinal, eye-muscle and cortical
processes. The main defect of this theory is, that it does not
loan itself to concrete application. It is too general to be of
service in actual explanation. It cannot be applied with profit
to the explanation of pleasing rectangular forms.

3. A far more serviceable theory is that which may be called
4 explanation through association.' Its adherents say that we
like a given form because of its resemblance to forms to which
we are already accustomed. Books, cards, envelopes, panels,
windows, etc., are made in shapes which are determined by
the purposes they are intended to serve. There are pretty
definite upper and lower limits, beyond which these forms do
not proceed. There is also a proportion of the sides, which is
more convenient in these objects, and which is embodied in a
very large number of them. We get accustomed to these
forms ; a facility in reacting to them is established ; and the
majority of people react most favorably to the proportions pre-
dominating in the environment in which they live. We like
therefore what we have seen most frequently, because of the
facility of its apperception. This makes the aesthetically pleas-
ing the predominantly useful. Its use determines its preva-
lence; and this in turn, determines the facility of its apper-
ception, and this again determines its power to engender aesthetic
feeling. The possibilities are that the reaction will be only
slight if it is agreeable, but if it be a disagreeable or unpleasant
reaction, it will be much more positive and intense. The reason
for this is that the nerve paths are less obstructed, and so lead
to a ready apperception.

4. Again it has been proposed as a specific explanation of
the pleasantness or aesthetic quality of a proportion, or asym-
metrical division or relation, that it is in reality a symmetrical
division or relation. In order to take pleasure in an asym-
metrical division of a horizontal line, we have to supply mental
material to the shorter end, in order that it shall come up in

252 T. If. HAINES AND A. E. DAVIES.

weight, so to speak, to the longer. The weight is a weight of
attention. From a meaning read into the shorter line, — say a
purpose that it is subserving, — it is occupying an equal share of
the attention with the longer line, and the two parts do really
balance for the onlooker. It is the same explanation which
sees in so many pictures a large object on one side, balanced
and compensated for, by the direction of the attention of the
onlooker to the other side, by leading lines, or the gaze of a
human subject. The physically unsymmetrical is shown to be
the psychically symmetrical.

5. Another theory accounts for the enjoyment of beautiful
objects, by movements of attention and imagination, coupled
with activity of comprehension and sympathy. Applied to our
own problem, this means, that in lines there is a wealth of
pleasurable association, of a kind that connects them with
movements, which we admire in living things. These move-
ments are those which suggest unconstrained activity. Besides
this ethical motive, the intellectual motive of comprehen-
sion enters also as an explanatory feature of aesthetic pleas-
ures. By comprehension is meant the possibility of bringing
a form, e. g., a line, under a single law throughout, so
that it may be understood. We see very soon what its inten-
tion is. The successive parts have novelty, and yet in the
end realize the expectation aroused. "The mental formula
or conception gained early in our perception of a line, is
found throughout to apply." Another factor is the organic
reaction which the movement and life attributed to the figure,
gently stimulate. Alteration in the breathing and circulation,
rhythmic change in tension of groups of muscles of the limbs,
neck, and trunk, accompany attention to the change in the
imaginary point along the observed figure. Organic factors are
however, only secondary ; their function is to react upon the
complex mental state, giving it ' body.' The object of our
aesthetic enjoyment, is not given in sense, but in a spiritual con-
struction out of material, which is often the very opposite of
aesthetic. Elementary aesthetic pleasure has the same factors
as our highest enjoyments and involves the awakening of intel-
lectual, ethical and religious ideas. If graceful form does not

PSYCHOLOGY OF AESTHETIC REACTION. 253

arise directly out of our ethical and religious nature, it does from
something akin to it, namely, our sympathy with well ordered
action and love of participation in such action.

Each of these theories possesses elements of value, but no
one of them is sufficient. Such an array of contradictory theo-
ries, is surely warrant enough for making a new excursus into
the field of the psychology of the aesthetic reaction. In order to
work within a narrow and yet quite typical field, the present in-
vestigation confines itself to rectangular forms. This is a desir-
able form to work with, because it puts the investigation in line
with what work has been already done in the same field, and
thus provides a means for checking results. But it is an import-
ant figure for another reason. Limitation of material is import-
ant in questions of this kind. Witmer has lost some advantage
that might have followed his method, by a lack of discrimination
among the selected forms studied. Take, for example, the
group in which the rectangle appears. It is put with the tri-
angle and ellipse only because they are all ' enclosed figures.'
It does not appear however, that they are all equally simple
forms. Moreover, whether they are capable of evoking an
equally primitive reaction aesthetically cannot be decided previ-
ous to investigation. We have taken one figure of this group,
thinking it well to keep the investigation as individual as possi-
ble. We have chosen the rectangle in preference to the other
figures, because it is, as an object, simpler than the other two ;
there are fewer distractions for perception in it, there is conse-
quently, less hindrance to the emergence of a purely aesthetic
response on its first presentation. The individuality of our
problem lies in the fact that one of the simple space forms, and
one that is constantly recurrent in experience, has been selected
and submitted to experimental treatment, with an aim to define
its aesthetic suggestibility.

Throughout the investigation, it is a matter of importance
to ascertain what figure is selected out of the series of rectangles
constructed on the same base, but varying in width, up to and
including the square. Is it the rectangle whose given sides are
as the golden section? In any given case, or number of cases,
do the judgments cluster around the golden section, as they do

254 T. H. HAINES AND A. E. DAVIES.

with Fechner's and Witmer's observers? But more important
still, is an insight into the mechanism by which the choice is
made. This inquiry was couducted so as to lead into the field
of explanation.

II.

The methods of Fechner and Witmer were neither of them
satisfactory. Fechner used ten cards, each having a surface
of six hundred and forty square centimeters. Beginning with
the square, the relation of the sides was as expressed by the

-f nlln winrr frsrtirmc • 1 6 6 4 29 3. 34 23 2 6 TMip>cp»

lonowing iracuons . ^ -g-, -j, ^, -2--^, §, T^, y^, ^, •%. i nese
were placed at random on a blank surface, the cards being
white, and the observer made a choice of one, if possible.
Sometimes two or more were chosen as equally satisfactory.
In such a case, 50 per cent., 33^ per cent., etc., was put down
for each one so designated, and thus the individual's choice was
distributed in making up the average. Now an objection to
this method is, that that the rectangle which is pleasing in one
position, often fails to satisfy when turned 90°. Another ob-
jection is that the judgments as to whether or not one likes a
given figure, must be comparative under these circumstances.
Judgments should be made as far as possible, in isolation from
other objects of the same kind. The distraction also incident
to having many shown at the same time, should be avoided,
thus securing uniformity of attention to each.

Witmer met the first objection by placing his cards in serial
order, and all in the same position relative to the same observer
But the comparisons and distractions were still incident in his
method. To avoid all three of these objections, we showed the
cards, one at a time. We placed the white card on the black
paper on the table before which the observer was sitting and
asked him to accept or reject each one in turn by a definite act
of pushing it away, or taking it into his hand. This action
called forth definiteness of choice in many cases which was a
surprise, even to the observer himself. This was perhaps, the
most significant factor in our method, different from the other
two mentioned.

The advantages from this method, over that of Fechner's and
Witmer's, may be set forth in detail :

PSYCHOLOGY OF ESTHETIC REACTION. 355

I. The Advantage of a Clear Objective Field of Vision. —
The accepted and rejected figures are removed to one side and do
not enter into competition with the last presented figure. The
advantage of this is to reduce the element of distraction to a
minimum, and to have attention free to react upon the presented
form. Both of these are absent from the method of both Wit-
mer and Fechner. We do not, however, exclude the compar-
ison of one figure with another in the same group as an aid for
judgment, but this is found necessary only when, for special
reasons, the mind has wandered from the matter under determi-
nation.

2. The Advantage of Serial Presentation. — Under this
condition, each figure makes its own impression — is presented
to consciousness in an unprejudiced way — and yet does not enter
consciousnesses an isolated or stray fact. Continuity within the
whole range of facts, covered by a series, is secured, thus mak-
ing change an element of the objective facts, not of subjective
attention, as is the case with Fechner and Witmer. And the re-
lations of every new fact are already laid down by the mental
disposition aroused by the serial order. In all this we have a
marked approach to conditions of normal life where, for ex-
ample, our interests tend toward groups, and the continuity of
conscious experience lies within these larger areas.

3 . The Advantage of the Motor Element. — The subject was
allowed to take any card in his hand about which he might be
in doubt, and frequently a decision was arrived at very quickly.
Also in connection with each judgment, whether it was favor-
able or unfavorable, the motor factor was introduced with the
removal of the card from the field of vision. We claim this
had the advantage of naturalness. Our judgments are, for the
most part, expressions of motor significance.

We used, in all, four series of cards. In each series the
length remained constant, the width alone varying. The sizes
follow :

Series i, length 80 mm., width from 25 mm. to 75 mm. by
2.5 mm. steps.

Series 2, length 90 mm., width from 25 mm. to 85 mm. by
2.5 mm. steps.

256

T. H. HAINES AND A. E. DA VIES.

Series 3, length 100 mm., width from 25 mm. to 95 mm. by
5 mm. steps.

Series 4, length 120 mm., width from 25 mm. to 115 mm.
by 5 mm. steps.

The following tables (I. and II.) present results obtained
from the use of the cards. The numbers 80, 100, etc., indicate
the length of the cards. The word « horizontal,' means the
card was so placed before the observer that its longest dimen-
sion was parallel to the line joining his eyes. ' Vertical,' was
at right angles to this position, and in the same plane. The
figures in the column mm., give the widths of the cards chosen,
in millimeters, and those in the percentage column, give per-
centage relations of these widths to the given length.

TABLE I.

Obs.

I/ength 80 mm.

Length 100 mm.

I/ength 120 mm.

Horizontal.

Vertical.

Horizontal.

Vertical.

Horizontal.

Vertical.

mm.

f

mm.

*

mm.

t

mm.

*

mm.

X

mm.

%

f 25.0

31

27-5

34

25.0

25

40.0

40

30.0

25

30.0

25

Cr.

. 27.5

34

37-5

47

50.0

50

7O.O

70

75-0

63

5°.°

42

j 6o.O

75

60.0

75

90.0

90

1 1 0.0

92

6o.O

50

75-o

63

Da.

5°-°

62

55-o

68

65.0

65

75-o

75

80.0

67

85.0

71

'30.0

37

67.5

84

25.0

25

EC.

. 55-o

68

70.0

87

35-0

35

65.0

65

35-0

29

105.0

88

75-0

94

75-0

94

90.0

90

90.0

90

105.0

88

1 1 0.0

92

95-0

95

95-0

95

115.0

98

He.

135*

43

35-0

43

50.0

50

95-0

95

65.0

54

115-0

98

157-5

72

57-5

72

95-o

95

115-0

98

Ho.

J37-5

47

30.0

37

45-0

45

40.0

40

55-0

46

45-0

38

142.5

53

25.0

34

50.0

42

50.0

42

(40.0

50

70.0

87

45-0

45

70.0

58

Ke.

1 45-0

56

40.0

50

60.0

60

60.0

60

80.0

67

(

85.0

85

[45-0

56

75-0

94

55-0

55

95-0

95

60.0

5o

115-0

98

Mai.

1 55-0

68

70.0

70

75-o

63

1 75-0

94

95-0

95

115-0

98

Mas.

("40.0
162.5

50
76

40.0
62.5

50
76

55-0
80.0

55
80

55-0
80.0

55
80

60.0
70.0

50
58

60.0
70.0

50
58

Mv

J42.5

53

42.5

53

50.0

5o

55-0

55

70.0

58

70.0

58

iviy.

I

60.0

60

f27-5

34

37-5

47

45-0

45

45-o

45

45-0

38

45-0

38

Pe.

I 37-5

47

57-5

72

55-0

55

55-0

55

70.0

58

60.0

50

I 75-0

94

60.0

75

I

75-0

94

f4o.o

50

70.0

87

45-0

45

65.0

65

35-0

29

90.0

75

"Re

1

50.0

50

50.0

42

JDWB«

1

65.0

65

80.0

67

I

85.0

85

90.0

75

PS YCHOL OG Y OF &ST11E TIC RE A C TION.

TABLE II.

'57

Observer.

Length So mm.

Length 90 ram.

Horizontal.

Vertical.

Horizontal

Vrri,, .:

mm.

Per cent.

mm.

Per cent.

mm.

Percent

mm.

Per cent

Ra.

35-0

43

35-0

43

4O.O

44

40.0

44

Me.

27-5

34

25.0

31

32.0

35

25.0

27

N.

30.0

37

27-5

34

27-5

30

30.0

33

Sh.

62.5

78

52.5

65

65.0

72

65.0

72

L.

35-0

43

60.0

75

65.0

72

50.0

55

Sc.

(40.0

50

70.0

87

35-o

38

25.0
80.0

27
88

T.

32-5

40

25.0

3i

30.0

33

32.5

36

(50.0

62

47.0

58

55-0

61

60.0

66

Ha.

J52-5

65

50.0

62

57-5

63

155-0

68

52-5

65

Po.

42.5

53

50.0

62

57-5

63

57-5

63

Do.

70.0

87

70.0

87

80.0

88

80.0

88

'37-0

46

25.0

32.5

35

30.0

33

40.0

So

27-5

34

37-5

41

32.0

35

Pa.

- 42.5

53

30.0

37

40.0

44

45.o

56

37-5

40

42.5

47

45-0

50

"45-o

56

52.5

65

80.0

88

80.0

88

Ko.

67.5

84

55-0

68

75-0

93

60

75

62.5

78

A superficial examination of these results, is sufficient to con-
vince one of the absence of any single principle determining the
choices of these various observers. Indeed the results given by
a single observer, as EC. or Mai. shows that very different types
are liked equally well. This is what is to be expected, if we are
controlled in making such choices, by the associations we con-
nect with such forms. That association does play a conspicuous,
if not an important part, is borne out by the introspections. The
observer likes two or more very different forms because of the
different purposes which they subserve, and different uses which
they suggest.

Thus, we find that the introspective notes reveal quite dis-
tinctly, that the reason for liking any particular figure or group
of figures is, either (i) because of the similarity of the propor-
tions of the chosen figure to those of figures met with previously,
or (2) because the chosen figure fits in with a group of interests
already developed in the case of the individual observer. As
examples of the former, we may cite cases in which there is a
likeness to ' calling cards,' ' note books,' « writing tablets,'

258 T. H. HAINES AND A. E. DA VIES.

1 posters,' ' looking glasses,' etc. As an example of the latter,
one observer says, that a given proportion of sides is good
1 because she would choose it for a canvass on which to paint a
marine sketch.' This observer paints marine scenes. It is thus
not merely likeness to other canvasses which have been seen,
but the arousal of personal interests that determine particular
choices. Another prefers given rectangles because they are the
' figures he likes to deal with in geometry.'

Beside these two general types of motive, we find also other
influences, which, while not so widely distributed, are suffi-
ciently prominent to deserve separate mention. (3) Solidity
and completeness, give rise to certain of our aesthetic judgments.
This is the case when a rectangle is spoken of as ' complete in
itself,' or ' has wide enough base to hold it up,' or when it pro-
duces the feeling that it can be 'left alone.' (4) A complica-
tion of motives often determines such choices. If a rectangle
suggests a panel, its selection is due to familiarity ; but if, in
addition, it is a panel suitable for the ' Gibson girl with a curl
over her shoulder,' the limitation of use to which the panel is to
be put, is imposed by the personal interests of the individual.
These introspections indicate that the results here presented,
are not wholly lawless, as might appear at first sight, and at the
same time they point out the direction in which an explanation
is to be sought. Two individuals may like different figures
because they are controlled by different motives. In the same
way, different grounds of choice, determine the same person to
like two very different figures equally well. Looking at the
tables, we readily see examples of what might be called type-
selection, rather than individual-selection. E. g., EC. in Table
I., shows clearly at least two t}^pes, in every series. So also do
He., Mas. and Pe. Pa. of Table II., is a good case of many
choices, but all conforming to a simple type. All four of these,
in the first column, e. g., ranging in width from 46 per cent, to
56 per cent, of the height, are equally satisfactory and for the
same reason.

The evidence, however, for the existence of types, is yet
stronger from the introspective notes. Some observers come
with preferences ready formed, but not well defined, so that

PS YCHOLOG Y OF ESTHETIC REA CTION. 259

anything within certain fairly wide limits 'suits them. The
figures conform more or less closely to what they demand. The
observer says, e. g., 'I always liked figures with even sides,' or
* These are exact proportions, in other words, a rectangle whose
width is about half its length,' or * a rectangle must be longer
than twice its width, and shorter than three times its width.' It
is often harder for the observer to select a given figure, than it
is to say that five or six suit him. In other words, the group
is easier selected than is a specific figure. But as you pro-
gress in a series, there comes a limit somewhere, that marks
off this type from others ; it is the limit of forms capable of
being assimilated under the ruling concept. If choices are
made beyond this, they are determined by other motives and
belong to other types. Thus there are comparatively wide
limits to a type, and yet types are clearly discriminated in given
individual observers, and between individuals.

These facts in regard to types, show the improbability of
the golden section being the aesthetic norm. They also point
out the error of applying the method of averages to such
results. To use the method of averages, in cases where
such significance attaches to individual results, is indeed to
kill the goose that lays the golden egg. So far from being a
source of embarrassment, these variations themselves contain
not only their own explanation, but in them is to be found the
germ of the explanation of this whole class of aesthetic judg-
ments. The variations then must be preserved. The method
of averages, as applied by Fechner and Witmer, vitiates their
results.

But the apparatus and the method of experimentation with
cards, also come in for a share of criticism. The steps in the
series may be too wide. E. g., in Fechner's series of only
ten cards, there are probably many cards left out which some
observers would prefer to any that are given. The same holds
of Witmer's series and our own. Clearly the ideal apparatus
for this experiment, would permit the observer to make his own
figure.

We used a piece of apparatus (see Fig. i) which allowed
of this possibility. It consisted simply of a screen of black

260 T. H. HAINES AND A. E. DA VIES.

card-board set up in a vertical position, 1.3 meters from the
eyes of the observer, when he was sitting in a chair fixed to
the floor. This screen was one meter square. In the center of
the screen was an opening 10 centimeters square. The center
of the opening was on a level with the observer's eyes. A
simple mechanism provided a means of moving a black card
across the opening, behind the screen. As a part of the frame
to which the black card of the screen was fastened, there were
two horizontal strips, A (Fig. 2), about 25 centimeters apart.
Vertical saw slits were cut in these. A slide, B, rested upon
the upper one of these pieces, and a card C, fastened to it, and
depending through the slits, moved back and forth from side to
side of the frame behind the window from the observer. Cords
D and E, from either end of the slide, passed through pulleys, F
and 6r, at the sides of the screen, and thence to the hands of the
observer on the other side. The depending movable card was
white, but in its center was pasted a black card, more than wide
enough to cover the opening in the screen. The edges of this
were exactly parallel to the opening of the screen, so that on
whichever side of the opening it might be, right or left, some of
the white card was exposed. Thus the figure presented to the
observer, in white, was an exact rectangle. A millimeter scale,
If, on the back of the screen, afforded a ready means of measur-
ing exactly, from behind the screen, from right or left, any width
of figure the observer might give. The work was done in a
dark room, the only source of light being a sixteen candle
power light held in a box directly above and behind the ob-
server. Light was emitted through a three-quarter-inch hole,
and even this was covered with tissue paper (Figs, i and 2).

By this simple apparatus it was possible to secure the motor
reaction on the part of the observer, which had been found so
useful with the cards. The observer had it in his power to
modify the width of the figure within the limit of 10 cm., by any
minutest step which he could perceive. Moreover, it was pos-
sible to go over uninteresting parts of the series with haste,
thus avoiding fatigue, gaining time, and obtaining results with
a more constant state of attention. The apparatus also pro-
vided a ready and efficient means of obviating two sources of

PS YCHOL OGY OF ^B5 THE TIC RBA C TION.

FIGS, i, 2.

262

T. H. HAINES AND A. E. DA VIES.

error which it itself might seem to introduce, namely, (i), by
the direction of approach from right and left, and (2), direction
in the series, /. e., increasing or decreasing width. Each of
these was obviated by the method of reversal. We approached
the question always in the four ways thus made possible :

1. Closing in from a white square from the right.

2. Closing in from a white square from the left.

3. Opening out from an all black from the right.

4. Opening out from an all black from the left.

Tables III. and IV. present results given on this apparatus,
with these four variations, by the same observers whose results
with the cards are presented above. The results given are
widths in millimeters for the four movements. The length was
100 mm. in all cases. The results therefore are also percentage
values of widths in terms of length.

TABLE III.

TABLE IV.

Obs.

i

2

3

4 , Obs.

i

2

3

4

Cr.

62.5

55-0

56.0

45-7 Ra

/4i.o

36.0

2I.O

28.0

Da.

65.5

61.0

54-0

55-0

\5o.o

54-0

31.0

41.5

EC.

f IOO.O

I 57.0

93-0

IOO.O

96.0
53-5

06.0 •»»•
? Me.

J39-o
\48.5

43-5
55-5

27.0

32.0

26.0
28.0

He.

f 92.0

97.0

93-0

93-7

f62.5

46.5

26.0

25.0

I 53.0

N

J 33-0

33-0

49-5

34-0

Ho.

/ 54.7

50.5

68.0

49-5

53-6

1 45-0

23.0
43-o

42.0

47.0
43-0

/ 51-0

51-3

49-3

49.0 Sh.

76.5

64.0

81.5

76-5

•

\ 49-5

47-5

45-0

45-0 I*

36.0

56.0

30.0

59-0

Mai.

64.0

70.0

75-0

71-0 Sc

/ 5T-5

56.0

52.0

48.0

TVTjic

/ 56.o

70.0

22.O

45-0

172.0

49.0

41.0

49.0

JJU.C19.

I 65.0

57-5

27-5

31.8 T

/ 42.0

56.0

45-0

44.0

My.

62.0

68.5

66.0

64-5

\46.o

45-5

46.5

44-5

Pe.

73-o

73-0

44.0

48-8 Ha

/ 64.5

65.5

70.0

70.0

f 72.0

61.0

59-0

X.i.CL.

5°-°

167-5

68.0

69.0

63.0

6<

1 63.0

61.0

38.0

47-o PO.

J6i.s

) Af C

57-5

58.5

55-o

T)n

(. 41-5
/87.7

39-°
85.5

56.0
84.0

52.0
83.0

U\J,

1 85.0

83.0

86.0

87.0

Pa.

/55-0

45-5

28.5

22.5

138.5

33-5

24-5

24-5

Ko.

57-5

81.0

91.0

80.5

Some advantages for this apparatus and method, already
anticipated, come out in the results here presented. Greater
facility in making a definite choice is secured. An observer
who gave five cards near together as being equally desirable

THE PSYCHOLOGICAL REVIEW, XI., 1904.
10 20 30 40 50

D.
Hor.

60 70

PLATE V.
80 90 loo

D.
Vert.

Wi.
Hor.

Wi.
Vert.

PSYCHOLOGY OF &S THE TIC RE AC TION. 163

would here give one as his choice: /'. *., one can more easily
make his choice, here, among several that belong to the game
type. It is evident, too, that sometimes a type is chosen very
different from any given by the same observer with the cards.
This is explained by the remarks of several observers. E. g.,
one says he is ' less limited by material here.' A card is a
card ; but here he is interested in form, as such. Again EC.
selects two types: the square, and one about 67 by 100 mm.
The square she says, she « just likes,' but the other form she
says ' would be a good form for a card if you dared have it.'
She was able to see it more purely as form. On the whole, it
is evident that the problem before us is more purely one of
form. The observer feels less hampered by the conditions, and
arrives at more definite judgments.

But the more serious and extended study of the problem is
yet to be reported. In it, we confined ourselves to this appa-
ratus. Altogether, there are seven observers whose results are
reported on the apparatus as above described in the horizontal
movement of the shutter. The screen was then turned 90°,
making the slide move up and down, thus giving the four
movements.

1. Closing from above. (Start with white square.)

2. Closing from below. (Start with white square.)

3. Opening from above. (Start with all black.)

4. Opening from below. (Start with all black.)

In this part of the work, there were in all, twelve observers.
Some of these worked an hour a week for several months.
Others worked only a few hours altogether. The amount of
work done by each observer is readily seen from the charts,
since each blackened square stands for a judgment. These
blackened squares indicate by their position what was the width
of the figure preferred, since there is a space for every milli-
meter of possible width. It has also seemed worth while to pre-
serve the absolute results, even with regard to each one of the four
ways of approaching the problem, since these are so strikingly
different in some cases.

In Chart I., under D. (observer) Hor., we find the results
presented as given by D. in the horizontal movement, for I, 2,

264 T. H. HAINES AND A. E. DA VIES.

3 and 4 as above explained. In (i) closing in from a white
square from the right, e. g:, we find he stopped once at a figure
39 mm. x 100 mm. But he gave most of them between 63 mm.
and 80 mm. wide, giving three each at 67 mm., 68 mm. and
70 mm., none at 71 mm. and one at 72 mm.

It is apparent at once in looking over the results indicated in
these tables that all the observers had quite different choices. It
is to be remarked that those observers who worked most with the
apparatus give the most scattering results ; and in the case of
one, D., where the height of the figure was changed, ' vertical,'
the choices are more widely distributed than in horizontal, in-
dicating decreased definiteness of choice. This, however, is
not to be taken as the natural effect of continued work. It is
explained in large part by an illusion of which he found it very
difficult to rid himself. The same holds of H., in a lesser de-
gree. This also was due in some measure to the same illusion.
In both these cases the direction of movement was responsible
for the scattering only in so far as it was responsible for these
illusions. Wi., Wa. and J. are the three other observers whose
results give us good ground for comparison of results in vertical
and horizontal movements. None of these show any consider-
able increase in scattering in the " vertical." All are clearly in-
dicative of definiteness of choice, under a single motive. K.
'vert' is another example of scattering results. It was very
noticeable in the case of this observer that, as the work pro-
ceeded from week to week, his limits of choice narrowed down.
It is safe to say that, in the last three hours of work with him,
all his choices fell between 38 and 48 mm. On the other hand,
the increasing scattering with H. and D. are no doubt due to the
increasing number of motives determining choices.

In general, the results show definiteness of choice. Where
two motives rule, the ground between the two groups is pretty
clear. See, D., ' Hor.,' and H., ' Hor.' Difference of types,
where two occur in the same observer, and in different observers,
comes out more significantly than in the card results given above.
It is pertinent here again to ask the question, What would be
gained by averaging the results of H., ' Hor.' to find how near
his choice fell to the golden section? Or what would an aver-

PS YCHOL OGY OF ^ES THE TIC REA C TION. 265

age of the results of Wa. and J. indicate? Of what imaginable
use could such results be?

The existence of two types in the same observer, for one
position of the apparatus, is clearly shown by D., * Hor.' and H.,
' Hor.' Both come to a wide one which they like * closing in,'
in movements i and 2, and to a narrow which is equally satis-
factory, ' opening out,' in movements 3 and 4. This interest in
two types from the first in these observers is worth considering
in relation to the later development of still other types. O. also
has a clear leaning to wide ones for movements i and 2, and to
narrow ones for movements 3 and 4. He said ' he would like
the narrow ones better in i and 2' if he could * content himself to
come to them.' H. is an example of overcoming the interest in
the narrow one, in about half the judgments in movements 3
and 4.

Another point of interest here is the relation between the ratios
of the two dimensions of the figure when the movement is hori-
zontal, *'. e.y height is greater than the breadth, and when the
movement is vertical, *. £., breadth is greater than the height.
Fechner found in his measurements of pictures that Genre pic-
tures, whose height was greater than breadth, had a normal size
of i. 202 by .992 Prussian feet, or ratio / : 82. Genre pictures
whose breadth was greater than height were 1.737 by 1.389;
ratio / : 78. Landscapes, height greater than breadth 1.890 by
1.330; ratio i :fo. Landscapes whose breadth was greater
than height, 2.271 by 1.571 ; ratio / : 68. These results were
the product of very extensive averaging, as already explained,
and they were not very far apart. Our more intensive study of
a few individuals presents no clear evidence for the one side or
the other. Wi. and L. clearly prefer a narrower figure when
it is turned on its side, /. £., their figure whose height is greater
than its breadth presents a smaller ratio of shorter side to longer
than do their figures whose breadth is greater than height. The
former for Wi. are about 65 by 100, and the latter about 57 by
100. This is in Fechner's direction and more so. But H. * ver-
tical,' compared with ' horizontal,' shows a clear widening of
both types ; J. also shows a clear broadening of the figure (one
type) when the figure is turned 90°; D., even with this scat-

266 T. H. HAINES AND A. E. DAY IBS.

tering, shows a clear broadening of both types ; Wa. is practi-
cally the same in both series.

III.

We turn now to a classification of the motives leading to the
choice. These are shown by the introspections of the observers
themselves. They come in answer to questions both as to why
they like those they do like, and why they do not like various
unpleasant figures which they either make themselves or which
the experimenter gives them. These are often very productive.
Now these reactions of pleasant and unpleasant character are,
of course, relatively simple likings and dislikings. They are
simple reactions of an observer to a comparatively simple content.
In any such case it is natural to find the reasons for liking,
sometimes placed in the object and sometimes expressed in sub-
jective terms. Thus the figure is now satisfactory in itself and
now satisfactory because it fits in with a particular purpose of
my own, or it 'conforms to an ideal proportion of mine.

This we make the basis of our first division of motives to
choice. We find five classes, as follow. They form a con-
venient though not strictly a logical division of motives.

1. Motives found in the object. These are what may be
called pure aesthetic judgments of simple type. The figures,
so judged, seem good in themselves. We do not want to do
anything with them. In the other classes the observer does
want to do something with the figure, or it expresses what he
wants to do.

2. The figure soon groivs from the immediate interest of
(i) to a definite suggested use. These we call suggestive.
Emotion and apperception function in these cases.

3. This apperceptive element precedes the reaction. There
is a preconceived ideal, as e. g.> in the ordinary association. Ex-
pectation is definite.

4. Simple sensory elements, sometimes illusions, make for
pleasantness or unpleasantness of the figure.

5. A pure motor element comes into prominence.
The notes will explain these.

i. In giving 30.6 immediately after 70, D. says, "That's

THE PSYCHOLOGICAL REVIEW, XI., 1904.

10 20 30 40 50

O.
Vert.

60

70 So

PLATE vii.

90 TOO

J.

Hor.

3

4
Vert.

3

4
R.
Hor.

E.
Vert.

PS YCHOL OGY OF ^£5 THE TIC RE A C TlON. 167

easy. I like it for itself. I wonder now how I like the broader."
Again he says, * there is an absence of demands.' Again,
' there is here a permanence of mood.' He mentions pulses of
feeling as he moves the shutter in. These, however, he finds
occur at definite places in the scale, as he tries it with different
rates of movement. Many observers use such expressions at
the figure is ' easy,' it is * comfortable,' * it is neat,' it * fits in,'
it is « complete in itself.' Again, * It is open and has variety,'
say D. and K. * The lines differ, and are yet in harmony,'
says L. 'It is substantial.' H. says of 67 'vertical' that * it
is substantial and complete in itself.' He says " that 75 as
compared with 90 ' horizontal ' is pretty satisfactory " ; 90 is
'too clumsy.' D. says of 70, « It is too bulky, but I could spare
very little.' L. says, ' The square is too bulky and has too
much sameness about it.' Moving horizontally to 35, D. says
' all wider ones are too squatty,' by which he seems to mean
they are too wide for their height.

2. This class is the direct outcome of i. In fact it is often
hard to assign a given case. These cases the observers describe
as being « sudden discoveries.' They are seized with a liking
for a figure, which was totally unexpected. " It was an arrest,
I had to do the stopping myself." They had failed to find any-
thing in it before. An « immediate liking' for it which is not
coordinated with any preconceived ideal or associated use,
seizes them. These interests may soon develop. They often
do. But at the moment of its ' bursting in,' there are no in-
terests or developed feelings. To K. and S. it comes with a
« thrill of satisfaction,' and Wi. likes it as he would * like a $10
bill handed him as a present.' D. says of 42, it ' somehow
fixed itself into shape.' Again it is described as * causing no
jar.' Another division of this subclass is differentiated by the
emphasis placed upon the suggestions arising out of the figure
which is at first liked in and for itself. E.g.> one observer
says, « I stopped for this ' and the next moment ' it looks like a
memorial tablet.' He ' finds an interest ' and this develops. In
other cases, a ' general idea of the place where a favorable
judgment is coming,' develops as the shutter moves. The sug-
gestion as to the goal of the process, seems to come out of the

268 T. H. HAINES AND A. E. DA VIES.

process itself. Again an observer refuses to go on because of
an unpleasant figure, to which the process seems to be leading.
The case of O. fearing to go to the narrow one which he said
he preferred to the broad one given, probably belongs here.
The middle ground between the broad and narrow contains
some such suggested bugbear. H. once refused 60, which
had been the previous jugment, but went to 39, because he did
not know ' what it (60) was useful for.' For him, suggestive-
ness is a positive factor, and distinct from associations, i. e.,
from preconceived ideals and uses. He demanded a certain
development.

The square, or what the observer called a square, was fre-
quently chosen. Some of these seem to be cases of the kind
we are here dealing with. The observer explores the whole
series and finds nothing that develops an interest or proves
suggestive. He takes the square because nothing else is inter-
esting. Not that it is positively interesting or suggestive ; but
in a series failing to produce a development, the observer makes
choice of a figure which is least objectionable. This is prefer-
able to no figure at all. He finds no goal in form itself, and
so sets himself to the more formal task of making the sides
equal. This is the only thing he can do. This may be in part
the explanation of the rising curve in favor of the square in
Fechner's results. (See table S. 195, Vorschule der Aestketik.}
Another motive for the selection of the square is present no
doubt. The square may be the preconceived ideal. But ob-
servers often said, ' I like this because I don't like anything
else.' This is given also for the narrower rectangles. E. g:,
' This is not bad, though I can't get what I want,' or, ' I like
this (38.7) because all others are unsatisfactory.' Again an
observer tires of a type, and takes something different merely
for the sake of a change. H., working on the motive of calling
card, made a ' good envelope.' O., getting tired of 30, which
he had been giving quite regularly, gave some in 60 and some
in 10 and 20. Such a motive often explains K.'s changes from
the norm which stands out so clearly in his results. Sometimes
the ' possibility of developing in two directions ' is a determinant
factor. E. g., 'I like 87.3, because it is partly in the square

THE PSYCHOLOGICAL REVIEW, XI., 1904.
10 20 30 40 50

L.
Hor.

L.

Vert.

K.
Vert.

S.
Vert.

PLATE VIII.
60 70 80 90 100

CHART 4.

PS YCHOLOG Y OF ^C5 THE TIC HBA C T/ON.

series, and yet is not a square.' Again 45.3, is « midway between
the broad and narrow. Possibilities in these two directions
bred the interest in this.'

3. In the last class of cases the suggestion grows. In con-
trast with it those to which we now turn, find the predominant
motive in a preconceived idea. We may for convenience call
these associations. We find a definite expectation, and the
figure * comes ' to fit it. It is described as the ' developing of a
negative.' L. says ' I have always liked pictures and such things
of this shape (54-62), /'. e., tall, high and narrow.' Another
sees a window shade, and stops when it is the right distance up.
D. says (60.5) : ' This is what I have been trying for.' Some-
times these ideals cannot be realized. The observer is certain
he knows what he wants, but he cannot find it. It eludes him.
" I need something narrower, but I can't find it." Another
says, « I cannot get this (60) any time I want it.'

Among these preconceived notions which control choices
are to be mentioned the things one knows about the golden sec-
tion and its theoretical place in such work. The class embraces
all notions of what ought to be the mathematical relations be-
tween the sides of a rectangle. E.g., H. says : « Mathematical
associations as of the golden section assert themselves once in
awhile.' This observer was giving, as a general thing, 28 and
90. This was also true of J. " General relation of golden
section present." No other observer acknowledged this as a
motive. Many of them had never heard of it. We made it a
point to preserve their naivete. Preconceived bad forms are
also often determinative. D. says : ' I thought I was coming to
something I would not like and so stopped.' In other cases it
is said, ' to go on would be to destroy it.'

Associations are of course very numerous. A given form is
good, because it is the shape of something which is familiar, or
which one likes. Such objects as slabs, blocks of marble, step
to entrance, coping over a doorway, sarcophagus, visiting cards,
envelopes, photograph cards, window sills, entrance to a large
house, open courtway, marble pillar, double door, church
entrance with arch, bronze memorial tablets, transoms, window
shades, loads of hay, and rising curtains, are found in the list.

27° T. H. HAINES AND A. E. DA VIES.

These vary from individual to individual and \vith the same
individual. H. was particularly fruitful in these. But all ob-
servers declare, many times, they have no associations.

4. The sensory elements indicated in this division are appa-
rent chiefly in certain illusions. Three observers get illusions
of movement. For W. (horizontal) ' with steady gaze, both sides
seem to move out,' when in reality, only one moves. D. had a
very lasting illusion of movement in the vertical series. He
seemed persistently to desire figures of as much as ten centi-
meters width, but wanted more height than he could get. Of
78.2 he says, 'from here, the sides come in until I get the
narrow one (47) and then the sides stand still.' Here there is
an entire change of type, for, ' when he narrows still more the
sides shoot out.' This in D. naturally leads to an illusion of
form. He says, 35.2 "ought to get higher. But instead of
doing this it gets broader." Of 55, he says, it needs to be
higher and lower at the same instant. If he increases the height
(vertical dimension) the sides come in, and if he decreases it the
sides go out, and so, in both cases, his aim is defeated. Some-
times a whole series is given where this illusion does not ope-
rate. But usually some one or more in the four is affected.
Where it is not operative, the wide figures, about 84-87, are
given. The trouble with those wider than 49, under the illusion,
is that the breadth leaves, or the horizontal lines are too short.
Of 78, again he says, ' I like it when I take it turned 90°.' It
is well supported. But moving along the horizontal line, as it
is, it is flat. Again, 60 is good so long as he keeps his ' atten-
tion off the horizontal lines, and so resists the tendency to
narrow it.' Control of attention can defeat the illusion. 17 is
the first place in closing where he could * defeat the horizontal
cramp.' This illusory transfer of movement may be responsible
for the greater spreading of D.'s results in the vertical series.

Another observer gets an illusion of the white card moving
to meet the black, whereas in reality, it moves with it. Another
sees the white as ' uncrumpling ' or moving out both ways with
the ' new substance welling up in the middle,' as the figure
opens ; and for him the white seems to be ' crushed in ' as the
figure closes. As to the form, again, the vertical lines are often

PS YCHOL OGY OF ^CS THE TIC RBA C T1ON. ^l

seen converging toward the top, in the horizontal movement.
The corners are too large ; the advancing line is • jagged ' or
« bulged,' or « wavy.' In his first work (horizontal ' moving in ')
D. says of 61, * It is jagged up to this point ; 61 assumes definite
appearance ; and smaller than this, is amorphous.' Another
likes the figure whose sides are broken and not exactly at right
angles to each other.

Another wants less light, * to remove harshness of outline.'
One says, * the narrow ones get into a shadow,' when there is
no objective shadow. Wi. says 64 is 'soft and restful to his
eyes.' * The glare of 95,' he said, « was what made the sides
slope in at the top.' He complains of 40 that « it is a little dark,
the corners do not show up well, and has to move to get it
clear,' though no complaint of darkness was made about 28,
immediately afterward, when 28 was preferred to 40. D. often
wants something bigger than he can get. Of 44.2, 'I want
something the size of the square and the proportion of this,' and
of 49, ' I suppose this is what I would like if I could see it large.'
These illusions and demands are all in a measure criticisms on
the method, but more particularly are they cited here to be reck-
oned with in explanation of the results.

5. The motor element, it is true, was brought in by the ex-
perimenter in every case when considered. That is he care-
fully questioned the observers after considerable work had been
done, as to the presence, especially of eye movements. Some
follow the moving edge of the black card, but they find this
does not give them a satisfactory view of the changing figure.
Sometimes, they move the eyes alternately, first in one direction,
and then in the other. Wa. says she probably knows when she
has the right figure, by eye movement. D. says of 74.7,
« There is a balance of the two movements of the eyes.' There
is a pain in the eye for further lateral movement. * But the
square (100) just before it,' is not explored by eye movements.
"And 29.5, (horizontal) is explored up and down, only. The
narrow top is easily measured without movement." He is very
certain that the feeling, * the figure is the right one,' does not
come as a development of kinesthetic sensations from the eyes.
All observers agree, that they really size up a figure with a fixed

272 T. H. HAINES AND A. E. DA VIES.

gaze, taking in the whole. K. says, ' when you get a satisfac-
tory one the eyes are still.' As one gets used to the apparatus,
he invariably drops exploration, finding the best results come
with steady gaze.

Of course, there is an important motor element in that the
observer makes the figure for himself by pulling on the cords.
One likes the narrow figures, 'because they are easy,' requir-
ing only a little movement. Another likes it, because ' I know
I am making it.' And wherever the observer has a ready-
formed purpose, this gets fulfilled by his own production, if the
quest does not prove futile, by reason of limitation of opportunity
or otherwise. His doing is an important factor in the resultant
satisfaction. By the balanced control of the pulls executed,
through the two hands, he has made that which fits his purpose.
It is his own, and he likes it. Some observers say, ' I like that
now, since I made it or found it myself. I do not know that I
should, if you had showed it to me.'

IV.

In the face of such a great variety of motives as above
described, can we hope to find a single principle of explanation
of the species of phenomena under consideration ? There are
very manifest differences in kind of motive. In fact, we find
in this single study as great variety of motives, as there are
theories occupying the field. In view of this variety, it is hardly
to be expected that any one of the current theories can be
adequate to a complete explanation of these aesthetic phenomena.
Associations, for example, are clearly inadequate to explain the
reactions of many of our observers. One observer never had
associations with the figures she liked. Only when asked why
she did not like the figures that were repulsive to her, did she
find associations. She could never give a reason for liking the
figure, except such as were descriptive of the figure itself. No
more can motor sensations be made the basis of a complete
explanation of aesthetic reactions. They are no doubt a factor,
but we find they are only one, and comparatively unimportant.

In fact, Stratton (Philosophiche Studien, XX.), has shown
very conclusively, that eye movements made in exploring a

PS YCHOL OGY OF s£S THE TIC RBA C TION. 373

pleasing figure, of either right or curved lines, do not follow
the graceful and easy lines of the figure. The movement is
jerky, and is interrupted by frequent stops. This, of course,
does not preclude obtaining a fairly accurate measure of the two
dimensions of the figure, by the eyes ; but it does warn us not
to rest too much upon the kinaesthetic sensations from the eyes
in the explanation of our choice of rectangles. The play of
motor elements which are not actually functioning, is of course
important. The figure is sized up by « staring at it* say some
observers. Of course, given retinal stimulations, may mean for
the observer, certain definite movements. The local signs are
fused, by means of the suggested movement, into space percep-
tion. The perception, as in all cases, grows only by the integra-
tion of motor with sensory elements, even though these motor
elements are only latent movements.

There is though, more importance attaching to the sensory
side of the process, so-called, than is sometimes allowed. In
such an experiment as our own, the simultaneous stimulations
of parts of the retina with light of greater intensity, brings out
the figure independent of eye-movement sensations. Having
given the adaptation of the eye to the dark screen, in our experi-
ment, what happens when the shutter is drawn open ? Clearly
there is a relation between the required adaptation to the brighter
surface now exposed, and the existing adaptation of the dark sur-
face, which must prove a determining factor in the size of the
chosen figure. The greater illumination in the centers of the
retinas, not only calls for an adaptation in these areas, but also to
a new adaptation or a readjustment in the surrounding portions
whose absolute stimulation has not changed. In other words,
McDougall's l explanation of simultaneous contrast applies here,
in part. This double process of readjustment, and the relations
between the two processes, explain many of the illusions of
form, such as jagged lines, sloping lines, and large angles. It
may also explain many of the demands for size, as well as the
expressions, * it is soft,' * comfortable,' and « easy on the eyes.'
There is of course, a motor element in these so-called sensory
processes. Retinal adaptation clearly has a motor side. Mc-

*Brain, 1903, Pt. II., pp. 183 ff. Sec also Mind, N. S., XII., pp. 473-488.

274 T. H. HAINES AND A. E. DA VIES.

Dougall shows that the after-image depends upon accommoda-
tion. And adaptation, doubtless, depends upon some motor
processes. But for any definite statement of what it is, we are
awaiting the future results of histological physiology.

Other motor sensations than those of ocular origin, no doubt
play an important role. We have already mentioned the arm
movements. One observer was very conscious of these. He
often put his forefingers out on the cords and thus controlled
the movements of the shutter through the more accurate arti-
cular sensations from the fingers. These sensations are a basis
of choice, in an observer who says, ' I like that because I can
make it easily.' But much more are they the means of doing
what one wants done, — they are the media for the realization
of our purposes. And this calls attention to a transition similar
to that made above. As the eye movement may be the basis of
the perception of dimensions, and so of the preferable figure,
and yet not become a movement, so our motor tendencies as a
whole, may be the basis of our choice of a rectangle, although
these lie wholly dormant. This is the basis of many of our
choises which show no definite motor factor. The motor side
of these aesthetic processes, is highly important for the purposes
of explanation, because it is so important a part of the process.

Another explanation of aesthetic judgments closely akin to
that of balance of eye-muscle sensation, is that which endeavors
to explain our likings by what may be called a balance be-
tween attention forces. It may be stated objectively as a bal-
ance of attention-drawing features in the object. They are
sometimes spoken of as forces in the picture or the line. But
they are forces of attention, so that it is ultimately and funda-
mentally a balance of attention and interest, which is made the
basis of explanation. Thus, in the case of our rectangles, it
would be said, that a given width is coordinated with 100 mm.
of height, and a given height with 100 mm. of width, because
this width or height has an interest or attention-drawing power
equal to the interest in the 100 mm. height or breadth, respec-
tively. This theory is valuable, in so far as it is an explanation
of the wide departures from the golden section. One's interests
change. The observer views the shorter (or the variable)

PS YCHOL OGY OF xCS THE TIC RBA C TION. 175

dimension, in different lights at different times. It has now
more of his attention, and now less, owing to the different rul-
ing interests. So its affinities are different. Hence also the
varying results, from observer to observer, as well as in the
same observer from time to time. Our objection to this theory
is, that it is not an explanation of the complex phenomena
under investigation. It is very aptly descriptive of the facts in
broad outline, but it leaves us with the problem of explanation,
where we start. It seems on the face of it to go back to very
simple elements. But attention is both the most baffling, and
the most complex of psychological concepts. We cannot hold,
therefore, that a theory is final which bases on attention, and
says the asymmetrical elements are harmonious because of a psy-
chical balance brought about by more attention or interest to
the shorter. Any balance brought about by a heightening of
the vividness of one element through attention, can by no pos-
sible means be ultimately simple ; for this heightening is itself
a complex process, not mere addition. It is a typical piece of
organic mental growth. This being granted, it is the first
business of the student of the phenonenon to gain insight into
the mechanism of this growth. And when he has seen the
component part processes of each, he will be in a position to
understand the aesthetic balance between two given whole proc-
esses. What we have called suggestiveness^ helps us, in some
measure, to the desired insight.

In our experiment, under normal conditions, the observer
was not consciously concerned with a balance. It was pri-
marily a question of satisfactoriness of form. To be sure the
form used is, to the analytic onlooker, only a composition of
the two dimensions. To consider the relations of these two
lines, each to the other, would be, a prior •*', a simple problem.
But our experience confirms us in the view that this form may
be considered — in fact is naturally considered — independent
of, and prior to, the relation between any two contiguous sides.
This is, in fact, a very primitive aesthetic experience, and in-
sight into the reactions of the class called suggestive, really
affords a basis for the explanation of the so-called balances of
attention. For it does not require there should be two new ex-

276 T. H. HAINES AND A. E. DA VIES.

periences in order that we should have an aesthetic reaction — a
balance between them. At least the restriction of the term,
aesthetic, to such limit, precludes the cases, which alone, as it
seems to us, give insight into the mechanism of the higher
aesthetic emotions.

In the cases called suggestive reaction, the experience comes
to a mind alert, but inactive, — attentive, but without any pur-
pose, save to get the experience that comes from the presented
stimulus, and to note the pleasantness or unpleasantness of this
experience. In some of these cases, the experience seems very
immediately pleasant. As contrasted with what we have called
purely aesthetic, there is a reason here for the pleasantness, in
the suggested use of the form. And they are different from
cases of association, too, in that the use is suggested by, and
comes out of, the experience itself. The emotional reaction
and the suggested use seem to come simultaneously. On this
point it is difficult to obtain definiteness by introspection. But
a -priori we would suppose that the suggested use preceeds and
that the pleasantness is an indication of the mutual appropriate-
ness of the experience and of the use, each to the other. This
a -priori view is supported by introspective evidence from
another class of suggestive cases.

In these, as the observer changes the figure, a purpose arises
which is not fulfilled. A goal is set for the process, as the proc-
ess itself is perceived, member by member. By the use of
this word goal, we do not wish to denote an end for action,
consciously present in the observer's mind. It is rather the
conative process set up by the sensory processes serially per-
ceived. This conative process is too large and too vague for
immediate execution, but it is that which determines the motor
outgo of the moment. There thus arises or exists a disparity
in the mind. Our clearest way of expressing this, is, that it is
a disparity between the goal for which the motor outgo starts,
and that which it realizes. It is, in fact, though, to the observer
himself, at the moment, simply a baffled emotional state. The
aroused emotional tendencies fail to get realized. The balance
here is between the ideal, however vague, which always means
conation, and the incoming experience.

PSYCHOLOGY OF ESTHETIC REACTION. 3??

The aesthetic moment is the moment of satisfied wants. In
so far as there is satisfaction with the new experience, there is
no further conative tendency. This is the case, in part, with
what we have called purely aesthetic reactions. It is character-
istic of the aesthetic mood. But right here in the sensi-motor
circle ' of the mental processes, at the point of impact of the
sensory processes, is the place where emotional tendencies or
ideals arise. The incoming experience either satisfies or it does
not satisfy the previous emotional tendency or conative impulse,
and in so far as it jars on it or fails to fit in with it, a new emo-
tional tendency arises, which, expressed in terms of intellec-
tualist theory, is an ideal. This in turn, leads to a new motor
outgo, which brings a fresh sensory process, and so the circle
repeats itself over and over; but it is mounting as a spiral,
and growth is occurring, through the organization of this men-

FIG. 3.

tal material called emotional tendencies. And the mechanism
of this growth is through the impulses imparted by the incoming
material jarring on the emotional incentive which led to the
motor outgo. And right at the point of impact of the sensory
income, as said before, is the place of the aesthetic feelings.
These are simply the satisfactions with what we have called
the^f/ of the incoming with the emotional incentive to the outgo,
as the unpleasant are the dissatisfactions with the misfits. Thus
we see the very fundamental place of these feelings in the
growth of the mind. Our meaning may be brought out better
by a diagram. Let A, A', etc. (in Fig. 3), represent succes-
sive sensory processes just arising through the action of stimuli
1 We refer to the circular process in mental life so admirably developed in
Professor J. M. Baldwin's Mental Development in the Child and the Rate.

278 T. H. HAINES AND A. E. DA VIES.

coming from objects D, D' ', etc. The psychic stream is rep-
resented by the line A-A"", and B, B' , etc., are incipient motor
outgoes. A-A', etc., are then places in the circular reaction
where the fit or the misfit of the new experience arises, this
being the feeling of satisfaction or dissatisfaction. In the case
of the latter, an emotional tendency to action organizes from
this point and emerges in action at B. The return sensory
process is at Af, and this leads to a new motor outgo, B' . So
the organization proceeds. This is simply to illustrate the place
as we conceive it, of the elementary likes and dislikes in the
sensi-motor circle before there is other psychic life.

Association and suggestion are such important elements in
our theory, and they are so likely to be confused, that some
added remarks are introduced at this point on their relation to
one another. To start with the characteristic difference from
the descriptive point of view, we may say that suggestion carries
with it the specification of a definite use for a given figure,
while association always affirms the likeness of a particular
figure to a class or group of figures. Put into general terms,
this means that suggestion is characteristically practical while
association is theoretical or scientific, so far as motive is con-
cerned. This has direct bearing upon the material and relative
simplicity of the two classes. As to material, we may say that
suggestion is predominantly emotive, but association intellective.
That is to say, the wants that are contemplated by each class
are different. Association seeks for a definite relation between
the given object and the observing subject, in which case the
aesthetic mood is held in abeyance, and is made to depend upon
the solution of a problem. For illustration of this we refer to
many cases where the square was chosen (see page 368). Sug-
gestion subordinates the relation between observer and the thing
observed, and allows free play for the feelings in determining
the aesthetic object. In this way, the ideal is found, not made.
It is discovered, not anticipated. Thus we may see that sug-
gestion is relatively much more simple than association. It
takes us back nearer to the elements of the aesthetic experience.

This may be made clearer by laying bare the processes in-
volved in suggestion and association. It is a commonplace of

PS YCHOLOG Y OF sBS THE TIC RE A C TION. 3J9

psychology to-day, that all psychic material is dynamogenic.
When we are at the roots of mental life, this means that the
question, 'What are you going to do ?' is much more funda-
mental than, 'What do you perceive ?' So that whether we
perceive anything or not, the primary arousements of conscious-
ness involve us in some kind or other of activity. It is not pos-
sible, on the basis of the present study, to go beyond this fact
of motor connection. It is however, the point which is most

FIG. 4.

emphatically borne out by the results which we have presented
above. This connection is the simplest form of psychic process
with which we are acquainted. What the connection is and
•where it is, we are not able to say. These remoter questions
are extremely interesting, but beyond our present work. All
we can now affirm is, that impressions coming from the object
D reach consciousness at the point A, and go out in motor dis-
charge toward the object from the point B (see Fig. 4).

FIG. 5.

The importance of the motor reaction consists in its making
secure the connection between A and B^ and probably, in giv-
ing greater definiteness to A directly. In contrast with this,
association is a more complicated process. Instead of defining
A through B we have here the interpretative idea C (see Fig.
5), which is a go-between for A and B. The definiteness
that comes to A from the motor outgo B in this case comes
through C. C is also a factor determinative of the motor reac-

280 T. H. HAINES AND A. E. DA VIES.

tion. That is to say, action is probably both ways in this com-
plex arc, as it is in the simpler one above.

Two other points of relation between association and sug-
gestion should be mentioned in passing. One is the difference
in the ways of their inception. The suggestive process starts at
A of the above figure and the associative process starts with C.
In the former, there is nothing until the sense experience sug-
gests it, and this runs simply to its own fulfilment in the simplest
cases, as a sort of psychological reflex. In the typical associa-
tive process, however, the preconceived idea C is the starting
point and controlling factor of the whole. The other point of
difference is already evident. The associative process is the more
general of the two. This is consequent upon the complexity
mentioned above. Suggestion leaves you with the particularity
that belongs to the individual object or form — it makes it more
concrete than at first ; association takes away the individuality
and substitutes the universality of the class. Association is
essentially a process of classification, and this implies empha-
sizing common features. This is what was meant by saying
that association comes from a scientific motive ; it is fundamen-
tally the process of classifying experience.

These suggestive and associative processes are not always
pure and clearly distinguishable. The suggestive, being the
more primitive, is always involved in the associative. There
are also many gradations between the two. The suggestive
process is often the starting point of an association. The process
starts as a suggestion, but the psychic processes develop so far
and so explicitly in advance of their realization that the idea,
or ideal use, of this kind of experience serves at once as a guide
to the motor outgo. Comparisons are prominent, and the whole
has been raised into the higher psychic plane. Whereas, had
the case remained simple we should have had a development
without conscious guidance, on the level of simple feeling and
mere conation. It is in rare instances, and usually under ex-
perimental conditions, that the human subject can know that he
has had such an experience. As another example of mingled
suggestion and association we mention those cases where an asso-
ciation helps on a suggestive process, but leaves it free to be

PS YCHOL OGY OF .<ES THE TIC ItEA C T1ON. 38 1

controlled by the objective interests. You may have a limit
imposed, as when you want a full length Madonna painted.
The size is a definite, associated restriction upon the process,
imposed by our ideas of use. But in many ways suggestion
works unhampered.

We thus get a glimpse of the way in which that which is
not perceptive or intellectual psychic material develops into that
which is intellectual. It consists in seeing how that which we
have called emotional tendency (A-B in Fig. 4) becomes ideal
or interpretative idea (Cor A-C-B in Fig. 5). The animal
that can feel an incrongruity and stay by it, adjusting itself
to its environment on the one hand, and adjusting the envi-
ronment to itself on the other, is the animal that will survive ;
and the fittest to survive, has most of this accomodative power.
Analytic power, which is the precursor of synthesis or general-
ization, which in turn is association, emerges at first very
crudely in a being able to react on an aspect of an experience,
neglecting unessential details. This kind of analysis at once
has great survival value, and the animal that can translate his
various impressions into an ordered experience will have begun
the ascent of mental as distinguished from organic evolution.
Thoughts are feelings objectified on the basis of wants through
motor response to existing environment. Feelings and cona-
tion are directly connected, and when feeling becomes generic^
t. £., tends to react in response to the object for less than the
object actually is, it is making possible that definite implication
of the object in consciousness, which is the characteristic of
intellectual life.1

1 The MSS. of this article was received on March 17, 1904.— ED.

CONCEPTIONS AND MISCONCEPTIONS OF CON-
SCIOUSNESS. 1

BY RALPH BARTON PERRY.

Harvard University.

Were the use of the term consciousness to be forbidden for
a season, contemporary thought would be set the wholesome
task of discovering more definite terms with which to replace it,
and a very considerable amount of convenient mystery would
be dissipated. There is no philosophical term at once so
popular and so devoid of standard meaning. How can a term
mean anything when it is employed to connote anything and
everything, including its own negation ? One hears of the ob-
ject of consciousness and the subject of consciousness, and the
union of the two in self-consciousness ; of the private conscious-
ness, the social consciousness, and the transcendental con-
sciousness ; the inner and the outer, the higher and the lower,
the temporal and the eternal consciousness ; the activity and the
state of consciousness. Then there is consciousness-stuff, and
unconscious consciousness, called respectively mind-stuff for
short, and unconscious psychical states or subconsciousness to
avoid a verbal contradiction. This list is not complete, but
sufficiently amazing. Consciousness comprises everything that
is, and indefinitely much more. It is small wonder that the
definition of it is little attempted. One of the most successful
efforts is that of Professor Ladd, who regards consciousness as
the difference (presumably from the sleeper's point of view)
between waking and dreamless sleep. This is equivalent to
the difference between more or less of something, and nothing
at all ; which is quite accurately true to current usage. Bald-
win's Dictionary of Philosophy defines consciousness, on the
one hand, as ' the distinctive character of whatever may be
called mental life,' and mind, on the other hand, as 'the indi-

1 Read before the American Philosophical Association, December 30, 1903.

282

CONCEPTIONS OF CONSCIOUSNESS. 283

vidual's conscious process, together with the dispositions and
predispositions which condition it.' But it is more customary
to say frankly that the term is indefinable. If it were taken for
granted that it is therefore better left unemployed in exact
thinking, there would be no occasion for objection. But its inde-
finability is more commonly attributed to the profoundness of its
meaning. Indeed the definition of being in terms of conscious-
ness is set down as the surviving and most illuminating truth of
philosophy. The hope is expressed that we may now postulate
it and proceed to more debatable matters. * And consciousness
so regarded as the fundamental ontological truth, is called
upon to carry and protect man's moral and religious interests.
Especially in the nineteenth century has this term suffered the
taint of eulogy, through being made the watch-word of non-
materialism. The advocates of the spiritual man, never over-
scrupulous in their choice of weapons, have in this case been
willing to confound the enemy by confusing him. * What are
you going to do with consciousness?' asks the idealist. The
materialist, not knowing precisely what it is, but convinced that
it bears no resemblance to a motion or a secretion, does nothing
with it. Whereupon the idealist shows him what he can do
with matter, and the materialist, who is a stupid adversary at
best, takes refuge in a general protest against metaphysics.
The defeat of materialism is not to be regretted, but there must
be no uncritical acceptance of the victor. The term conscious-
ness as at present employed is too reminiscent of this contro-
versy. It stands for a general propaganda, which runs some-

1 " There are certain accepted doctrines of modern philosophy — e. g., that
knowledge is only of phenomena, not of anything unrelated to consciousness,
and that object and subject are correlative — from which this conclusion seems
to follow so inevitably, that anyone who has adopted it must enquire anxiously
why it is not more generally recognised. If nothing can enter into knowledge
that is unrelated to consciousness ; if relation to a subject is necessary to make
an object, so that an object which no consciousness presented to itself would
not be an object at all ; it is as difficult to see how the principle of unity, through
which phenomena become the connected system called the world of experience,
can be found elsewhere than in consciousness, as it is to see how the conscious-
ness exercising such a function can be a part of the world which it thus at least
coSperates in making." Green's Prolegomena to Ethics, pp. 14, 15- Compare
the more recent statements of C. A. Strong, in Why the Mind has a Body, pp.
166, 183, 186. The criticism of such views as these is undertaken later in the
present article.

284 RALPH BAR TON PERR Y.

what as follows : psychology and transcendental logic disprove
materialism, seat God on his throne, and prove the immortality
of the soul. When one ceases to look upon them from the
standpoint of the counter-thesis of materialism, these are impos-
sible allies. Consciousness cannot mean everything and yet
mean anything. As a name for the psychological aspect of
experience, it may be shown to mean something definite and
important ; but consciousness so interpreted is confused and mis-
conceived when called upon to serve as a metaphysical account
of being, and is no safeguard of man's spiritual interests. Con-
sideration of morality and religion will be omitted from the
present discussion, which will attempt first to account for and
define a concept of consciousness, and second to criticise its
metaphysical use.

Only a succssful analysis can justify the proposal to account
for this concept in terms of psychological experiences, the more
so since the term « psychological ' must be defined at the same
time. But the arbitrariness of the procedure is at a minimum
when we begin where the race and the individual have presum-
ably begun to learn of these matters. Before a certain moment in
the development of reflection the self is theoretically indistin-
guishable from body, and conceptions of it throw no light on
the idea of consciousness ; while after that moment the self is
conceived with definite reference to a specific type of experience
which has come to be noted and differentiated. That which
makes this difference between the early cosmological, and the
later radical or critical conceptions of the self, is the distinct
employment of a set of ideas signifying seeming or appearance.
While primitive experience is entirely free from any general
idea of the dependence of objects upon the knowing of them,
there are certain accepted cases in which an experience is
definitely recognized as my experience, or certain facts which
are regarded as deriving existence from &for-me relation. This
is a very different idea from that of the functioning of the
sense-organs. That I see and hear and taste is a commonplace
of all experience, and I may study what I see, or the manner
of my seeing, without effecting any discontinuity in my prac-
tical or scientific world of things. But to believe that what I

CONCEPTIONS OF CONSCIOUSNESS. 185

see is constituted by my seeing of it, is to define a new realm,
an anomalous science, and possibly a new philosophical method.
Such a belief must arise very early in connection with discred-
ited or illusory experiences. Illusions so vivid as dreams are
doubtless in the beginning often regarded as unusually signifi-
cant experiences of objects, but such can scarcely be the case
with all dreams, with fever-deliriums, and with wanderings and
inventings of the imagination. And these adventures are
homogeneous with certain very familiar and normal happen-
ings. Experience is constantly correcting itself and discredit-
ing its earlier content. Observation and identification is a
process of self-correction. The surviving judgment is the last
of a series of discarded judgments which were once as living
as itself. They are not the object A, but « what I thought,'
« the way it seemed to me then,' my mistake, or confusion. To
be sure, such retrospect is not demanded for the direct purpose
of observation or identification, but they cannot altogether
escape the notice even of the man of affairs. They tend, as in
the case of the double images, to be neglected because not im-
portant. They become important, however, whenever the task
of thinking becomes specialized, and interest is aroused in con-
ditions that tend to determine its success or failure. Error and
confusion come then to be attended to, and designated as a
realm of idiosyncracy, to be corrected or repudiated by the wise
man. The appearance of these ideas in early Greek philosophy
is familiar history. They determine the common distinction
between ' truth ' and « opinion ' ; and the Protagorean doctrine
is an inference from them.1 The aspect of experience recog-

1 A special interest attaches to the earliest statements of this idea in philos-
ophy. The following are representative :

" It is not meet to act and speak like men asleep." "The waking have
one and the same world, but the sleeping turn aside each into a world of his
own." Heraclitus, Fragments 94 and 95 in Burnet's Early Greek Philosophers.

" Welcome, noble youth, that comest to my abode on the car that bears tbee
tended by immortal charioteers. It is no ill chance, but justice and right that
has sent thee forth to travel on this way. Far, indeed, does it lie from the
beaten track of men! Meet it is thou shouldst learn all things, as well the un-
shaken heart of persuasive truth, as the opinions of mortals in which is no true
belief at all. Yet none the less shall thou learn of these things also, since thou
must judge approvedly of the things that seem to men as thou goest through
all things in thy journey." Partnenides, in Burnet, oft. cit., p. 184.

286 RALPH BA R TON PERR Y.

nized in this old epistemological criticism has played an im-
portant part in modern philosophy, where it appears notably in
Spinoza's conception of modality and inadequate ideas, in
Kant's manifold of the internal sense, and in Hegel's doctrine
of subjective spirit. It furnishes the most likely definition of
the field of psychology, and with reference to its bearing upon
this problem, let us consider the analysis independently of its
history.

The first intent or bearing of experience is objective,1 as
expressed in the judgment, that is A. But experience proves
to be self-corrective. The content of A grows in the direction
of its own completeness. A is in the first instance more or less
problematical, and increases in articulateness. While the direc-
tion or interest remains the same, this experience is homogene-
ous, an experience, we say, of the same thing, or context of
things. But an act of attention is possible whereby the direc-
tion is reversed. With this new interest there now appears a
series of corrected experiences, to any degree of inadequacy.
These specific limitations may be noted and attributed to specific
conditions. In this wise the corrected and replaced experience,
in contradiction to the corrective experience, is viewed as merely
my experience, a term of my blindness and struggle. Since I
have now apprehended the thing itself, I can define my more or
less successful purpose with reference to the thing. In ordinary
experience I have my face to reality and my back to such of the

"And the soul is like the eye : when resting upon that on which truth
and being shine, the sou perceives and understands, and is radiant with intel-
ligence ; but when turned towards the twilight of becoming and perishing,
then she has opinion only, and goes blinking about, and is first of one opinion
and then another, and seems to have no intelligence." Plato, Republic 508 D,
Jowett's translation. Cf. 510, 511.

"The senses are variously named hearing, seeing, smelling ; there is the
sense of heat, cold, pleasure, pain, desire, fear, and many more which are
named, as well as innumerable others which have no name ; with each of them
there is born an object of sense,— all sorts of colours born with all sorts of sight
and sounds in like manner with hearing, and other objects with the other
senses." From Plato's exposition of Protagoras in Theaetetus 156 B, Jowett's
translation. Cf. 157.

1 1 am at pains in this part of the analysis to avoid any verbal suggestion
of the indispensableness of the subject-object relation. I shall, therefore, so
far as possible, use the terms ' thing ' and 'real ' rather than the equivocal term
'object.'

CONCEPTIONS OF CONSCIOUSNESS. 287

cognitive process as I have passed by. But I may turn and be-
hold the way I have come, together with its stages ; and these
latter I now denominate points of view in contradiction to that
which may be so viewed. Such is the psychical fact and the
reflection required for the identification of it. Let us turn to
the consideration of examples.

The most unequivocal instance is the dream. This is a
definite type of invalid experience, recognized as such from the
standpoint of a valid corrective experience. Were there only
dreaming, there would be no dreaming. Either I must myself
awake or have my illusions observed by another, who both
knows them and knows beyond them. The waking and the
dreaming differ in that the former not only succeeds the latter,
but includes and replaces it ; while the latter on the other hand
knows nothing of the former. The waking experience defines
my dreaming, and in the presence of the real judges it to be
unreal. When I wake up to the actual situation, my dreaming
takes on the duller hues of a subjectivity and fancy which I
significantly call my own.

There is a similar distinction between the narratives of the
historian and the eye-witness. The historian corrects the ex-
periences of the eye-witness by marshalling contemporaneous
events and by eliminating the more accidental sequences and
coincidences of observation. In view of the real order of
events, the uncritical report of an individual may be circum-
scribed and identified as such. A continuous series of maps of
the battle-ground, with the formations and movements of the
combatants, would so include and transcend the order of occur-
rence in the experience of a soldier of the ranks.

Let us turn to those instances that are due to the deliber-
ate psychological interest. The need both of an included
and of a supervening experience is here determinative of a
method, and is most clearly in evidence in the case of compara-
tive and experimental research. The experience of the animal,
child, savage, or abnormal subject, is viewed as within a valid
world of experience, and interpreted in terms of specific and
characteristic limiting conditions. In experiment these limiting
conditions are in part artificially provided, and with them is

288 RA LPH BAR TON PERR Y.

coordinated the report of the subject, the whole being contained
in the presumably or practically unlimited experience of the
investigator.

But lastly let us consider the more crucial case of introspec-
tion, and in particular, introspective attention to perception.
How is the psychological manifold differentiated from the thing-
manifold where there are no social relations involved? The
possibility of it is clear, the manner of it obscure. I can analyze
my perceiving on the one hand and the object of my perceiving
on the other with quite different results, and yet in the perception
they are indistinguishable. The difference must lie in my
interest, of in the direction of my attention, and it appears here
also that one interest is fundamentally determinative. Indeed,
the method is essentially identical with the judgment, ' I have
been dreaming,' except that in this case the invalidity of the
corrected experience is less radical. Introspection is retrospec-
tive attention to an experience which I now surround and sur-
mount. That more or less complete apprehension which can
now become a distinct manifold for me because I compare it with
the occasion itself, I call my state. The actual method employed
in this type of investigation is commonly hidden on account of
the rapid alternation of interests. My objective experience is
constantly awaking from new dreams. I must oscillate rapidly
between the standpoints of experimenter and subject. From
my standpoint as experimenter, my experience as subject is the
relatively inadequate experience whose boundaries I may now
view retrospectively and whose limiting conditions I endeavor
to analyze. Consider the case of my perception of a house,
which tends to reveal to me its true geometrical form, together
with the totality of its exterior and interior. In ordinary ex-
perience I have it so present to me ; practically, as is attested by
my dealings with it, and theoretically, as is attested by my de-
scription of it to another. But I may compare with this valid
experience the inadequacies which are contained and compen-
sated for within it. My corrected spatial perspective would
constitute such an inadequacy, and I may analyze this as respects
its content, and as respects the manner and the means where-
with the correction is made. In such procedure the house has

CONCEPTIONS OF CONSCIOUSNESS. 189

been regarded as the culminating event in a process of mind,
and the factors determined by such an analysis are called states
of mind. This interpretation of the method of introspection
might be further and more readily illustrated with reference to
imagination and memory. The same method holds in the case
of feeling, this psychical factor appearing in the experience, /
want or like A, in contradistinction to the experience, A is good.
Feeling is an invalid judgment of worth. In each case the field
of psychology comes into view only when an incomplete ex-
perience is recognized as such from the standpoint of an experi-
ence regarded as objective. The corrected or discredited
experience so determined critically in an experience of things,
is regarded as merely my experience, and may be analyzed as
such. But we must have passed beyond the psychical to be-
come aware of it. These psychical data cannot be called
things or reals in the same sense as the standard objects, for
they are completed and replaced by the latter. We therefore
provide a radically different category for them, and recognize
that their content is common to themselves and to things, while
their specific character is given them by their limitations and
context.

Accepting for the present this definition of consciousness in
terms of relativity, let us examine the attempt to construe it as
a philosophy. Such a theory might properly be designated as
psychological idealism, and is known under the names of per-
ceptual idealism, phenomenalism and sensationalism. This
theory arises from the thought of the possibility of indefinitely
extending the psychological manifold. Every corrective experi-
ence may, and tends to become in turn, a corrected experience.
There is no experience of which one may not come to say, * it
is my state,' or, « it is your state.' " At first sight," says Walter
Pater, who styles himself a new Cyrenaic, "experience seems
to bury us under a flood of external objects, pressing upon us
with a sharp and importunate reality, calling us out of ourselves
in a thousand forms of action. But when reflexion begins to
act upon these objects they are dissipated under its influence ;
the cohesive force seems suspended like a trick of magic ; each
object is loosed into a group of impressions — colours, odour, tex-

290 RALPH BARTON PERRY.

ture — in the mind of the observer. * * * Experience, already
reduced to a swarm of impressions, is ringed round for each one
of us by that thick wall of personality through which no real
voice has ever pierced on its way to us, or from us to that which
we can only conjecture to be without. Every one of these
impressions is the impression of the individual in his isolation,
each mind keeping as a solitary prisoner its own dream of a
world." l On such grounds one reaches the generalization that
every knowable object is someone's perception, or the more
radical persuasion that every knowable object is his own per-
ception. The only definable being is seeming. In terms of
the above analysis, this is equivalent to the proposition that
everything so far as knowledge is concerned is invalid experi-
ence. To assert this proposition is, of course, to plead scep-
ticism. But even as scepticism it is not tenable, since it is a
criticism of experience according to a principle. There can be
no experience of a world in which each mind keeps ' as a soli-
tary prisoner its own dream of a world.' The Protagorean
mind must itself have awakened and broken from its prison.
Madame Ackermann is quoted as saying : " My last word will
be : ' I have been dreaming.' " 2 But in that moment she will
at last have ceased to dream. Relativism will not do as a doc-
trine, though it may serve as an apology for silence. And
where the Protagorean principle has been asserted, it has almost
invariably been associated with a deeper metaphysics calculated
to make this principle itself a psychological one. In the case
of Protagoras himself, the world was conceived with Heraclitus
and Democritus as essentially motion. Perception is itself a
type of motion, and so incapable of fixing upon permanent
being. But motion itself is otherwise and distinctly conceived,
so that perception is defined in terms of being, and as -within a
world. Such is clearly the case with all definitions of the per-
ceptual realm in terms of so-called ' secondary qualities.' Where
the motive of the physical sciences is the determining one, and
this is very commonly the case, the world gets itself divided into
the physical and the psychological realms, the former being

1 The Renaissance, pp. 247-248.

2 James, Varieties of Religious Experience, p. 63.

CONCEPTIONS OF CONSCIOUSNESS. 291

employed as the standard and defining world. And here a
subsequent reduction of knowledge to psychological terms is
evidently contradictory.

The perceptual idealism of Berkeley announces subjectivity
as an ontological, and not merely an epistemological principle.
The famous dictum, « esse est percipi,' is the ontological
counterpart of the more ancient dictum, « flfo-ciov %fjrl/IdT(oi>
fisTpov dvdpatxoz.' But it appears shortly that to be is rather to
perceive, or to cause to perceive. The soul and God are the
real terms of the perceptual relation, and they are themselves
revealed in another order of cognition. Berkeley's later tendency
to abandon his perceptual idealism for one of the Platonic type
is well-known, and emphasizes his inability to make an objective
order out of the psychological realm. But he persisted in this
course so far that he made content and subjectivity coextensive,
and was then under the necessity of adding the objectivity all
at once and abstractly. The same necessity is interestingly
exhibited in the case of J. S. Mill, whose category of * possible
experience ' functions similarly as objectivity conceptually and
artificially superadded to a content that has been stripped of it.

Sensationalism in its other modern and contemporary phases
scarcely warrants serious treatment. It commonly defines
sensations as events within a physical world, and then gravely
announces that these sensations, as the simplest terms of intro-
spective analysis, are the ultimate beings. But the perfection
of this contradiction is enlightening. The sensation is the
quintessence of relativity. It signifies objectivity at a minimum
and subjectivity at a maximum. Simple pressure, or the lonely
and unrecognized sound, are the first dawning or the last wan-
ing of objects. But they are such vanishing points only in the
light of their all but entire inadequacy. In themselves these
pulses of experience are objective, and are remarkable only
when we come to consider the great degree of their deficiency.
Sensationalism means the attempt to define being in terms of
what it is not. Indeed, such a plan is virtually announced in
the language of all relativists. The Protagorean proposition
stated ontologically would read : all things are the human
measure of them, which contains the same substitution of a

292 RALPH BAR TON PERR Y.

passive for an active or neuter verb that is remarkable in the
Berkeleyan principle. But any account of being in terms of
another than itself is as unprofitable as it is contradictory.

The transcendental idealist would doubtless regard the dis-
cussion up to this point as a stage in the development of his own
argument, and he must now be reckoned with quite independ-
ently. His doctrine is established with direct critical reference
to psychological idealism. The impossibility of defining ob-
jects in terms of relativity is allowed to conduct the thinker dia-
lectically to the conception of the absolute. The sequel to my
error or exclusiveness, is truth or inclusiveness. The outcome
of this dialectic is determined by the symmetry of the antithesis.
Corrected experience implies a last correcting experience ;
partial cognition, complete cognition ; empirical subject, a
transcendental subject ; finite mind, an absolute mind. Hence
being is definable as for a standard, complete, transcendental or
absolute consciousness. Now it is evident that the validity of
this reasoning depends upon the degree to which the limiting
adjective determines the meaning of the substantive. If con-
sciousness means limitation, then absolute consciousness is a
phrase but not an idea. Where consciousness is recognized as
relative, what does it mean apart from that relativity? This
question has remained unanswered so far as transcendentalism
is concerned. But if consciousness known as experience rela-
tive to a point of view, is not defined save in terms of that cir-
cumscription, then to retain the concept of consciousness for a
realm defined as free from just that factor of circumscription, is
sheer absurdity.

Let us consider briefly the Kantian foundation of transcen-
dentalism. The Critique of Pure Reason taken as a whole,
informs us that the object, so far at any rate as knowable,
can be neither inside nor outside of my private consciousness.
The dilemma is solved by defining the object as apperceived
by a transcendental ego which is the ideal cognitive subject
logically immanent in my consciousness. This subject remains
for Kant a law of my consciousness and thus dwells in
that logical realm which is neither soul nor nature, until it
enters into the real world under the form of faith. But its

CONCEPTIONS OF CONSCIOUSNESS. »93

relation to the realm of knowledge is such as to define nature
as phenomenal on the ground that it falls between the unthought
world on the one hand, and the completely thought world on
the other; between the residual objectivity of the perceptual
experience, and the ideal objectivity of the conceptual exper-
ience. There is too much or not enough of consciousness in the
natural world to permit of its being a world of things ; too much
because space and time are merely subjective necessities, and
too little because these forms of perception are such as prevent
the realization of the ideal of subjectivity itself. The phenom-
enal realm is distinguished from such a realm as would have
its being independently of thought, and such a realm as would
have its being in the perfection of thought. So far as theory
is in question Kant leaves us here. For the post-Kantian who
wishes to define a metaphysical doctrine, there are two possi-
bilities. He may conceive that successful thought ceases to be
distinguishable as thought, and therefore realizes being as
independent of thought ; or he may conceive that successful
thought is still essentially thought and therefore realizes only its
own consummation. Kant's phrase ' intelligible intuition ' per-
mits either interpretation ; the former is the way of realism of
the Platonic type, the latter the way of post-Kantian idealism.
But the only account of mind that is offered even by the
idealist is an account in terms of its practical function with
reference to the things which it seeks to evisage. As in
Kant's delimitation of the realm of the internal sense as psy-
chological, the realm of physical or external experience becomes
for the moment a realm of things ; so in any delimitation of the
phenomenal world as a whole, the noumenal world becomes a
realm of things. Now if I define my real world to escape my
subjectivity why should I call it my transcendental self? Kant
himself refused to do it when he maintained that the logical
subject, or transcendental ego, was not a real. I might as well
call it an objective subject, or an absolute relative. The con-
tradiction is only thinly disguised in the common language of
idealism. This theory finds no difficulty in an absolute point
of view (or Absolute's point ofview), as though anything ab-
solute could be a point of view at all. To transcend my point

294 RALPH BAR TON PERR Y.

of view, I am to employ a transcendent point of view. Since
in knowledge I must escape subjectivity, I resort to a supreme
subject. It is like defining riches as transcendental poverty, or
satiety as transcendental hunger. Suppose an orifice through
which light shines upon a wall : the disk is then due to the ori-
fice. Remove the orifice, and the generally diffused light is due,
according to the transcendental idealist, to a transcendental
orifice.

But possibly we do this type of idealism an injustice through
not advancing in its behalf the direct and positive argument for
consciousness in consideration of its synthetic function. This
argument is sufficiently obscure to make one fearful of stating
it in behalf of another ; but it seems to mean that truth is a
gathering up, systematizing, or relating of terms, and that such
is exclusively the property of thought. Now I may see the
logical evidence for a connection without seeing any evidence
for the dependence of that connection upon my seeing. My
judgment does not attest its own indispensableness. Only a
later judgment can so define my first judgment as a judgment
at all. The judgment so discovered has, moreover, an indi-
viduality or numerical uniqueness that forbids the definition of
its object in terms of it. Were the triangle constituted by the
defining thought of it, there would be a triangle for every such
judgment, but no such thing as a triangle. That truth is a syn-
thetic activity of thought must be a psychological truth, i. e., it
has reference to my access to truth rather than to truth itself.
It is biographically true that when I apprehend a law, or prin-
ciple or definition, I comprehend a number of terms together and
in relation. I reach the truth by combining, as, notably, in the
case of my knowledge of similarity. But it would be folly to
claim that therefore things are made similar by their combi-
nation in my experience. Things are not made similar by
seeming similar. In seeming similarity there is doubtless a
peculiar unity. Two similar seemings will not make a seeming
similarity. But this has to do with the peculiar relational char-
acter of the psychical manifold, and not with the truth of simi-
larity. It is true, of course, that a succession of feelings is not
a feeling of succession, but this does not point at all to the

CONCEPTIONS OF CONSCIOUSNESS. 295

dependence of the former upon the latter. It is the transcen-
dentalist's favorite complaint against the empiricist that he con-
fuses psychology with logic, but his own arguments for ideal-
ism turn upon this very confusion. His psychology of thought
is an improvement upon the crude associational theory, but they
are none the less psychology. And in the metaphysical use of
his theory he identifies the object of knowledge with the know-
ing. He makes being out of the psychology of logic, and by
a dialectic that is in this respect essentially indistinguishable
from that of the sensationalist, he defines the real in terms of
that activity, purposiveness, or category of objectivity which
he regards as the most important factor of the knowing state.
Indeed, he quite frankly acknowledges that metaphysics and
psychology coincide in the conception of the self. There is
space here for only a brief independent consideration of this con-
ception, but sufficient to do justice to its serviceableness as a
general ontological principle.

Self-consciousness is introduced to terminate the series of
relativities defined by a perceptual idealism. If A be for /?, B
for C, and C for D, there must eventually be an Mt such that
while A, B, C and D are for M,Jlfis for itself. The difficulty
here centers in the proposition, ' A is for itself,' which for our
critical purposes we may treat in a purely dialectical manner.
If there be no difference between M and * itself there can be
no relation between them except that of identity, M is M, which
is the category of the thing. If M and ' itself ' are not alike,
then Ml is for Af2, and the original perceptual series is pro-
longed interminably, or Afl — M2 must be regarded as a unique
and organic relationship itself constitutive of a new thing IV,
which itself does not derive existence from relation to a mind.
So we must either content ourselves with a world that is phe-
nomenal and face the contradiction that is virtually contained
in such a proposition, or consent sooner or later to regard the
terminus of thought as a thing not constituted by that thought.
And such a consent is in reality prior both temporally and logic-
ally to the conception of subjectivity. The error here is sub-
stantially the same as that which lies at the root of the other two
transcendentalist arguments ; the terms of psychology are mis-

296 RALPH BARTON PERRY.

applied to a totality of which by definition they signify only an
abstracted aspect. The term consciousness has reference to
relativity and exclusion within a world of reals, and therefore
cannot signify a principle constitutive of that world itself.

It is the chief interest of faith that certain values shall sur-
vive and be consummated. If consciousness be either a specific
and unique kind of thing, as certain so-called ' spiritualistic '
philosophers would have us believe, or a general form of all
being, it cannot be centrally important in such an issue. But
if taken to signify selection within the realm of things, then,
though it cannot be the ontological first principle, yet as the
most general category defining a self it will apply either to psy-
chology or the religious aspect of metaphysics. It must be
admitted that error is an outstanding problem. But that
circumstance is at least equally difficult for the subjective
idealist. Grant him his absolute subject, and finite experiences
with their relativity and exclusiveness are a totally new problem,
which the general and innocuous pervasiveness of consciousness
does nothing to solve.1

1 The MSS. of this article was received on April 12, 1904. — ED.

RETINAL LOCAL SIGNS.

BY DR. WALTER F. DEARBORN,
New York.

RETINAL LOCAL SIGNS.'

While the existence of some sort of local sign is generally
regarded as almost self-evident, the hypothesis regarding their
exact nature vary materially not only in their account of the
constituent elements, but also in their estimation of the relative
significance of these elements. Lotze, in his initial discussion
presented, as is well known, three different hypotheses. Under
all the various modifications, however, he conceived the local
sign as a motor consciousness supplementary to the original
retinal impression. His first two hypotheses are alike in their
general nature in that each presupposes a physiological mech-
anism by which the stimulus from each point of the retina is
transferred to the nerves of the orbital muscles. According
to the first hypothesis the stimulation of each point through an
' interweaving ' of the nerve fibers from the surface of the retina
and the ocular motor nerves, causes an eye movement definite
enough to bring the fovea immediately to the point of excita-
tion. In the second hypothesis the retinal points are not sup-
posed to have such an exact coefficient of movement, but with
each point is associated only sufficient motor impulse to start a
movement in the appropriate direction. Lotze regarded this
latter hypothesis as the more probable. As a third hypothesis
a theory of association is merely outlined and rejected without
further elaboration.

Recent experiments have greatly increased the improbability
of Lotze's second hypothesis, *'. c., of a continuous succession of
motor impulses arising from a successive stimulation of the
retinal points between the point of original excitation and the

1 Read in part before the Section of Anthropology and Psychology of the
New York Academy of Sciences.

»97

298 WALTER F. DEARBORN.

fovea. It is now pretty well established that except for move-
ments in the vertical and horizontal planes, the eye does not
move in straight lines, and, in the second place, it has been
shown l that during continuous eye movements from one point
of regard to another there is no new effective stimulation of the
retina which could occasion the successive motor impulses.2

The purpose of this article is to present some new experi-
mental data as a contribution to the discussion of the first Lot-
zean hypothesis which still appears as a more or less important
factor in practically all current accounts of retinal local signs.
In a report of their experiments on the angle velocity of the
eye, Dodge and Cline called attention to certain errors in the
immediate fixation of eccentric visual stimuli. As these inac-
curacies evidently concur the normal functioning of the motor
impulse it seemed worth while to determine their extent and
frequency.3

As a result of preliminary experiment the position of the
stimulus to movement was arbitrarily fixed in the first series of
measurements at forty degrees to the left of the primary fixa-
tion point. The stimulus itself was a bright point of light
about eight tenths of a millimeter in diameter, made by expos-
ing a ground-glass incandescent bulb behind a perforated
screen. The observers A and B were respectively Professor
Dodge and the writer. The ratio between the angular displace-
ment of the eye and the displacement of the lines on the photo-
graphic negative was determined empirically. Since these
measurements were concerned with errors occurring only in the
vicinity of forty degrees to the left of the primary line of regard,
it was necessary to determine the value on the negative of some
unit of movement between thirty and forty degrees. Under
the conditions maintained in the succeeding experiments the
lines representing on the negative five-degree movement of the
eye between 30°, 35°, 40° had the following values :

1 PSYCH. REV., Vol. III., pp. 454-465.

2 Ibid., Vol. VIII., pp. 145-157.

3 The experiments were performed at the Psychological Laboratory of
Wesleyan University and were undertaken at the suggestion and under the gen-
erous cooperation and supervision of Professor Dodge. The registering appa-
ratus used was a modification of the Dodge-Cline photographic camera described
in this REVIEW, Vol. VIII., pp. 147-151, and in its recent form in the American
Journal of Physiology, Vol. VIII., pp. 308-310.

RETINAL LOCAL SIGNS.

199

For A the average of twelve five-degree movements was .54
mm., with a M.V. of .039 mm., i. *.,

i° = .108 mm.

For ^the average of seven five-degree movements was .5 mm.
with a M.V. of .06 mm., *'. e.,

i° = .10 mm.

The following table, Table I., shows the frequency and ex-
tent of the corrective movements occurring in the movements
of the right eye through forty degrees (/'. *., from approxi-
mately the primary position of the eye to a point of stimulation
forty degrees to the left of the primary fixation point). They
are in some cases positive', *'. e., the first movement of the eye
fell short of the point of stimulation, making necessary a sup-
plementary movement in the same direction, and in other cases
they are negative. The positive movements are denoted by
plus signs and the negative by minus signs on the right of each
column.

TABLE I.

TABLES SHOWING THE EXTENT OF THE CORRECTIVE MOVEMENTS MADE BY

THE EYE IN MOVING FROM THE PRIMARY FIXATION POINT TO

A POINT OF STIMULATION 40° TO THE LEFT.

I.

Corrective Movements of A.

II.

Corrective Movements of B.

I.

.20 mn

2.

oo

3-

• 14

4-

.12

5-

OO

6.

•13

7-

.10

8.

.27

9-

oo

10.

.18

ii.

•13

I.

oo mn

2.

oo

3-

.14

4.

.12

5-

.19

6.

00

7-

oo

8.

.13

9-

.24

10.

.12

n.

.26

12.

.22

Average, .116 mm. (= i° 42') Average .118 mm. (= l° lo/48")

Summary.1 No. Per Cent. C. Mvts. No. P. & N.

A
B

II.
12.

72
66

M.

V.

S+, 3 —

.116 mm. (= i° 42')
.118 •« (=i°io/4

A .056 mm. (=31' «")
-5.079 " (=47/40//)
1 No. = number of eye movements. Per cent. C. M vts. = percentage of
corrective movements. No. P. & N. = the number of positive and nexadve
movements. M. = The average or mean extent of corrective movments. M,
V. = Mean variation iti the extent of corrective movements.

300 WALTER F. DEARBORN.

One corrective movement is omitted from the above table of
B. It is clearly an abnormal break and divides the whole
movement into about two equal parts. Its extent is 1.98 mm.,
and thus represents an eye movement of about twenty degrees.
If it entered into the computation, the average error for B
would become .24 mm., z. <?., an average corrective movement
of 2° 24'. Those instances in which the corrective movement
is given in the table as zero must be interpreted in view of the
limitations of the physical measurements of small differences,
more or less exaggerated by certain pecularities of our photo-
graphic negatives. The cathetometer read only to .02 mm.
making it impracticable to measure any movements of the eye
half a degree or less. Moreover, with our present knowledge
it is impossible to distinguish minute corrective movements from
the slight eye movements found in all attempts to maintain fixa-
tion. Arid, finally, errors due to inaccuracies of definition on
the negative which become evident only when the lines are
magnified, lead to a probable error in all cases. While this
would probably be evenly distributed positively and negatively
in the cases actually measured, it makes it impossible to meas-
ure minute angles, and consequently renders the mean value
that is given slightly too small.

In discussing further the above results, the question arose
whether they might not in some measure be modified by short-
lived motor habits which the eye seems to acquire with more or
less facility. In order to determine the matter a second series
of experiments was made in which the angle of movement was
varied. In the process of refocusing the relative position of
the camera and the source of light was somewhat changed, as
was also the distance between the lens and the photographic
plate. These changes necessitated a redetermination of the
empirical ratio between the angular movement of the eyes and
the displacement of the lines on the negative. The stimulus
was in this second series exposed at one of three points — at
20°, 30° or 40° — instead of at only 40° (as in the first series).
The ratio between the angular displacement of the eye and the
.lines on the negative was determined for the angle of 30°, 40°.
In accord with the results thus obtained .106 mm. in the case

RETINAL LOCAL SIGNS.

301

of A, and .107 mm. in the case of D, is used as the equivalent
of a movement of the eye of one degree at forty degrees from
the primary fixation point.1

The results are given in Table II. Under A the measure-
ments of the corrective movements of A are given in three
columns, under a the 40°, under b the 30° and under c the 20°
movements and similarly under B the measurements of B are
given.

TABLE II.

A.

I.
2.

3-

4.
5-

6.

a (Stimulus at 40°).
.16 mm. +

.16

.19
.22

.20

.22

-f

+
+
+

+

b (Stimulus at 30°).

c (Stimulus at *>°).

I.

.12 mm. —

I.

.05 mm. -f

2.

oo "

2.

•30 " +

3-

oo "

3-

00 "

4-

.14 "

4-

.28 " —

5-

00 "

5-

.14 " +

6.

00 "

Average .191 mm. =1° 48//. Average .043 mm. Average .154 mm.

a (Stimulus at 40°).
i. .29 mm. -f-

b (Stimulus at 30°).

2.

3-
4-
5-

6.

•34
•36
.28

.36
.29

Av. .32 mm. =2°59/24//.

c (Stimulus at ao°).

I. oomm.-r

(a)2. .07 " +

.18 " +

3- -14 " +

(a)4. .08 "

.10 •• -f

5- -09 " +

Av. .14 mm. Av. .134 mm.

I.

.23 mm,

.+

2.

.08 "

—

3-1

.21 "

-f

.10 "

+

4-

•13 "

+

5-

.18 "

+

6.

.24 "

+

7-

OO "

8. «

.14 "

-f

.12 "

—

Disregarding the error of movement when the stimulus was
20° and 30° from the primary fixation point, a summary of the
40° movements is given in Table III.

'The mean variations for both these units of measurements was (M.V.)
.009 mm.

2 In computing the total error of the double corrections, I have added their
absolute values and determined the variation of this sum from the mean rather
than the variations of the single corrections from the mean.

8 For abbreviations see note on page 299.

302 WALTER F. DEARBORN.

TABLE III.

Subject. No. Per cent. C. Mts. M. M.V. No. of + or — .

A 6 100% .191 mm. ( = i°48//) .023 mm. 6 +

B 6 100% .32 " ( = 2°59'24") .036 " 6 +

These tables show, as was expected, a somewhat increased
inaccuracy in fixation in the forty-degree movements and tend
to substantiate the hypothesis that the earlier results were modi-
fied by the successive eye movements of the same angular dis-
placement. At the same time they introduce some new factors
into the problem. The unexpected accuracy of fixation in the
thirty-degree movements — especially those of A in comparison
with the 20° movements presents a phenomenon whose explan-
ation is impossible without further investigation. Secondly,
double corrections, i. e.t two separate corrections in one move-
ment, appear twice among the twenty-degree movements, and
twice among the thirty-degree movements. In two of these
cases both movements are positive, but in the other two the cor-
rection is made by one positive and one negative movement.
This is interesting as showing the great variability of eye move-
ments.

In order to compare the inaccuracies of motor innervation
with the threshold value of local discrimination, it was necessary
to measure the latter for both subjects as exactly as possible.
For the sake of more general comparison both the usual method
of the discrimination of two points, and a second method, which
will be described later, were used. The two points of light
employed according to the first method were about 8/10 mm.
in diameter, 60 cm. from the axis of rotation of the eye, and at
a maximum of 40° to the left of the primary line of regard.
They were exposed either simultaneously or but one at a time,
and the subject was asked to state at each exposure whether he
saw one or two points. The results were in brief that in a series
of ten exposures when the points were 5 mm. (or 28'35//)1 apart,
they appeared as one to both observers. When 71/4 mm. or 41'
27" apart in a similar series of ten they could be distinguished

1 As the distance from the axis of rotation of the eye to the point of fixation
was 60 cm., the circumference of the circle of which this is a radius was 3771.12
mm. This gives 10.47 mm. as the equivalent of i° on the circumference of
the perimeter, or i mm. = — 5'43".

RETINAL LOCAL SIGNS. 303

as two by A but not by B. When 10 mm. or 57' 10" apart they
were distinguishable by B. In some cases, before the thresh-
old was reached, the subject reported that the single point of
light seemed to appear broader than at other times. These
cases of course, show the proximity of the threshold.

The thresholds therefore lie between 28' 35" and 41' 27"
for one observer (A) and 41' 27" and 57' 10" for the other ob-
server (B). These numerical limits might have been more
exactly defined but the results of the second series of experi-
ments seemed to make this unnecessary. The latter series
depends upon what appears to the writer to be a more satisfac-
tory method of measuring the fineness of retinal discrimination.
Stern l determined the width of a just perceptible black line
dividing an otherwise continuous whole surface, and regarded
the result thus obtained 15" as a measure of the threshold of
space discrimination. But it seems very doubtful whether this
is a measure of spacial discrimination at all ; it seems rather to
be a measure of the least intensity or smallest extent of the
stimulus necessary to produce a sensation of blackness, and
would doubtless vary for different colors. Gilbert2 in repeating
the experiments under somewhat modified conditions, reduced
Stern's measurements to 2.5". Stratton5 has measured the
threshold by means of exposing motionless points of light in
immediate succession one above the other. The disturbing
effect of irradiation was thus largely eliminated. His results
for the angle of 30° are as follows :

Subject. Angle. No. of Cases. Length of Arc Discriminated. M. V.

A. 30° 5 29> 5-2'

Bd. 30° 3 18.3' 2.a/

P. 30° 4 63.7' «.?'

In my experiments, a continuously moving point of light was
employed and the smallest extent of movement which gave a
definite clue to its direction indicated the threshold value. It
has been considered by Stratton a possible objection to his ex-
periments that although the points were actually motionless and
only exposed in succession they produced the appearance of

^Zeitsch. f. Psych, u. Phys. d. Sinnes., VII., 321.
* Psych. Rev., IX., 435.
*Ibid., IX., 436.

304 WALTER F. DEARBORN.

continuous motion. Since there was, actually, however, no
objective motion and the effect was due purely to suggestion,
he considered that the subjective illusion might be disregarded.
The real justification of the method, however, seems to me not
the fact that there was no motion involved, but rather that some
form of local discrimination was distinctly included in the
process. The important consideration is not the absence of
movement, but the recognition of a definite change in position.
Stratton has shown that there is no reason for considering these
two processes, /. e., the perception of movement and the dis-
crimination of position, independent.1 For our purpose, how-
ever, it is sufficient to point out that, whether the processes are
dependent or independent, if observation is always made not
only of the fact of movement but of the direction of movement,
some element of local discrimination must be involved.

The stimulus used in these experiments with a moving point
was, as before, the light of an incandescent lamp with ground-
glass bulb which was exposed through a small aperture (8/10 mm.
in diameter) cut in a moving slide. The extent of the move-
ments of this point of light was limited by transverse slits,
which were one, two, and three millimeters in breadth. By
means of these a movement of the point of light of from one to
three millimeters in length could be made at the option of the

1 His conclusion, that the perception of motion is simply a perception of
the fact that ' a sensation is changing its space relations, ' has however hardly
been established. The results clearly contradict the evidence upon which
Exner's theory is based and leaves that the more improbable theory. It is
recognized, however, that in all such experiments sensations of movement
occur without any perception of direction. In the present experiments this
was the case in several instances despite the fact that the observer was attempt-
ing to discover any clue that he might of the direction of the given movement.
Observations of a similar character were made by Stanley Hall and Donaldson
with relation to tactual discrimination (Mind, 10, p. 571) : 'there is no doubt
whatever that a distinct sensation of motion occurs without giving any impres-
sion of direction in many cases.' This may be due simply to a lapse in atten-
tion, or a return to the more habitual reactions, that is, as in the case of these
experiments, we are much more accustomed to observe small movements with-
out consciously noting their direction, or it may be due to other causes which
may easily be suggested. But it has not been shown that any of these more
evident explanations will cover all the facts of the case. The processes in-
volved are evidently complex, and need further investigation and analysis
before the relation of the sensations of motion to those of direction may be
established.

RETINAL LOCAL SIGNS. 305

operator. The contrivance also admitted of the movements
being made in either the vertical or horizontal planes. It was
found by this method that at forty degrees to the left of the pri-
mary fixation point the discrimination of the direction of a single
point of light in motion is from seven to ten times greater than
the discrimination of two discrete points.

The length of the first movement experimented with was
2 mm. (n' 26"). The exposures were made in three series,
each of eight to twelve exposures ; in the first the subject was
told that all movements would be in the horizontal plane,
in the second, that all movements would be in the vertical
plane, and in the third, in either the horizontal or vertical planes.
The direction of all the movements in the first series was cor-
rectly perceived, in the second series three vertical movements
were reported to be horizontal by A and one by B. But, even
when the movement was recognized as being in the vertical
plane, both subjects questioned in several cases the accuracy of
their further judgment of the direction, *. e., whether up or
down. In the third series, all the horizontal movements were
again correctly given by both subjects, but the vertical move-
ments were reported horizontal by A, and in the case of B two
out of four vertical exposures appeared as horizontal.

When the movement was decreased to i mm. (5' 43"), the
horizontal movements were still reported correctly, but the ver-
tical were variously stated as movements from right to left or
left to right, as oblique — e. g*., from upper right to lower left
— although the subject knew that the apparatus in use did not
admit of such movements being made, as a mere appearance
of light without motion and as a movement whose direction was
not perceived. Small changes in the velocity of the movement
also affected the ability to perceive the vertical movements. As
these movements did not, however, primarily concern this inves-
tigation, no attempt was made to determine the most favorable
rate of movement. The appearance of these factors is evidence
that the threshold for vertical movements has been reached
more quickly than the threshold for the horizontal plane. The
latter is obviously below i mm. (5' 43"), but the form of appa-
ratus did not admit of smaller movements in this plane.

306 WALTER F. DEARBORN.

Several peculiarities were also noted which were somewhat
characteristic of these movements in either plane.

First, the movements, even when correctly interpreted, often
seemed to both observers like the passing of some opaque
object behind the aperture. Secondly, the movements appeared
to be larger than they really were, and finally, B especially
noted occasionally an act of judgment in which the direction of
the movement seemed to be inferred from changes in the size
of the stimulus as it came into view and disappeared. Obvi-
ously, however, all of these peculiarities must depend on some
real differences in local coloring.

A similar series of experiments was made with a different
form of stimulus. Instead of a point of light a piece of white
carboard, 4 cm. square was used. It could be moved as the
former stimulus with exactness in either the vertical or hori-
zontal planes. With this method there were no marked pecu-
liarities evident, such as have just been discussed, the former
upper limit of value of local discrimination was somewhat
reduced in both the horizontal and the vertical movements.
Both observers perceived the direction of all movements over
0.9 mm. in extent. As this was the smallest movement experi-
mented with, the threshold value must be even lower.

This refinement of special discrimination (5' 8") has been
approached only by the questionable method of Stern. The
results of Stratton, as stated above, show 18.3' as the lowest per-
ceptible area of movement at ten degrees nearer the primary
point of regard that in our experiments.

The noticeable disparity between the results of our two
methods, i. e., as between the methods in which the threshold
is determined with the aid of movement and that in which
motionless points are used, does not admit of ready explanation.
It might seem quite plausible a priori that a high degree of deli-
cacy in the discrimination of two points might be brought about
by means of repeated observations of changes in position. Un-
doubtedly the irradiation of light and the lack of sharp defini-
tion must be taken into account, but just how far this operates
to lessen the discrimination of adjacent points is undetermined.1

1 The similarly increased discriminative sensibility in the case of motion
on the skin was first noticed, I believe, by Stanley Hall and Donaldson, Mind,
X., 563-

RETINAL LOCAL SIGNS. 307

A summary and comparison of the different parts of this
investigation leads to the following conclusions. It has been
shown in the first place that the mean extent of corrective
movements of the eye in fixating a stimulus forty degrees from
the primary fixation point was, when no effort was made to
counteract the effect of short-lived motor habits, for A i ° 42'
and for B i° n' ; when the attempt was made to prevent the
formation of such habits, the mean extent of corrective move-
ments was larger, for A i° 48', and for B 2° 59'. In the
second place, it was found that the threshold value of local dis-
crimination as determined by method I. was for A between 28'
35" and 41' 27", and for B between 41' 27" and 57' 10" ; as
determined by method II., 5' 43" or less for each observer, and
by method III., 5' 8" or less.

Even under the most unfavorable circumstances therefore
(Method I.) the local discrimination at 40° from the fovea in
the horizontal plane is much finer than the adjustment of motor
impulses as shown by the mean extent of corrective movements
made in fixating a point at the same distance (40°) from the
primary point of regard. That the delicate local differences
shown by methods II. and III. could result from such grossly
inaccurate motor impulses seems out of the question.

These discrepancies between the accuracy of the motor im-
pulse and the delicacy of local discrimination necessitate, as I
believe, some modification of the traditional view in regard to
the nature, or at least in regard to the relative importance of
the motor factor. Moreover, the variability of the latter, as
shown by the presence of corrective movements of varying
magnitudes, is altogether in contrast with the relative constancy
of special discrimination.1

1 The Mss. of this article was received on March 4, 1904. — ED.

STUDIES FROM THE CALIFORNIA PSYCHOLOGI-
CAL LABORATORY.

VI. SOME PECULIARITIES OF FLUCTUATING AND OF INAUD-
IBLE SOUNDS.

BY KNIGHT DUNLAP, PH.D.

I. THE EFFECT OF PHYSICAL INTERRUPTIONS IN
SUBLIMINAL PHASES.

Eckener 1 states that ' subjective ' fluctuations in a minimal
sound may be readily distinguished from ' objective ' inter-
ruptions, and in particular, that in phases where the sound is
inaudible owing to the ' subjective ' fluctuation of attention, an
' objective ' or actual physical interruption may be perceived if
it occurs. To test this phenomenon, I conducted some experi-
ments in the spring of 1899, in the course of which some inter-
esting developments appeared.

The sound employed in these experiments was that emitted
by a telephone receiver in circuit with the secondary element of
a DuBois-Reymond induction coil, the primary current of
which was supplied by two gravity cells and interrupted by an
electric diapason of one hundred double vibrations. The tele-
phone receiver was placed close to the left ear of the subject,
who sat in a ' silent room ' from which external noises were
excluded by padded double doors and specially constructed
walls. By means of a noiseless key controlling an electric
circuit the subject operated an indicator in the experimenter's
room ; either a sounder or a kymograph marker, as the case
might be ; and by a prearranged code of signals was thus able
to give his reports. By means of a key the experimenter could
at any time break the secondary circuit, and thus silence the
telephone receiver without in any way interfering with the

1 ' Untersuchungen iiber die Schwankungen der Auffassung minimaler Siu-
nesreize,' Philos. Studien, Bd. VIII., S. 365.

308

FLUCTUATING AND INAUDIBLE SOUNDS. 309

primary circuit. The diapason, induction coil, kymograph, and
all accessory apparatus, were in the experimenter's room.

In the first set of experiments the subject's key was con-
nected with a sounder, and he signalled the instants of appear-
ence and disappearance of the telephone snarl by appropriate
clicks. By varying the position of the secondary coil an inten-
sity was found at which the fluctuations occurred well, and then
in certain of the intervals indicated by the subject as * empty,'
(*. £., in which the sound was inaudible), the sound was phys-
ically interrupted for a brief period. If the subject perceived
this second death of the sound, he indicated it by a special
signal.

Table A gives the numerical aggregates of the results ob-
tained in this way. The first column of figures gives the total
number of cases for each subject in which the sound was
physically interrupted in an apparently * empty ' interval ; the
second column gives the total number of cases in which the
subject indicated such interruption as having been perceived ;
and the third column gives the number of cases in which each
subject indicated a physical interruption when really no such
interruption had occurred. The third column is then a record
of errors.

TABLE A.

PHYSICAL INTERRUPTIONS AFTER PSYCHOLOGICAL DISAPPEARANCE

OF SOUND.

Subject. No. Made.

S. 23

R. 9

G. 47

W. 25

A. 10

Bi. 75

Bo. 17

In the second set of experiments the conditions were the
same as in the first set, except that the subject's key was con-
nected with a stylus writing on the drum of a kymograph, and
the operator's key for breaking the telephone circuit was con-
nected with a second stylus writing directly over the first one.
Thus the actual course of the experiment was recorded for

No. Perceived.

No. Imagined.

IO

I

3

0

24

0

17

3

4

i

34
6

10
0

No. Perceived.

No. Imagined.

32

7

9

0

5

0

109

27

10

3

18

0

3 1 0 KNIGHT D UNLAP.

leisurely examination. The results of this set are given in
Table B.

TABLE B.

INTERRUPTIONS AS IN TABI,E A.

Subject. No. Made.

S. 55

R. 16

A. 69

C. 123

Bi. 27

G. 29

The length of the physical interruption was about 15^ sec.,
while the length of the average ' empty ' interval was over four
sec., so the apparent perception of the physical interruption was
not due to a failure of the sound to reappear at the expected
time. Such expectation, again, could hardly operate, owing to
the great irregularity in the period of fluctuation ; for so it would
have produced a large percentage of reports of interruptions
when none occurred.

That the perception of the interruption was genuine in almost
every case is evidenced by the small number of false reports, in
conjunction with the fact that in the majority of empty intervals
no physical interuptions occurred, thus rendering large the op-
portunities for errors of imagination. Some of the errors were
undoubtedly correct judgments delayed in the registration on
account of hesitation of the subject to accept his experience as
real. This delay occurred in several cases concerning which
the subjects were questioned, but usually there was no oppor-
tunity of making inquiries until after the exact occurrence had
passed from the subject's recollection. The recorded errors in
almost every case occurred in empty intervals in which a
physical interruption had been made a little earlier and not re-
ported, but in obtaining the data of Table B. an arbitrary rule
was adopted, according to which no report of an interruption
was accepted unless recorded on the drum within one millimeter
(three fifths of a second) after the break in the current. The
majority of the accepted reports were however recorded within
one half of a millimeter after the break. It should be noticed
in regard to subject C. in Table B, that almost all of his errors

FL UC TUA TING A ND IN A UDIBLB SOUNDS. 3 1 1

were made in the early part of the work, and few after he be-
came accustomed to the conditions of the experiment.

Three of the subjects made rather definite analysis of the
conditions attending the perception of the second death of the
sound. R. was so surprised upon receiving the interruption for
the first time (having previously been given no information at to
the nature of the phenomenon sought), that she stopped the ex-
periment to describe it, and even after she had become familiar
with it could not repress a feeling of astonishment at each oc-
currence. It finally became clear to her, however, that while
just before the physical interruption she was sure that she was
not hearing the sound, as soon as the interruption occurred she
felt from her new point of view that she had been hearing it all
the time, but had only now realized it.

Quite different was the report of C., who was just as sure
after the interruption that he was not hearing the sound as he
had been before. The occurrence was not for him the cessation
of a sound, for that had absolutely disappeared previously. It
was rather the * bottom dropping out ' of some indefinite ele-
ment in consciousness ; to be compared to nothing except the
feeling attending the downward start in a rapid elevator. The
experience of S. was similar, but of different characteristic.
For him the occurrence was largely a feeling of relief, as
though the sound had been absent previously, but that now he
felt sure there was no further use in looking for it. Sometimes,
however, his experience would incline towards that of R., and
he would feel that he had really heard the sound all the time.1

This experiment shows conclusively that a minimal sound
which has become imperceptible through the so-called fluctua-
tions of attention, may yet be ' heard to stop.' But a yet more
striking phenomenon was observed by subjects S., R., A., Bi.
and C. The sound was in this case made so weak that it was
not perceived at all, and was then physically interrupted at
irregular intervals ; the result being that in a large number of
cases the interruptions would be correctly reported by the sub-
ject, who however, in the typical cases did not perceive any-
thing corresponding to the commencement of the sound at all.

1 See Stratton, Experimental Psychology, pp. 9°"9i-

312 KNIGHT DUN LAP.

That is, nothing was perceived except the stoppage of the im-
perceptible note. Two of the records of single tests on this
part of the experiment are given in Figs. i. and 2, in which the
depressed portions of the upper line indicates periods of subli-
minal stimulation, and the elevated portions indicate no stimu-
lation ; while perceptions of any thing connected with the sound
are indicated by breaks in the lower line, exclusive of the first

FIG. i.

break, which was made with the drum at rest to indicate the
relative positions of the two styli.

The relatively greater perceptibility of the stoppage, as com-
pared with the commencement, of the note, can not be due to
any instrumental peculiarity, since we are dealing here with an
alternating current of 100 alternations per second, in conse-
quence of which the interruptions by the key, which did not
disturb the primary current, would produce no special effects
upon the current at the moment of interruption. Moreover, a

FIG. 2.

similar relation was observed in using distinctly perceptible
sounds, in the following branch of the experiment.

Series were taken by commencing with a distinctly audible
intensity of the telephone snarl, and decreasing the intensity
slightly between each test and the following one, until it passed
completely below the threshold. During each test, which lasted
about twenty-five seconds, as in previous work, the sound was
interrupted, sometimes regularly with three-second intervals,

FLUCTUATING AND INAUDIBLE SOUNDS.

3>3

sometimes irregularly, the intervals in the latter case averaging
in the long run about three seconds. The subject indicated
these stoppages and commencements as accurately as possible,
by depressing his key at each beginning, and releasing it at
each ending, his reaction being recorded on the drum along with
the physical series.

After determining by careful tests that the latency of the
magnet markers was not measurably different for the two move-
ments, i. £., raising and lowering, careful measurements were
made of the distances between each stoppage or commencement
of the sound, and the registration of the same by the subject.
The delay at the beginning was then subtracted from the delay

• 3 .

Ffg.

3 O

at the ending, and the results for each interruption in a given
test averaged together. Averages for corresponding tests from
the other series for the same subject, reckoning backwards from
the point at which the sound passed below the threshold, were
then averaged together.

We have then for each subject, a series of numbers, positive
or negative, or both, indicating the average excess in the delay
in recording the ending of the sound over the delay in record-
ing the beginning, through various intensities from perfect dis-
tinctness down to complete disappearance. Such a series we
can represent by a graph, letting the ordinates represent the
values of the successive numbers in the series, and the abscissas

3 I 4 KNIGHT D UNLAP.

the points on the scale of diminishing intensities to which these
numbers correspond. It is evident that a negative ordinate, or
a negative value in the series of numbers represented by it, sig-
nifies that at that point the stoppage was recorded more quickly
than the commencement (cf. figs. 3, 4).

Figs. 3 and 4 show graphs derived from the results of sub-
ject S., which are characteristic of the results for all. It is
clearly evident from these that the stoppages are in general re-
ported more quickly than the beginnings, and that the difference
in this respect increases as the intensity of the sound diminishes.

Since there is no reason for supposing that the reaction time
•per se differs materially in the two cases, and since there is no
known aural condition which could account for this peculiarity,
the explanation is most probably to be referred to the general
conscious conditions governing the superior discriminative ad-
vantages of a position in silence harking back to sound over a
position in sound harking back to silence. The fact that con-
trast does play a large part in the perception of the weak sound
is further shown by the fact that in all cases a continuous sound
was found to be inaudible at a much higher intensity than an
intermittent sound under the same general conditions.

The cases of the second death of a sound referred to above
are most probably not different in character from the phenom-
enon just considered. A steady sound, inaudible continuously,
or intermittently through the ' fluctuation of attention,' becomes
perceptible at the instant of physical stoppage because of the
sharp transition from feeble sound to silence. The introspection
of the subjects referred to above favors this explanation de-
cidedly. Of course the neural conditions of the phenomenon
are yet to seek, but the psychological facts may be settled in
advance ; in fact they must be, before the neural facts may be
safely inferred.

II. THE FLUCTUATION OF DIAPASON AND GAS FLAME TONES.

W. Heinrich1 came to a remarkable conclusion that -pure

tones (i. e., tones unmixed with noise) do not fluctuate, but are

i'Zur Erklarung der Intensitatsschwankungen eben-merklicher optischer
und akustischer Eindriicke,' Bull. Internal, de VAcad. des Sci. de Cracovie,
Nov., 1898, 365-381 ; also, ' De la Constance de perception des tons purs a la
limite d' audibiliteV ibid., Jan., 1900, 37-45.

FLUCTUATING AND INAUDIBLE SOUNDS. 315

either heard continuously or not at 'all. This conclusion was
based principally on observation of the tone produced by the
singing gas flame. Titchener1 repeated Henrich's observations
briefly, and found that his results were apparently confirmed.
H. O. Cook,2 in Titchener's laboratory, had previously found
that tones of diapasons show the ordinary fluctuations quite
clearly.

The important bearing of these conclusions upon the whole
matter of the fluctuation of attention, so called, as well as upon
the special features of my own work, led me to investigate the
behavior of both diapason and gas-flame notes under the most
favorable conditions. This I did in the fall of 1903.

The subject was placed in the silent room mentioned above.
A lead pipe has one termination within a recess in this silent
room, and the recess is provided with a door by which it can be
noiselessly closed, either partly or completely. From the
silent room the pipe extends to an ordinary research room on
the other side of the building. In this second room was in-
stalled the diapason, or other source of sound waves. The in-
tensity of the sound in the silent room was regulated by varying
the distance of the source from the one termination of the pipe,
or by varying the position of the door over the other termina-
tion in the silent room, or by both. The subject, seated at a
fixed distance from the orifice, indicated by a prearranged code
of signals as before, the periods in which he heard the sound
and the periods in which he heard nothing.

An electrically driven diapason of 500 d. v. was first used.
The resulting note in the silent room, even when made very
loud, was not only surprisingly pure, but was practically simple.
The noise of the electric contact, as well as the overtones,
seemed to be practically eliminated by the transmission through
the long pipe.

The results on four subjects confirmed those obtained by
previous investigators ; the fluctuations clearly occurred. No
experiments on the second death of the tone were made, be-
cause of the difficulty of bringing the note to an abrupt termi-
nation.

1 ' Fluctuations of Attention to Musical Tones,' Am, Jour. Psy., XII., 595.
*/6id., XL, 119-123.

3 1 6 KNIGHT D UNLAP.

Next, the tone of the singing gas flame was employed in
the same way. The results were in direct opposition to those
obtained by Heinrich and by Titchener, for under favorable con-
ditions the fluctuations were unmistakably observed by each of
the five subjects employed. Since the note could be abruptly
terminated by bringing a card over the upper extremity of the
tube, it was possible to investigate the second death in this
case. This phenomenon was clearly observed by at least three
subjects, and interruptions made in a subliminal tone were also
clearly observable.

I said that the fluctuations occurred when the conditions
were favorable. These conditions were somewhat difficult to
secure, and demand first, that the tone shall be rather -pure,
second that it shall be relatively simple, and third, that it shall
have a fairly constant pitch.

Let us consider the first condition. Heinrich claims that
that gas flame tone is normally pure ; much purer than the
diapason tone. He argues that the latter is impure from the
supposed fact that it will not set up standing waves in a room,
while the gas flame tone will. Now the fact of the case is that
the diapason will set up standing waves, as can be readily de-
monstrated in a room not too large. The nodes nearest to the
wall may be easily observed, especially if the diapason be in-
sulated from the table and other resonant material so that there
may not be waves emanating from points widely separated.
The only difference between the diapason note and gas flame
note is due to the relative weakness of the former and its rela-
tive simplicity, which makes the nodes complete instead of par-
tial as in the case of the highly complex gas flame note.

The diapason note, when the noise of the electric contact
(provided one is used) is eliminated, is really very pure. The
fourth is the only partial which is strong in comparison with
the first and is so weak as to be indistinguishable when the in-
tensity of the note is reduced to moderate intensities ; being
absolutely neglectable in the case of minimal tones. There is
therefore nothing in the tone under these conditions which
could generate ' noise.' Nor can it be reasonably claimed that
the pipe transmission employed in my own experiment gener-

FLUCTUATING AND INAUDIBLE SOUNDS. 317

ated « noise.' It might possibly have modified the pitch, and
that is an inconsequential possibility. The only observable
effect was a relative simplification of complex tones due to the
disappearance of weaker partials, and a complete elimination
of faint noises.

That the gas flame is the source of a very impure tone is
readily observed. With certain pressures of the gas supply
beats of great strength and as slow (with a four-foot tube) as
one per second, may be obtained. Or, they may be increased
in rapidity beyond the point of counting. In most cases there
are several systems of beats in operation at once, and it is
practically impossible to obtain by any adjustment a tone which
is entirely free from the roughness (/. *., noisiness), of rapid
beats. But with certain adjustments this roughness was slight,
and being further reduced by transmission through the pipe,
did not prevent the fluctuation. The production of the beats is
probably due to the great strength of the upper partials ; which
leads us to the second condition.

By relative simplicity is meant the condition that one note in
the complex shall predominate considerably in intensity over the
other partials. With a two-foot tube it was almost impossible
to get an adjustment of gas and tube such that a note of about
three-foot wave-length and one of nine inch length were not
about equally strong. When reduced to extreme faintness this
compound was heard continuously, if at all ; but when made
strong enough so that the pitch was perceptible, the note heard
alternated between the two, showing a selective fluctuation of
attention which would prevent the disappearance of the note
completely.

As regards constancy of-pilch, the gas flame tone is also apt
to be deficient at times, as may be noticed by observing the vari-
ability in the number of beats which it makes with a steady tone
from some other source. This inconstancy is at other times so
small as to be neglectable, but there is always a possibility of
its occurrence. It is perhaps due to variations in the temper-
ature of the air column in the tube.

In addition to these factors which tend to prevent fluctu-
ation, there is another which is not peculiar to the gas flame

3 i 8 KNIG HT D UNLAP.

experiment, but yet seems to be especially strong therein. This
is the persistence of after-images of the sound, which tends
to bridge over gaps which would otherwise be caused by the
psychological disappearance. At times, when the tone was
shut off without abruptness, the subject would continue to hear
it for a minute and a half after its physical disappearance.
Why there should be more difficulty from this cause with gas
flame tones than with other tones, I do not know ; but such
seems to be the case, so far as my observation goes.

Such inference regarding the influence of ' purity of tone '
as can legitimately be made from my own and others' experi-
ments, is that the advantage as regards case of fluctuation is on
the side of the purer tone ; and on the side of the simpler tone
as against certain types of complex tones, as regards disappear-
ance at least. The fluctuations of a certain type of complex
tone of moderate strength may however be more marked than
those of a simple tone, if we consider the selective fluctuation
mentioned above.

The apparent contradiction of the first of these inferences
by the easily obtainable fluctuation of such * noisy ' tones as the
telephone snarl and watch ticks is not serious. The telephone
snarl is highly complex, but not very ' noisy ' after all, espe-
cially when reduced to its lowest terms ; and the complex is not
capable of ready resolution by selective attention. The ticks
of a watch, on the other hand, vary physically so that they
cannot be used as a criterion of any sort.

Noisiness, as distinguished from mere complexity of tone,
can be considered as nothing else than beat quality; i. e.,
shock, or jar, which is the characteristic thing about beats in
an otherwise smooth tone. The function of beats in preventing
fluctuation is closely connected with the ' second death ' of the
minimal tone. The sharp contrast between the minimal sound
and the succeeding silence is perceptible although the sound
was imperceptible up to that moment. Such contrasts occurring
with sufficient frequency may prevent the pulsations of the
thus marked off sound from ever clearly disappearing unless the
intensity be weakened below the point at which the contrast is
perceptible ; in which case no sound will be heard at any time.1

1 The MSS. of this article was received on April 2, 1904. — ED.

SOME OBSERVATIONS ON VISUAL IMAGERY.

BY DR. H. B. ALEXANDER.

With all that has been written on the subject of mental
imagery, it yet seems certain that there is much to be done and
said before we are to have a serviceable clearing up of the phe-
nomena. It is possible that we have learned all that we may
by the usual * breakfast table ' questionary, or like methods '
relying upon uncritical introspection. There is too much danger
of suggestion and fable and the trickery of language for the
results to be perfectly reliable. When it is possible for one
deaf and blind from childhood to use the language of the lost
senses with the grace and truth shown by Miss Keller, more
than ever must doubt be cast upon ordinary wordings as descrip-
tion of psychical events. The tact and deftness of such descrip-
tions as ' white darkness ' applied to dense fog, * luminous
shadows ' to a river's reflections of trees, the most consummate
literary skill well might envy and only an unusual visual sensi-
tiveness would seem likely to suggest.2 But while the need for
caution is thus stressed, the desirability of careful observation
is not at all gainsaid. Rather, now that the general character-
istics of imagery are fairly determined, it is the more urgently
called for; theories as to imagery cannot be on sure ground
until we have fuller detail, and the detail is only to be acquired
from trained introspection.

The nicety of discrimination demanded for just observation

1 Such as that of M. Ribot : " J'ai toujours proc£d£ de la meme manicrc,
en disant au sujet : ' Je vais prononcer plusieurs mots ; je vous prie de me dire
imtn£diatement et sans reflexion, si ce mot n'eVoque rien dans votre esprit, ou
s'il eVoque quelque chose et quoi ?' La r^ponse £tait notee anssitot; si elle
tardait plus de cinq & sept secondes, elle 6tait conside're'e comme nulle ou doo-
teuse."—L'£vobttton des idtes gtnirales, p. 131. The present writer con-
ducted in 1896 a series of similar experiments interestingly corroboratiYe of M.
Ribot's results.

2 See The Story of My Life, by Helen Keller, pp. 21 and 39 (New York,

1903)-

319

H. B. ALEXANDER.

can only be known in the attempt, arid at the best the results
are far from indubitable ; the mind too inevitably permutes and
falsifies its own states, too constantly alters its phenomena under
the influence of interest. I emphasize this merely that I may
not seem to invite disregard of the personal equation nor to fail
to reckon in the idiosyncratic distortion unavoidable to the study
I undertake. This study is based upon memoranda covering
a period of several years, noting the experiences as they occurred,
and so having the advantage of representing psychical happen-
ings induced in the mind's normal course, usually with no
thought of special observation. I might add that the person-
alism necessary to the discussion appears to be permissible only
in view of the general significance which I believe the facts to
have. These facts are in themselves of an essentially tame and
usual type, — which should constitute their chief value.

I. GENERAL ANALYSIS.

I shall begin with a general analysis of grades and types
of visual imagery. With reference to vividness, three grades
or intensities are to be discriminated.

1. There are the fleeting images of common thinking.
They are vague, fragile and ephemeral, only by chance to be
observed, the moment they attract attention transforming them-
selves to grade 2. They may be pictures of things or typo-
graphical images.1

2. Whenever any image is fixed in attention, it becomes
distinct in outline and seemingly filled in with substance. The
liminal quality in this grade is of a tenuous, * watery ' sort, but
if the thing imaged is naturally of distinctive color, the color is

1M. Ribot {op. cit.) recognized as a peculiar variety of his 'type visuel
typographique ' the imagery of compositors, finding his differentiae in the facts
that the compositors he examined ( i ) saw the words given in a particular ty-
pography, sometimes named, and (2) for semi-concrete words had accompanying
picture images, for abstract words no images. I myself worked some years at
the case, and I have many typographical images. But such images are certainly
not traceable to the types, which seems to be M. Ribot's conception. The
compositor always reads the letters upside down, and in handling the type
depends quite as much on touch as on sight. As for the instances where the
type face was named, I, as a typesetter, might very well describe my typograph-
ical images as in 'ten-point roman,' but such description would mean no more
than technical familiarity with type sizes.

SOAfE OBSERVATIONS ON VISUAL IMAGERY. 3*1

sure to appear. Thus dog calls forth only a colorless (gray)
figure of a dog, but Spot summons the image of a particular,
black and white, shaggy dog. The color element varies from
the black and gray (pure white is difficult) of typographical
images to a versimilitude hard to differentiate from reality.
With the exception of the typographical, the images all appear
small, i. <?., as miniature copies of the things they represent.
Furthermore, they are commonly isolated; there is no back-
ground of non-related objects. A landscape may be presented,
but it is seldom full or distinct unless enlarged to grade 3.

3. In the third grade images are richer in detail ; there is
illumination as of a stronger light thrown on, though the color
scheme qua color is not altered ; there is added substantiality
and a filling in of background ; and the images appear to be of
the actual or approximate size of things as viewed from a
normal standpoint.

It will be seen that the three grades pass perceptibly into one
another. In general, the variation in distinctness of outline or
figure is minimal ; variation in definiteness of content (as smooth
to shaggy) is greater ; variation in substantiality is still more
pronounced, and variation in color, luminosity and apparent size
is greatest of all. In imagery subject to volitional retention the
vividness is directly proportional to the attention bestowed. Of
the imagery as a whole, the conspicuous characteristic is frag-
mentariness ; seldom save by effort is a whole object seen, and
within the given content there is usually one portion more strik-
ing, distinct and detailed than the remainder.

In addition to the three grades, there are to be distinguished
two types or classes into which the images fall. These are :

a. Voluntary or memory images ; all images that may be
called to mind or retained by an act of will. Memory images,
in the simplest sense, afford the typical instance, but I include
along with simple reproductions all images consciously con-
structed from remembered elements, as, for example, a geomet-
ric figure, a landscape ideally composed in accordance with the
elements furnished by a description, or a mechanical device
illustrated in imagination.

/?. Spontaneous and irrelevant images, the salient character-

322 H. B. ALEXANDER.

istic of which is that they seem to determine their own occur-
rence, coming and going of their own accord. Of course these
images can be retained or reproduced in memory, but the reten-
tion or reproduction involves a change of quality : it removes
that asset of surprise and perversity which gives so much of
their forcefulness, and usually it projects them into new asso-
ciational environments and new spatial contexts.

By ' spontaneous and irrelevant ' I imply a partial disjunc-
tion. Spontaneity characterizes the whole group ; irrelevance
is a usual but not invariable feature. By irrelevance I do not
mean necessary absence of suggestion, but its lack of rationality.
Indeed, if the province were dominantly ratiocinative ' irration-
ality' would be the better word. To instance : If in reading, say,
the description of Cleopatra journeying to Cydnus, I imagina-
tively build up the picture, though with all the royalty of Shake-
spearian imagery, the act differs not greatly in kind from an act
of memory, for I can use only materials selected from the mem-
ory store.1 But if, as I read, in place of Cleopatra

O'erpicturing that Venus where we see
The fancy outwork nature, —

I meet here, suddenly, exasperatingly, a grotesque negroid face,
this I call an irrelevant image ; it is a perverse and irrational
intrusion, insulting alike to Shakespeare's lines and to the intel-
ligence before which it presents itself. A ^posteriori there is
traceable some associational suggestion, — Cleopatra being an
African ; but taken as a type, it is very like a prank of secondary
personality. Indeed such images argue two planes of intelli-
gence, in the apprehension of meanings : there is, first, the plane
of constructive coordination, or rational synthesis ; there is,
second, the plane of uncritical association, with an isolative
intelligence incapable of just or proportionate assembling of
imagery. Frequently, however, images of the /3 type show no
discoverable connection with the conscious context ; spontaneity
and temporal independence are their sole differentiae .

1 1 purposely discount the imaginative fervor which really gives such con-
struction more vitality than is common to memory.

SOME OBSERVATIONS ON VISUAL IMAGERY. 3*3

II. PROJECTION AND SIZE.

A satisfactory criterion of the externalization of images is
difficult to obtain. In my own experience, all visual images
appear in a field of vision, therefore, in that sense, as external ;
they never seem to be ' in the head.' They are not, however,
all external in like sense ; not all appear in the space which I
sensibly perceive or am conscious of as real ; there seems to be
truth in the notion of a ' mental eye,* and it seems to perceive
in space-relations of its own. Thus it is quite possible to com-
pare a memory image with an after-image, although the two
are assuredly not present in the same space. Nor is the discrim-
ination of spaces dependent upon vividness ; the memory-image
is often more vivid than the after image, and it may, as at dusk,
be more vivid than an actual perception. William Blake
affirmed of his own imagery — from which his wonderful draw-
ings were directly copied — that it was * infinitely more perfect
and more minutely organized' than ordinary perceptions. Fur-
ther, Blake discriminated these images very explicitly from ex-
ternalized or projected images — a ' ghost,' as he put it, he had
never seen but on one occasion, when he beheld a repulsively
spotted being on the steps of a house.1

I distinguish, then, as 'mental' images, those which have
disjunct spatial contexts of their own ; bona fide projections are
images that intrude upon what one is conscious of as real space.
This does not mean that the imaginary space is not external in
the sense of being in a field of vision ; but it is distinctively an
imaginary space, related to the mind's eye. To talk about its
internality or externality at all borders on absurdity ; it is simply
not included in the space system of which body and head form
a part.

But it is not unusual for images in imaginary spaces to
plunge into real space. The mark of their advent, and, I take
it, the safest criterion of projection, is their modification of ac-
tual sense-perception. When a portion of the perceived sur-
faces of things is blotted out to give place to an image — as
when a picture appears upon a printed page and hides or blurs
the print — then projection takes place. Thus it is relation to
objects of perception that is the determining factor.

>See William Blake, by Alfred Story (1893).

324 H. B. ALEXANDER.

Effort has been made to distinguish very sharply between
imagery that affects sensation and imagery that is purely idea-
tional on the hypothesis of distinct cortical centers,1 and this
may very possibly be a just discrimination, but it certainly can-
not sustain any very sharp cleavage of a subjective character.
When I speak of images being projected, I mean that sensible
ideas become sensation (that is modify sense-perception) with-
out any material alteration of character and content. This is
not a case of hallucination or of pseudo-hallucination (if this
means hallucination), but it is the case of a visual mental image
becoming a visual physical image, and being recognized as
such. I imagine that images so projected are really what is
meant by ' pseudo-hallucinations.' Professor James' account of
these, characterizes very well images of the /9 type, when pro-
jected as I have described : * From ordinary images of memory
and fancy, pseudo-hallucinations differ in being much more
vivid, minute, detailed, steady, abrupt, and spontaneous, in the
sense that all feeling of our own activity in producing them is
lacking'; further, they are 'projected outwards.'2

As a matter of fact, there is no hard and fast line between
sensation and sensible imagery. Even the criterion suggested,
of spatial interference, is not of indubitable application. I recall
an instance in which an apparition-like image appeared directly
over the shoulder of a person with whom I was conversing, so
attracting my attention as to elicit query concerning what I was
seeing. The image occupied a perfectly demarkable locality,
toward which my eyes were visibly accommodated, yet J was
not then aware that the real space was in any sense infringed
upon and was perfectly aware that the image was of the fancy.
As bearing upon the correlation of spaces, I note that many
times projected images appear sequent to winking; I have

1 See The Nature of Hallucination, Boris Sidis, PSYCH. REV., N. S., XI.
2. The statement (p. 128) that ' ideas and sensations differ fundamentally, they
differ in kind and no amount of ideational activity can ever be made to become
sensory in nature,' if it really means that mental imagery cannot be projected in
the sense defined above, can hardly be accepted ; the reverse is observably true
in every respect except continuity of spatial context and even the ideational
space may be considered as subject to projection when we consider the anoma-
lous smallness of some projected images (see below).

* Psychology, II., 116-117.

SOME OBSERVATIONS ON VISUAL IMAGERY. 325

repeatedly become aware of previously unnoticed imagery in
this way. It would seem that a blanking of the retina to outer
stimuli were necessary to the apprehension of these images,
which, afterwards, are enabled to maintain themselves for an
appreciable time in competition with sense-perceptions. Such
images always appear in specific space, — on the wall, the page
of a book, etc.

In this connection, it may be in place to offer demur to
Professor James* categorical assertion that it may be stated * as
a universal proposition that after-images seem larger if we project
them on a distant screen, and smaller if we project them on a
near one, whilst no such change takes place in mental pictures. "
This is a usual, but not an invariable rule ; some mental images
follow precisely the law of after-images, indicating similar
retinal excitation. For example, after an evening at chess or
whist, I have observed, even at an interval of an hour (surely
an impossible intermission for the advent of a positive after-
image),2 projections of chessmen or pips suffering the same dis-
tortion and alteration of size as ordinary after-images. Again,
the occurrence of an image on the page of a book is plainly
conditioned by the angle at which the book is held and is liable
to distortion to suit this angle.

From the foregoing it will be seen that there is no universal
distinction between ' mental ' and projected images, and after-
images. It is true that the latter are manifestly of sensory occa-
sioning, but it seems also certain that all projected images
involve retinal excitation. A difference of importance, though
a relative one, is the ephemeralness of the evicted central
imagery, — in my case, quite unable to stand up against incom-

1 Psychology, II., 51.

2 These images also differ from after-images in being wholly ephemeral,
having not more than a second's duration, '.whereas an ordinary after-image may
endure a minute or more. Professor James cites Meyer : " Most of these sub-
jective appearances, especially when they were bright, left after-images behind
them when the eyes were quickly opened during their presence. For example,
I thought of a silver stirrup, and after I had looked at it a while I opened my
eyes'and for a long while afterwards saw its after-image." I do not see that this
is not a case of simple projection, unless, as Professor James infers, negative
after-images are meant. I have observed colored images to project themselves
in black and white, but this I lay to their inability to compete with sensuous
luminosity and to the fact that the color element is the weaker in my visualuation.

326 H. B. ALEXANDER.

ing stimuli. The various differences between after-images and
imagination-images which Fechner has noted,1 I find, to hold
either not at all or to only a partial extent. After-images are
for me no more coercive or involuntary than images of the /?
type ; they are neither brighter nor sharper in outline than other
images may be ; and though usually * unsubstantial and vapor-
ous,' this is not necessarily the case. On one occasion I placed
a new silver coin on the palm of my hand, in bright sunlight,
for the purpose of getting a negative of the design. I had no
thought whatever of the hand. On closing the eyes it was with
a shock of utter unpreparedness that I beheld a tiny hand as
substantial, detailed, truly colored and formed as an actual per-
ception. The surprise was so complete that for an instant the
image was a veritable hallucination. With 'the eyes open it
enlarged and dissolved in the usual fashion.

The tininess of this image is worth note. It brings up the
whole interesting question of size in visual imagery. With me,
and I presume with most persons, all after-images appear small
when viewed with closed eyes or in pitch darkness. A window
seen from a distance of seven or eight feet, as an after-image,
contracts to a seeming height of two or three inches. The con-
traction is not instantaneous, though it must be particularly
heeded to be observed in transitu. The size indicated remains
nearly constant (so long as the eyes remain closed) during the
changes attendant upon dissolution.

The small size characteristic of after-images with the eyes
closed holds of most mental images where no effort is made to
realize a just magnitude. Smallness, in fact, seems to be one
feature of the spatial independence of these images. For the
shrinking of after-images, the taking away of all the hints
by which we commonly judge perspective might seem a suffi-
cient account, and similarly abstractness an explanation of the
smallness of centrally occasioned imagery ; but a -priori there
seems to be no reason why an image should assume one size
rather than another. Some help may be gained by reference
to actual space perception. Chance juxtapositions often startle
us by their emphatic contractions of the third dimension, and by

1 Cited by James, Psychology, II., 67.

SOME OBSERVATIONS ON VISUAL IMAGEHY. 3*7

a sufficient effort one may very nearly free oneself from per-
ception of this dimension. By attentively envisaging a house,
for example, one may observe it gradually flatten out and shrink
into a diminutive toy house.1 But there is a limit to this contrac-
tion, a norm of diminutiveness beyond which it cannot go.
Apparently the factors that determine this limit include besides
narrowing of the attention, so as to eliminate the signs that
ordinarily give perspective, some direct relation of physical and
conceptual conditions.

I venture as a guess that the normal shallowness of space is
a sort of sense generalization, a perceptual construct, determined
by the focus of most distinct vision — that is, at about reading
distance, — and that the size of images, relative to the area stimu-
lated, is dependent upon the horopter at this focus.2 An object
at this distance stimulating an area equal to the area stimulated
by a house at a hundred feet would then give the abstract or
normal size of the visual idea of the house, which must hence
appear very small if envisaged as a whole. But it is obvious
that if all images are reduced to the same or approximately the
same absolute size, they will not all be reduced in like propor-
tion ; objects naturally smaller will seem relatively less reduced.
This is actually the case, the limiting instance being furnished
by typographical images which appear undiminished because
they are presented at the normal reading distance. We thus
get an explanation of the usual lack of true proportionality of
mental images.

It is not to be supposed that small images are universal.
Many people either do not see images as small, or, if they do, they
still reckon with space and judge the objects as if seen at a dis-
tance sufficient to account for the size.' Possibly it is only the

1 Apropos, I vividly recall that on the first railway journey of which I have
recollection (at about the age of four), I was amazed to see tiny people, tiny
cattle and houses, all but a few inches high yet all as real as reality. There
were whole fairylands through which I was whisked, but with imperishable
glimpses retained. No scepticism of my elders could abash my conviction, for
what I saw I saw with my mortal eyes.

2 So far as I have been able to judge the process is independent of i
accommodation, which would rather discount the kiniesthetic element,
reverse case, however, of imagining an image enlarged, kiniesthois »cem«
clearly present.

3 So a friend testifies.

328 H. B. ALEXANDER.

bookish who find the ideal world crowding so close ; and in any
case it is possible to enlarge all imagery subject to volitional
retention either by a fiat extension of the third dimension or by
imaginatively filling in the detail that creates perspective,
though each process involves effort.1

III. THE IMAGERY OF DREAMS.

The imagery of the state bordering on sleep is so distinc-
tive in an empirical way that I have a somewhat coercive feel-
ing that it ought to be placed in a class by itself. At the same
time I can discover no certain characteristic differentiating it
from the type /5 ; the main differentia from other instances of this
imagery is greater duration with an aptitude for evolving under
the eyes from one form to another, but the evolution is so pal-
pably a consequent of the duration, and the duration so plainly
due to the freedom from sense or thought competition in the
state of semi-sleep, that this differentia is of little account.

It is possible that these images arise in connection with
idioretinal light or some internal disturbance of the eye,2 after
the manner of ' fortification' images, but in no case will this
account for the intrinsic character of the images — that is, as
representations of ideal things, — and in any case there remains
a residue of manifest central origin ; there is no ground to sup-
pose that the retina can itself create imaginary portraits or

1 Not to place too much weight on the mechanical features of vision, it is
well to bear in mind the analogous smallness of auditory images which to many
people, especially when overworked and nervous, take the form of imaginary
whispering. Though weak in auditory imagery I have occasional snatches of
music having a peculiarly phonographic diminutiveness. It should be noted, too,
that there is some variation in the focus of imagery — if I may so call it. An
after-image of a lighted lamp, with closed eyes, appears at about the distance
mentioned ; but if the image be taken with one eye only, the other being left
open, it is appreciably nearer. Of course the contrast with the extension seen
by the open eye might very well account for this.

2 Compare the statement of Dr. Sidis, in the article cited: "Often in
closing my eyes and keeping quiet, so as to become somewhat drowsy, and
watching the field of vision, not directly, but, so to speak, from the corner of
the eye, animals, figures, faces, can be seen forming and dissolving into mist.
These phantoms can be directly traced to specks of light and masses of color
coming from the retina and especially from the macula lutea." The certainty
of this origin seems to me far from apparent.

SOME OBSERVATIONS ON VISUAL IMAGERY. 3*9

evolve ideal representations consistent with, yet differing from,
reality.

The persistently sensational feel of this imagery would, how-
ever, seem to warrant the inference of some physiological ele-
ment, obscure though it may be. That there may be a kinas-
thetic factor involved is at least suggestively hinted by an
experience which I may cite : While reading, both hands
beneath the book, I became drowsy, my eyes closed and my
head seemed to fall forward. Just as, or just after, my eyes
closed, a face appeared directly before them, — a man's face,
thoroughly Mongoloid, excepting the eyes which were full, open,
brown and very bright. As I saw the face, it seemed to fall
toward mine, the eyes at the same time closing so that I saw
the lids droop ; there was also a vague image of a hand raised
to catch the head. The image appeared with a vividness suffi-
cient to rouse me completely. It is to be noted that my eyes
did close, my head nodded, and although my hand did not rise
up, my first feeling after the surprise was that this hand some-
how seemed unsatisfactory, as if it ought to move. On several
other occasions I have noted the same phenomenon of an image
of closing eyes in connection with the drooping of my own lids.
Such images seem to indicate the translation of an affective
state into a form of cognition, of a subjective feeling into an
objective representation. In a grade of conscious life where
mental process has reference solely to near activities the like
function might well possess vital significance.

The general character of borderland images is fairly de-
scribed by Professor James' account of pseudo-hallucinations ;
they are relatively ' vivid, minute, detailed, steady, abrupt and
spontaneous.' I find them also diminutive, projected to but a
few inches, mutable or self-changing, usually fragmentary, and
frequently grotesque or odd. To give instance : On one occa-
sion I saw what seemed to be a living cartoon of the face of a
friend, a microscopic exaggeration, yet perfectly recognizable.
Again, a woman's head and shoulders, the head thrown back
with strong light and shade on the features ; the face of a primi-
tive savage type, prominent cheekbones, wide zygomatic arch,
prognathous jaws, the mouth open showing large irregular

33° H. B. ALEXANDER.

teeth ; the eyes were in the shadow of heavy brows, the hair
was low on the forehead, the neck very muscular. On another
occasion, a Dantesque nose and brow which gradually filled out
to full profile and lengthened into a vapid expression of mock
piety, the whole irrelevantly giving place to a thick-lipped mouth
with rows of teeth widely shown.

Preponderatingly these images represent faces of men or
imaginary animals (the order of frequency I judge to be : eyes,
mouths and noses, profiles, full faces, bodies), and it is not an
uninteresting speculation — if the like experience is as common
as I take it to be — whether such tiny personifications may not
have had a deal to do with the development of belief in fairies 1
or diminutive night-folk who come, like Queen Mab,

In shape no bigger than an agate stone
On the forefinger of an alderman,
Drawn with a team of little atomies
Athwart men's noses as they lie asleep.

Between the imagery of semi-sleep and that of dreams proper
there is one marked distinction — that in dreams objects appear
to be of the size of real objects. Further, in dreams images do
not appear to be images at all, but seem to be the real things
themselves. But the latter distinction is subject to our very fre-
quent consciousness of the unreality of the dream state, and even
the distinction as to size must be received with caution. In the
dream state there is no competing space relation (unless some
vague sense of the body's extension), and hence no criterion by
which to judge size outside the somewhat vague remembrance
of dream consciousness. If a comparison of dream space and
real space could be made, it might be found that dreamland is
confined to Lilliput. I guess this from two confirmatory experi-
ences. The first was an awakening in the midst of an early
morning dream in which a horse drawing a sleigh seemed to be
approaching, growing in size as he did so with the rapidity and
a good deal of the effect of the moving picture produced by a
kinetoscope. Wakened, I became aware of sleigh-bells outside,
the approaching sound of which undoubtedly suggested the

1 At least this seems no more improbable than the usual derivation from a
race of palaeolithic dwarfs. Compare, also, the conception of the human soul
as diminutive and image-like : Tylor, Primitive Culture ; I., p. 450.

SOME OBSERVATIONS ON VISUAL IMAGERY. 331

dream ; in addition, I became conscious of having heard the
bells in the dream, and also that the visualization had been
diminutive, enlarging as the eyes opened. The dream had, so
to speak, telescoped, and was retrospectively given a proper
extension through the influence of the continuous stimulus fur-
nished by the bells. A second case was the projection of a
dream image of a human figure upon sudden awaking. It was
one of a group in the dream, and when my eyes suddenly
opened, I was surprised to see the figure lengthen out exactly
as does an after-image.

On more than one occasion I have had a distinct dream con-
sciousness that the dream experience was one of imagery, not
reality ; but this I conceive to be a late development, partly due
to the nature of my interests. Probably as fair a test of the
relative real-seeming of dreams as can be gotten is the duration
of dream impressions as compared with memories of real ex-
periences.1 My own most long-standing memories include at
least two dreams which occurred when I was between three and
four years old, as I know from the fact that both are concerned
with the death of my mother at that time. Except from inter-
nal evidence (their inconsistency with reality as I have come to
know it), I should not be able to distinguish these dreams from
real memories, and of other memories of early origin I am not
even now certain. Nowadays, however, I am frequently dream-
conscious, and even find myself fully alive to the theoretic inter-
est of a dream state, which I am yet unable to influence. A
recent instance curiously illustrates. On January 7 I dreamed
that I was being shown a collection of pen and ink drawings
and instructed as to their merits. One of the drawings was of
a mediaeval landscape, and on examination the clouds in the sky
proved to be swathed cherubim in horizontal attitude. In my
dream this picture seemed very unique and impressive, and
on waking I remembered it with great distinctness. On Feb-
ruary 15 I dreamed that I was shown this same drawing again,
that is to say, I recognized it in my dream ; but on awak-

1 This test is if anything unfair to the dream, since, especially in later life.
it offers fewest points of attachment to that systematic apprehension of reality
which experience generates and which in turn so largely selects those memories
in which experience is preserved.

33 2 H. B. ALEXANDER.

ing I remembered that what I had seen in this dream had in
fact been a chalk drawing on slate of two vertical, or erect,
swathed figures, with no landscape at all.

In these dreams and their memory images there is a curious
interlocking of the dream world and the real world and an inter-
connection of dream interests somewhat bearing out the sug-
gested conception of a secondary plane of intelligence. Al-
though dreams are able to create their realities at whim, they
yet reveal a rudimentary consistency which is the beginning of
law and order. Linking together in memory, even if errone-
ous, is one step ; discrimination of the dreamer's self from the
dream world, though it be but a feeling of helplessness in the
presence of dream mastery, is a greater step ; and finally the
advent of intelligent interest allies the dream state very closely
with rational experience. In a few recorded instances dreams
have actually solved problems which have baffled normal intelli-
gence,1 and it is worthy of note that they have done this in the
form of sensible imagery and in what might be called the dra-
matic mood of reason.

IV. THE INFLUENCING OF IMAGERY.

From foregoing discussions it should be apparent that lines
of distinction between different classes of imagery are nowhere
fixed and determinate. After-image and imagination-image,
projected and non-projected, dream image and waking, all
are united by liminal, indecisive cases. The distinction which
maintains itself most steadfastly is that between images of the
a and ft types, between voluntary and spontaneous imagery,
but even this distinction will be found to break down to a con-
siderable degree when we carefully examine the influences to
which imagery is subject.

In restricting the influence of volition to a certain type of
imagery, I have allowed myself an accommodatingly indefinite
use of the term. It is a question whether new combinations of
elements are ever volitionally predetermined, while on the other
hand it is certainly possible voluntarily to influence the occur-
rence, if not the content, of images other than memory images.

JSee the cases reported by Professor Newbold, PSYCH. REV., III., 132.

SOME OBSERVATIONS ON VISUAL IMAGERY. 333

After-images, of course, are found by paying attention to them,
and in the same way, if one keep on the look-out for imagina-
tion images one is sure to perceive them ; indeed, I am not at
all confident that their liveliness and frequency may not be
directly dependent upon an attitude of expectancy toward them,
— upon an assumption of incipient rapport with their plastic
objectivity. This is not saying, however, that there is involved
any predictive certitude as to character; the images always
have some degree of surprise in store. Even in the most
abjectly servile imagery there is always some spontaneity. It
is sometimes held that it is impossible to will an indefinite end,
that desire must be conceptualized before it can become an
object of volitional effort. Were this really so, we should never
will anything at all. The calling to mind of the simplest idea
or the merest recollection always introduces some incommensur-
able element ; there is always at least possibility of surprise in
the summoned presentation. For example, I entertain the idea
of an elephant, willing the image, and instead of the full figure
which I have a right to expect, there appears only head and
trunk. Further (I cite a specific occurrence), it appears at a
certain distance and of a certain (life) size ; the head is turned
slightly to one side, the tusks are absent. It is perfectly plain
that this experience represents a considerable degree of incom-
mensurableness ; the formal and final causes by no means square
with one another. Volition is thus largely discredited at its incep-
tion, wholly unable to free itself from irrelevance and chance.

Moreover, even the general influence of volitional attention
in assuming the attitude of expectancy must be guardedly ex-
erted if it is to be effective. If too direct, attention really seems
to defeat its object in many cases ; often it must be oblique, side-
long, in order to catch that for which it waits. There is needed
a peculiar kind of mental passivity — even in connection with
the keenest curiosity — where images are concerned. If one too
deliberately tries to manipulate them, they simply balk, and
more than likely disappear altogether. Of course this does not
apply to memory images, though even a memory image is apt
to obtrude itself in unexpected garb if left unwatched.

If one may so put it, then, the most direct influencing of

334 //• B. ALEXANDER.

imagery must be indirect, and the further one proceeds from
mere reproduction the more the indirectness grows. In ordi-
nary thinking there is seldom a consistent rational nexus between
thought and image, doubtless because the thought is too rapid
for the imagery to keep pace. At the same time a reflective
state of mind, one in which the ideas are chief concern, is de-
cidedly conducive to vivid imagery, however inclined this may
be to sport-like characteristics. If the interest be imaginative
rather than abstruse the images are naturally more abundant and
are also more apposite ; and if the imaginative thinking is of
sufficient intensity or lasts for some time, it alters the whole tex-
ture of thought, frequently for days at a time. On one or two
occasions I have undertaken some line drawing, and on these
occasions have found a large accession of images in black and
white line-work. These images were characteristically spon-
taneous, with a strong element of surprise, and often they were
projected. Usually the specific suggestion or association was
quite untraceable, though in one case I recognized the nucleus
of a striking presentation in a forgotten penciling of six weeks
previous. The image had been subconsciously incubated during
this period and at the end hatched forth with all the self-suffi-
ciency of a lusty chick.

While interest is thus to a degree able to contract the range
of imaginative experience and intensify its content, it is unable,
so far as introspection shows, to furnish any account of actual
variations. No element of suggestion, no law of association
will quite explain the phenomena. Images frequently bear not
even a hypothetical relation to their contextual consciousness.
For example, I am quite at a loss for an explanation of the
curious ethnological traits or yet of the grotesqueness prone to
characterize my imagery. Again, even in the case of memory,
images are seldom enough mere reproductions ; neither are they
mere erosions of former impressions ; rather they are veritable
transformations, growths, the spontaneity of which is sufficiently
attested by the fact that they sometimes surprise to laughter.
The problem of their origin and function may be worth a few
paragraphs of speculation.

SOME OBSERVATIONS ON VISUAL IMAGERY. 335

V. IMAGERY AND RATIOCINATION.

Sumarizing the general phenomena of occurence, it may be
stated : First, that the mind tends normally to inhibit a mass of
inflowing imagery, either directly through refusal to entertain
the presentations offered, or indirectly through ratiocinative or
perceptual preoccupation. Second, that the mind is able to
create, through interest or suggestion, an imaginative diathesis
which, in a measure, overcomes the inhibitory function. Corol-
lary to this is the tendency of the mind so predisposed to sink into
an attitude of mere curiosity towards it images, which hence
achieve relative independence and self-direction. Third, that a
class of images (the /? type) bear internal evidence of a process
of subconscious synthesis, — that is, the elements of which they
are composed in the relations in which these elements appear
belong to no recognizable historical experience. This incom-
mensurable element is present to some extent in memory images,1
but ex hypothesi is there subject to limitations.

These facts to my mind make plausible the supposition of a
primitive form of plastic intellection, intuitive rather than reflec-
tive, yet having the true characteristics of rational thinking —
namely, dissociation and ideal synthesis. It is much the habit
of writers on ratiocination to credit all mental power of abstrac-
tion and construction to the naming-function of language. Even
if the generic image is possible, maintains Stout,1 we have yet to
show that it has not been rendered possible by preceding word
analysis. It seems to me that such writers rely too strongly
upon the supreme significance of language in their own, exact,
thinking. Metaphysical abstrusities naturally demand a mobil-
ity of ideas which no form of thought limited to space-bound
presentations could yield. But this is hardly true of thought
in general. Galton found scientists to be nearly devoid of
imagery, but women and children were freely endowed with it.
The kind of ideas which one entertains makes all the difference
in the kind of medium demanded, Language is undoubtedly
the supremely facile agent of thought, but it is by no means the

1 Consult the interesting discussion of la mhnoire pittorcsque, by M. Paul
Souriau, in V Imagination de V artiste (Paris, 1901).

1 Analytic Psychology, II., 176.

336 H. B. ALEXANDER.

only efficient one. For my own part, I find that consciousness
of language is keenest when I write ; the words group them-
selves and fall in order far ahead of the pen ; yet when endeav-
oring to master a difficult abstrusity, it is necessary to lay the
pen aside ; consciousness of language then sinks to a minimum
and the conception if grasped at all is grasped in a sense of
relationships, it is not mentally expressed. So far as I can
observe such thinking quite transcends all use of language or
other sensuous element. In an opposite direction, effort to
understand geometrical or mechanical conceptions is vastly
furthered by mental pictures ; an illustration is much the most
effectual definition of a machine, a diagram the aptest solution
of a problem in space relationships.

It is certain enough that we do reason to some extent in
pictures. Whether such reasoning actually precedes linguism
may be doubted, but it seems to me that it presents a type of
reasoning which might be fairly developed among the higher
animals. Where the end of thought is proximate action — and
no animal, we may be sure, reasons for any great future —
thought should be as rapid and instinctive as possible ; and for
the saving of mental confusion and waste no surer device could
arise in nature than subconscious specialization of thought-
processes ; reflective consciousness is decidedly the product of
biological leisure. Moreover, since an animal's interests are
practically all in the world of space, visual images answer his
highest need for abstraction, especially since his thought's symbol-
ism must needs be of anunconventionalized, pictographic order.

The mass of imagery, then, which so presses in upon our
normal intellectual life may be no more than a ghostly reminder
of what was once the sum and substance of thought. The fact
that the function of the mental picture has been so largely taken
over by language would go far toward accounting for the irra-
tional, dreamlike texture and sequence of the images ; they are
mere residual mental organisms, pursuing a natural course of
degeneration. Possibly the seeming degeneration may be
accounted for as the natural inferiority of the mode of thought
itself ; the image failing to keep pace with the normal growth
of mental life. Some such conception, in the case of dream

SOME OBSERVATIONS ON VISUAL IMAGERY. 337

consciousness, seems to be entertained by Dr. Sidis : '« It may
be claimed that the dream consciousness is to some extent a
reversion to the earliest forms of mental life, when the race wa«
as yet undisciplined by the accumulated experiences of ages of
social life." l

As bearing upon the suggested view, I venture to cite, also,
a suggestion of Dr. Bosanquet's : " It may be that in early soul
life this reproduction [of ideas by other ideas] is unconscious,
and that its results, the images which it brings before the mind,
are not used as ideas, i. e., are not distinguished from fact or
known to be symbolic of a content other than themselves. The
results of experience may be made available for the guidance of
an animal through suggestion effected by reproduction, but not
distinguished as suggestion from any presented reality. In this
process we have something that does the work of judgment and
inference, and that has the same fundamental nature with them."*

Apropos, it might be added that the better part of all our
reasoning is unconscious. I, at least, have never been able to
fixate with the mental eye an actual case of conscious disen-
tanglement of puzzle or of clear-seeing conceptual synthesis.
What I do find is a preliminary state of confused hetero-
geneity, mere puzzle, and then a succeeding state of illumination.
Order appears from disorder, plan from chaos, but in response to
no urgency or coaxing. Knowledge always comes as a precipi-
tate from the psychic solution. Occasionally a significant phrase
forms itself and presents itself, perhaps as a typographical
image, coming as from nowhere and with no apparent incentive,
but usually the process is merely the process of seeing through ,
and it happens with an utter spontaneity not to be distinguished
from that of the apparition of imagery.5

lLoc. cit.t p. no.

2 Logic, II., p. 15.

3 The MSS. of this article was received April 13, 1904. — ED.

INCIPIENT PSEUDOPIA.

BY REV. CHARLES CAVERNO, A.M.,
Lombard, Illinois.

One of the miseries of my life has been a supersensitive
retina. The sunshine on snow or a bare road is often tor-
ture. Any cone of artificial light is unendurable. To sit in an
electric-lighted room, even with eyelids closed and perhaps
with a shade over the eyes, is to forfeit the efficiency of the next
day. If I have congestion about the head from grip, fever or
cold, then it is an affair of bed, black bandages and darkened
rooms. Out of this experience have come some observations
which are of interest to myself and may be to others.

The doctrine of the specific action of the senses is well
understood. The nerves of the retina are made to respond to
light and they answer even mechanical pressure in terms of
light. I have been interested in watching, in my hours of
darkness, the illumination which I get from the inflammation
of the retina. The glory of light seems then turned on at full
head. In the interest of clear psychological distinction I wish
to say that this light is not subjective — a thing of imagination
— it is objective; as it comes from the pressure, against the
optic nerve-filaments, resultant from congestion.

Further on we shall see where psychic action comes in, as
plainly distinguishable from this illumination, or any of its
phenomena, as is such psychic action in any perception which
is derived from light that comes from the external world. The
field of vision in my case is confined to the optical apparatus
itself.

I am satisfied that I see the march of the blood through the
arteries of the retina itself. I feel my pulse by seeing it in the
retina. It is with some timorousness that I make such a state-
ment, but I make it. The statement rests on the perception of
the same phenomena recurrent for many years. I have seen,

338

INCIPIENT PSEUDOPIA. 339

under a microscope, the corpuscles of blood dart along the
tissue in a frog's foot ; and the motion there is similar to that
which I detect in my eye ; only the march is in open order in
the frog's foot, whereas in my retina the order is close. I have
tried to verify my conclusion in various ways. The steady
onflow I cannot stop nor turn aside by volition. I can talk
with friends or open my eyes, but when I settle back and the
eyes are shut, there is the same procession of bright particles
moving in the same way. This phenomenon comes in when
the tension from inflammation is high enough and departs as
that tension ceases. In this experience I am sure that psych-
ically I am a passive percipient, I see what I have to.

But the show does not end with the sensation of light and
the perception of the procession of the blood. I get an unlim-
ited variety of pictures projected before this interior vision.
They are the products of imagination I know, but I have no
will power over them. They begin at the edge of this inward
horizon, pass over the field and then depart. They come and
go at their own sweet will. I can never anticipate what is
coming. I can call up, alter or modify nothing. I am as
helpless about what the exhibition shall be as one sitting in the
midst of an auditory is over the movements of a panorama on
which he is looking. I have tried again and again, when a face
would begin to appear, to have it take the form of some one
familiar or loved, and have never yet succeeded. I am com-
pelled to watch the show that is brought up before me. While
imagination furnishes some of the elements of the spectacle, I
am sure inflammation furnishes others. For instance, some-
thing begins to move along before me and it develops into a
full-blown iron-foundry. Everything in it takes on the glow
that flashes out in a real foundry when the furnace doors are
opened ; only the color is uniformly distributed over everything.
The tools, the walls, the iron rods, the cinders are of molten-
iron color. A heap of cinders is simply a heap of gems, an
iron rod is burnished gold. I charge the flame-tint up to
motion which I interpret in terms of light. But the foundry
and all its appurtenances move along and disappear. I cannot
stop it. I cannot call it back. The glow may continue but it

34° CHARLES CAVER NO.

is formless. I have many sights savoring less of inflammation
than that of the foundry. I am entertained with landscapes of
all sorts ; and here again I am helpless. If a landscape begins
to move out before me I must take it as it is set. I have tried
over and again to turn a starting picture into a vision of my
old home, or of some familiar scenery, but it always develops
in its own wilful way. There is a beautiful ravine twenty
miles from Chicago with a magnificent elm at its head. I have
supposed myself starting up that ravine many a time but I have
never yet seen that elm in these compulsory imaginative excur-
sions. Within three months I have been sure that I was at the
lower end of that ravine and have had high hopes of reaching
the elm, but the vision refused to go in that direction and turned
me out, apparently on the edge of a wood in central Illinois
that I had not called for, and I

" Was left lamenting."

Perhaps I had better give some of the conclusions I have
formed. My helplessness in respect to imagination assimilates
the phenomena of my visions to those of dreams. But then are
we not in our waking moods more than a little helpless in re-
gard to imagination? Do we not then and there have to take
pretty much what volunteers in the field? It is a startling, dis-
quieting thought that we have little control over imagination
even in our normal conditions. It is the most elusive, evasive,
the least tamed of our faculties, the one most inexplicable.

Throughout all these visions I find sensation and psychic
action tethered together as in normal sight. The picture is
carried forward because it is physically connected with the
steady onflow of the circulation of the blood. In these visions
I am on the open road to the pseudopia of all delirium. Be-
tween what I see and what a victim of delirium tremens sees is
only a matter of degree. Carry the pressure by congestion on
the filaments of the optic nerve far enough, and let some of the
brain centers used in coordinating thought and act be involved
in the inflammation, and it is easy from my point of experience
to see that the vision of imagination might catch and hold the
attention to the exclusion of the vision derived from the external
world. I might then start to run up my fictitious vale at Brush

INCIPIENT PSBUDOPIA. 341

Hill or begin to set things to rights or wrongs in my mentally
invented foundry. Out of my long experience I have no
memory of unesthetic visions. Medically speaking they have
been "benign." Landscapes predominate in my panoramas.
Though I cannot call up the countenance of a friend the faces
I do see are not ugly. Rarely does a human being figure be-
fore me. Though my foundry appears to be in working order
to the last touch there is never a soul besides myself in it.

I am not fond of pain, but I have had my compensations in
experiences of the above-described sort and in observation and
reflection upon them. I have had my exhibitions when I have
not used opiates, or rather they have had me. It would make
no difference in the ultimate physical and psychic analysis how-
ever if opiates were used. They explain nothing. The ques-
tion still is, what are the laws of such visions or why their
lawlessness?

NOTE. — Since the above was written I have had another
experience with pseudopia. I lay in bed a few days from a
cold and congestion about the head. One evening the visions
came on and I described them aloud to my wife for half an
hour. This experience was mainly of a geological sort. Rock
views passed before me such as one would get in riding over
the mountains in Wyoming on the Union Pacific Railroad.
The rocks were at close range so that I could see all the bands
of stratification in the sedimentary rocks. Sometimes the rocks
were divided perpendicularly as well as horizontally so as to
suggest a wall laid up artificially. But I saw no tools and no
workmen. In fact I did not see a single human being in the
whole half hour show. The primitive rocks appeared in all
their amorphous condition. I could not get vista or prospect
among or over them though I tried with all my might. But
view after view of rocks in endless succession came on and went
off. The only modification of bare rock (no tree or shrub or
grass was in sight) was once when the view slid off to a river —
drift bank in which were tumbled large boulders as is usual.
One boulder was broken across the face and plainly showed the
mixed gray, pepper and salt mottling of granite. This did not
long continue and then the vision went back to the lithological

342 CHARLES CAVERNO.

vision which was on before. Nothing was repeated. The
variety was only equalled by the actuality as you climb the
Rocky Mountains and go over the crest in Wyoming. All of a
sudden the rock exhibition failed and there came out a beauti-
ful landscape in which nothing of vegetation was wanting to
completeness in effect. From embowering trees I looked down
a long vista over a grain field filled with bunches of grain as
farmers put them up before stacking. The field was appar-
ently of acres in extent and the further edge faded out down a
declivity with an intimation of a broad valley lower down, be-
yond. This closed the show and I have seen nothing since and
do not expect to till I have another season of congestion about
the eyes.

I do not care about drawing conclusions ; but I do not see
why I might not have the visions of frenzied oracles, bacchants,
hermits, devotees — any of the sights of sinners or saints, if my
mind was bent in their particular directions.

N. S. VOL. XL No. 6. November, 1904.

THE PSYCHOLOGICAL REVIEW.

THE CLASSIFICATION OF PSYCHO-PHYSIC
METHODS.

BY DR. EDWIN B. HOLT.

Every one who has undertaken experiments on the relation
of stimulus to sensation, or even read considerably on the sub-
ject, must have become aware of the inadequacy of the histor-
ical and still current classification and designation of the psycho-
physic methods. For many cases arise in practice which have
no immediate place in the classification, as for instance those in
which it is necessary to recognize and more especially to evalu-
ate judgments other than those of * less,' * equal ' and * greater '
(as say * much less ' * uncertain ' and ' much greater *) ; and other
cases arise whose place is doubtful because they have features
of several of the methods but have not all the essentials of any
one of them.

If for instance one were to wish to find the acuteness of a
given sense in an interesting pathological patient, and if from
independent reasons (as may well happen) one were obliged to
adopt the so-called method of right and wrong cases, one might
be brought to pause, if the subject were impatient or perhaps
hysterical, by the appearance of adverse emotions and fatigue,
due to the almost endless repetition of just two stimuli which
this method requires. And yet perhaps one could get from the
subject one fifth the required number of judgments on each of
five different pairs of stimuli, or else one tenth of the number
from ten pairs. Now clearly if one were careful to have one
stimulus common to these five or ten pairs, one could calculate
by the method of least squares from the five or ten resulting

343

•

ED WIN B. HOL T.

groups of judgments the coefficient of precision (' Pracisions-
mass ') with as much accuracy as otherwise from one group of
ten times as many judgments, on a single pair of stimuli. Yet
in such an instance one would not be using the method of right
and wrong cases. For although the point of departure and the
answer yielded conform to the descriptions of this method, yet
the actual data gotten and used in calculating the result are pre-
cisely such as one would have gotten from the method of least
differences, or minimal changes (Abstufungsmethode der klein-
sten Unterschiede) if peculiarities of the sense-organ investi-
gated or exigencies of apparatus available had forced one to use
few and coarse gradations of stimuli and to present the different

pairs at random. For in both cases the data would consist in
judgments of « less ' (<), ' equal ' (=) and « greater ' (>), which
would be distributed as shown in Fig. i, where the abscissae
represent the value of the stimuli greater or less than the stand-
ard stimulus (o) which is common to the ten pairs, and where
the ordinates represent the number of judgments.

In short the experiment described does not come under any
one of the traditional methods, since it has also features of at
least one other method, nor can it properly be called a * com-
bination ' of the methods, since it has not all the essential fea-
tures of any one.

CLA SSfFICA TION OF PS YCHO-PH YS/C MB THODS. 345

Apart from the confusion which every reader or experimenter
in psychology has probably felt in assigning a given experiment
to one of the four or six « methods,' the unsatisfactory state of
the whole matter is amply witnessed by the historic errors made
by the most distinguished experimenters in designating the
methods they have used, by the confused and overlapping defi-
nitions of the several methods invariably given in text-books,
and by the amendments offered from time to time by various
theorists. The errors of designation, although several were
made by Fechner himself l are chiefly of historic interest. Of
systematic interest, however, are the current definitions and
descriptions of the psycho-physic methods, and the several
emendations which their inconsistencies have elicited.

The case above cited of an experiment which, by having
features of two methods and yet lacking the essentials of any
one, belongs properly under none of the four or six methods, is
not wholly unprovided for in the treatises. Thus Wundt,1 in
defining the method of mean gradations, describes, for the case
that the middle stimulus varies irregularly, a certain simple way
of deriving the mean value from the raw data : but this way
need not be used, he says, since the raw data can also be treated
by the method of right and wrong cases, in which case the
whole procedure becomes ' a combination of the method of mean
gradations with the method of right and wrong cases.' Now
the simpler way of evaluating the data applies exclusively to the
method of mean gradations : is it then essential to this method ?
Clearly not, since it may be abandoned in favor of the method
of right and wrong cases. But then the method of mean grada-
tions is not, as is generally supposed, a method for both getting
and then evaluating raw data, but only for getting them. Con-
versely, too, the method of right and wrong cases is not at all a
method for getting data but only for treating them when gotten.
But if this is true, these two methods are not alternatives to be
chosen between, but supplementaries to be used in combination;
a conclusion which is at variance with the present theory and
practice.

'Miiller, G. E.; Zur Grundlcgung der Psychophysik, Berlin, 1879. S. 56 ff.
• Wnndt, W. ; Grundz. d. physiol. Psych., ste Aufl., Leipzig, 1901, I Bd., S.
480.

346 EDWIN J5, HOLT.

This contradiction arises, of course, from the confused defi-
nitions of the methods, whereby these latter, instead of being
four mutually-exclusive logical classes, as they should be, are
defined as being now exclusive, now partially identical, now
coordinate, and now not coordinate. And this confusion extends
through the whole tissue of the methodology, at least the
Wundtian. For this author, after dividing all methods in two
(mutually-exclusive ?) groups, of gradation (Abstufung) and of
telling-off (Abzahlung), explains l that ' Among the telling-off
methods the method of mean errors is most nearly related in its
origin to the gradation methods or more particularly to the
method of minimal changes.' Indeed,2 ' The method of mean
errors arises from the method of minimal changes in case one
limits oneself to taking the just not-perceptible differences of
stimuli.' By this is presumably meant that the raw data in the
method of mean errors are identical with a part of the data in
the method of mean gradations ; and the fact is left out of ac-
count that the data are further quite differently treated in the
two cases, in ways which by no means ' arise ' the one from the
other. In short Wundt's systematic treatment of psycho-physic
method stops short of logically exact and consistent definition.

Yet these ambiguities are by no means peculiar to Wundt ;
firstly because most other modern treatises follow, or perhaps
even copy, the veteran psychologist, secondly because the same
ambiguities have prevailed from the very first. Thus while
Wundt finds the method of minimal changes merging into that
of mean errors, Fechner3 found that it 'goes over' into the
method of right and wrong cases ! The key to the situation is
this : the four psycho-physic methods are historical develop-
ments bearing the marks of their growth and of the accidents
they have met. The methods have not been defined, they have
been used; and where a user has had to modify a method he
has generally not modified its name, so that there are several
varieties under every method, and these bear the most diverse
and unsystematic relations to one another. So great has been

/., s. 473-

2 Ibid., S. 472.

3 Fechner, G. T. : Elemente d. Psychophysik, Leipzig, 1860, I Th., S. 75.

CLA SSIF/CA TION OF PS VCHO-PH YS/C MB THODS. 347

the dignity of tradition in this matter of method, that a thorough-
going revision and consistent systematization of the procedures
has never been achieved.

To systematize the procedures is not necessarily to analyze
or revise their mathematical details, — an ambitious performance
indeed, — but the frame-work of methodology can be recon-
structed in and for itself. This frame-work so recast will carry
with it undisturbed, as tent-poles the canvas, the vast multitude
of details.

To commence, one must first survey the methods in their
confusion, then either discover or postulate some one feature as
the sole essential of each method, and then deduce the logical
consequences. The Wundtian account is the most character-
istic and may well be made the starting point.

The methods fall in two groups, those of gradation and those
of telling-off. The former group has two classes, the method
of minimal changes and that of mean gradations ; the latter also
two, the method of mean errors and that of right and wrong
cases.

i. THE METHOD OF MINIMAL CHANGES.'

" In this method one seeks to determine at different points
on the scale of stimuli such a change in the intensity of stimulus
as produces a barely perceptible change of sensation." And,
in detail, one finds what interval between the standard stimulus
and a lower, compared stimulus is needed to make the two feel
different, and the same between the standard and a higher
stimulus ; and again the interval is found between standard and
lower or higher stimulus which will just not make the two feel
different. The average of these four intervals is the threshold
wanted.

This is good as a scheme, but on coming to the practice one
finds that there is no interval which just is or is not always felt
as a difference. There are intervals which are generally felt
as one or the other, but in order always to be so felt the interval
must be either so large or so small that it tells nothing about
the accuracy of discrimination which one is trying to find. The
scheme appears to be applicable if the compared stimulus can

'Wundt, op. cit., SS. 470, 476-479.

348 EDWIN B. HOLT.

vary about the standard continuously, for then one commences
with the compared stimulus plainly larger, say, than the stand-
ard and diminishes the former until it no longer seems larger ;
one calls this interval the upper threshold of difference, and
finds the other three desired intervals similarly. But the con-
tinuously applied stimulus tires the sense, while habit retards or
expectation hastens (presumably according to the temperament
of the subject) the moment in which the relation of the stimuli
is felt to change ; and it has not been proved that these factors
cancel themselves out. Thus the result yielded by such a pro-
cedure would be modified by the temperament of the subject
and the capacity of his sense-organ to resist exhaustion, and
would not be a pure measure of his discrimination.

If the compared stimulus is such as cannot be varied con-
tinuously then the procedure in question is, for all careful
work, out of the question, for the final result will be found to
be not a little dependent on the size of the step-wise gradations
which one arbitrarily has adopted in the series of compared
stimuli. And furthermore the case will often come up and
have somehow to be taken into consideration that one interval
will yield a perceptible difference, the next smaller will not,
while the next smaller than that will again do so.

In short, whether the compared stimulus is to be varied con-
tinuously or step-wise, it is advisable (as Wundt himself admits ;
S. 478) not to present these stimuli in their orderly progression,
but rather in a random succession. But ' this procedure has at
the same time the character of a combined method, since it ap-
proaches the telling-off methods.' Whether * combined ' or not,
it yields three kinds of judgments ( * greater ' ' equal ' and
' less ' ), and they show the arrangement which we have met
before in Fig. I.

What is now essential to this method of minimal changes ?
It is not the use of a standard stimulus and compared stimuli, for
the method of mean error also uses these ; nor is it the * mini-
mal' gradations in the compared stimulus, for the methods of
mean gradations and of mean error require these ; nor is it yet
the ascertainment of the intervals above and below the standard
stimulus which barely do not give a difference in sensation, for

CLASSIFICATION OF PSYCHO-PHYSIC METHODS. 349

the method of mean error involves also this. Essential to the

* method ' of minimal changes is only the project of finding that
difference between stimuli * which produces a barely percept-
ible change of sensation.' This, however, is not a method,
but a problem : furthermore the quantity so sought needs a pre-
cise definition (which the phrase quoted by no means gives) and
gets this only from the method finally fixed on for solving the
problem. Now the methods of evaluating data described by
Wundt as peculiar to the ' method' of mean gradations have been
shown to be inadequate for careful work ; and in order to be
precise, as Wundt himself advises, one must use a method which

* approaches ' (to put it plainly is) the method of mean error.
In a word, the ' method of minimal changes,' is not a method
at all, but a problem which one sets oneself. It can be solved
with precision only by such a procedure, and by any such, as
yields at least three classes of judgments grouped as in Fig. i.
How from these a solution is to be gotten is not told by the

* method of minimal gradations.'

As if so much confusion were not enough, the account pro-
vides only for the case of finding that change of stimulus which
produces a change of sensation, that is, the threshold of dis-
crimination ; whereas the case that one wishes to get the thresh-
old of sensation may properly come under one of the methods,
and certainly can come under no other than this of minimal
changes. Finally this threshold of discrimination is an average
or two thresholds, those of just perceptible and just impercept-
ible difference, each of which has often been made the basis
of investigation, although in either case the method has generally
been called minimal changes. Thus the ' method of minimal
changes ' not only is a problem and not a method ; it is not
even a definite, single problem.

2. THE METHOD OF MEAN GRADATIONS.'
This method, which Wundt substitutes for the older, some-
what more natural, and at least equally important • method of
over perceptible differences,' consists in finding a mean stimulus
which shall seem to lie equally far from two chosen extremes.
1 Wundt, op. cit.t SS. 471, 479-481.

35° ED WIN B. HOL T.

Two intervals are to be made equal. ' But in order to get
reliable results,' Wundt says, ' it is necessary to combine this
method either with that of minimal changes or with one of the
two telling-off methods about to be described.' In fact it ap-
pears that when the mean, variable stimulus is presented with
the two standards, it will seem to lie nearer sometimes the lower,
sometimes the higher, standard ; sometimes just half-way be-
tween. This gives rise to three groups of values for the mean
stimulus, those for which it is judged to lie more and less than
half-way and just half-way, between the two extreme stimuli.
And these three groups are arranged, once more, as in Fig. I.
If only a rough approximation to accuracy is wanted, the in-
formal procedure described under minimal changes can be used
in evaluating the data. But as has been shown, this procedure
is not permissable if it is a question of careful work. Rather
after the three groups of values shown in Fig. i have been
found, these must then be in some so far unexplained way eval-
uated ; in a way, presumably, which will ' approach,' ' resemble '
or ' shade off ' into one of the telling-off methods about to be
described.

In short, mean gradations are no more a method than were
minimal changes, but once more a problem. There the prob-
lem was to find the threshold of discrimination between sensa-
tions ; here it is to find the threshold of discrimination between
intervals ; there the just perceptible difference, here the just
equal over-perceptible differences (intervals), between sensations.

Wundt's account of this ' method,' which is doubtless designed
to intone the importance of Merkel's ' law,' is peculiar inasmuch
as it gives the impression that one must always take the extreme
stimuli as fixed and vary the mean stimulus ; whereas by far
the most work on over-perceptible differences has been done by
taking either the upper two or the lower two stimuli as fixed,
and the variable stimulus below or above them. When this is
done the data still consist in three groups of values for the vari-
able stimulus, according as the variable interval seems greater
or less than, or equal to, the .standard interval : and these
groups still shape themselves as in Fig. i.

Thus both of the so-called gradation methods are no methods,

CLA SS/F/CA T10N OP PS YCHO-PH YS/C MB TUODS. 35 1

but are problems ; namely, how to find objective values (meas-
uring the stimuli) which somehow correspond to just perceptible
or to equal over-perceptible differences, respectively, between
sensations. And either problem requires for its solution, firstly,
a procedure which will yield three classes of values of the vari-
able stimulus (corresponding to the judgments « greater,' * equal '
and ' less ') grouped as is shown in a general way by the curves
of Fig. i. Secondly, the problem requires a procedure for
deriving from these three curves a measure of the discrimina-
tion, respectively of difference between sensations or of equality
between over-perceptible intervals. The former procedure
affords the raw data, the latter evaluates them. In these two
processes will be found the real division of psycho-physic
methods.

3. TELLING-OFF METHODS : THE METHOD OF MEAN ERROR.'
This method * arises from the method of minimal changes in
case one limits oneself to taking the just not-perceptible differ-
ences of stimuli.' If this were all, then this method would be
not an independent method at all, but one of the subdivisions of
the so-called method of minimal changes. But it is not all.
For while the raw data group once more as in the curves of
Fig. i, we are now for the first time told how to evaluate these
data, and to get from the curves a measure of discrimination.
The procedure becomes here for the first time precise ; one
finds by definite rules the raw mean error, the variable mean
error, the constant mean error ; not to mention, as Wundt does
not, the probable error, the standard error or deviation, the
coefficient of variability, and the relative variability.1 The fine-
ness of discrimination is usually taken as the reciprocal of the
variable mean error ; or, to put it accurately, the discrimination
is defined as equal to the reciprocal of the variable mean error,
called for brevity's sake mean error. By other conventions the
other « errors ' are utilized more or less advantageously to give
light on the workings of the sense organs under investigation.

1 Wundt, op. dial., SS. 472-473, 481-482.

8 Cf. Yerkes, R. M. : PSYCHOLOGICAL BULLETIN, 1904, Vol. i, No. 5, p. 137
and Myers, C. S., Report of the Cambridge Anthrofwlog uat Expedition. Vol.
II., Pt. II., p. 212.

35 2 EDWIN B. HOLT.

But this variable mean error measures the sense discrimina-
tion not only for the case that one is seeking the just not per-
ceptible difference between stimuli, but also when one wants
the just perceptible difference (the quest which is confusingly
called the ' method ' of minimal changes), or when one wants
the just equal over perceptible differences or intervals (called
the method of mean gradations). In both these cases also the
variable mean error, with the method of finding it, is the pre-
cise definition of the terms * just perceptible difference ' and
' equal over-perceptible difference ' — terms which were so far
loosely used in stating the problems of minimal changes and
mean gradations. In short the mathematical manner of eva-
luating the raw data, which always group as in Fig. i, is the
only exact definition of the measure of discrimination which is
being sought.

The method of mean error is then both a problem and a
method. Its problem is to find the just not perceptible differ-
ence between stimuli. Its method is a mathematical affair of
averaging errors and of other details which we are not now
concerned with going into ; except to note that here at last is a
true method. But this method has no intrinsic affiliation with
the problem of not perceptible differences : it is equally neces-
sary in evaluating the data on just perceptible differences and
just equal over-perceptible differences or intervals. This is a
fact which the traditional classification quite obscures, although
the actual practice of psychologists will be found to accord
with that fact and to belie the tradition.

The three ' methods ' of minimal changes, mean gradations,
and mean error have amounted so far to three problems in three
kinds of discrimination, each calling for a special kind of data
to be gotten by experiment ; and then, one universally applicable
mathematical method for evaluating these data. The three
problems in discrimination may now conveniently be called —
just perceptible differences, orj.p.d. (from the method of mini-
mal changes), equal intervals that is equal over-perceptible dif-
ferences, or e.o.p.d. (from the method of mean gradations), and
not perceptible differences, or n.p.d. (from the method of mean
error). The one mathematical method for evaluating any of

CLA SSIFICA TION OF PS YCHO-PH YS/C MS THODS. 35J

these data, with its variations, will be called through the rest of
this paper the method of mean error, proper. One more prob-
lem may well be added — that of the threshold of sensation, or
/. ofs. (in comparison with which, as will be recalled, the dis-
criminations just mentioned are often named thresholds of
difference).

4. THE METHOD OF RIGHT AND WRONG CASES.'
This method consists in the repeated presentation to a subject
of two stimuli which are so nearly equal that they will frequently
be judged to be quite equal, and the greater be sometimes
judged even less than the lesser and conversely. Thus the
judgments will be sometimes right and sometimes wrong — a
fact which every author sapiently points out, as if the same
were not true of judgments found under every one of the other

V

\ /

• t

V

0
FIG. 2.

« methods.' But the essential fact is that, although these data
cannot be grouped like those of Fig. i, yet they are exactly
such a part of Fig. i as would lie in an ideally narrow vertical
section taken near the zero point of Fig. i. This is shown in
Fig. 2. That is, this method yields as much of the Fig. i as a
standard stimulus compared with only one other stimulus can
yield. The judgments gotten are that the compared stimulus

>Wnndt, op. cit.t SS. 473-4, 482-90.

354 EDWIN B. HOLT.

is equal to, greater or less than, the standard ; and in order to
complete these data into a figure like Fig. i, it would be neces-
sary only to bring in more compared stimuli, i. e., to use more
pairs. This was shown in the second paragraph of this paper.
Why sometimes many pair of stimuli are used, and why some-
times only two, does not for the moment concern us.

Now we have seen that the data of Fig. i can be evaluated
and made to yield a measure of discrimination, by means of a
method which we have called the method of mean error. The
method of right and wrong cases presents us, in fact, with a
second method of evaluating data into a measure of discrimina-
tion, and relatively fragmentary data at that, i. e., judgments
on but two stimuli. It is not the purpose of this paper to discuss
these mathematical methods of evaluation in their details ; and
it is sufficient to recall that a ' Pracisionsmass ' is derived from
the judgments on two stimuli by means of Gauss's formula for
the error curve, or more readily by the use of Fechner's table
of integrations derived from that formula. The method assumes
that the theory of errors may be applied to the mistakes made
in comparing two stimuli, that is, the ' wrong cases ' ; another
fact which is invariably emphasized, although the averages
taken in the method of mean error involve exactly the same
assumption.

The method of right and wrong cases is seen, therefore, to
be no new problem but in fact a new, real method of evaluation,
to be used (one is told) when for any reason the stimuli com-
pared have been only two. Right and wrong cases are gener-
ally used for finding a ' Pracisionsmass ' of the just perceptible
difference, but there is no intrinsic reason why the use of two
stimuli and Fechner's integral table should not give a measure
of equal over-perceptible differences, or even, with the standard
stimulus made equal to zero, of the threshold itself.

We have so far analyzed the four so-called psycho-physic
methods into four problems (not parallel with the original four
* methods ') and two real methods. The problems are those of
the j.p.d., e.o.p.d., n.p.d., and that of the threshold of sensa-
tion, or /. of s. The methods are those of mean error, m.
of m.e., and right and wrong cases, m. of r.-w.c. Let us

CLASSIFICATION OF PSYCHO-PHYSIC METHODS. 355

now see what is the relation of these two methods to each
other. The m. of m.e. is commonly recommended ; except
where the data are derived from the comparison of only two
stimuli, in which case the more cumbersome m. of r.w.c. has
to be resorted to. But this is not the real distinction between
the two methods. It will be recalled by experimenters that the
measure of precision yielded by the m. of r.w.c. is virtually
neither more nor less than the steepness of the curves shown in
Fig. i, or to be more exact, it is a numerical representation of
the steepness of the curve of ' greater,' or that of « less,' judg-
ments (for these two curves are assumed to be ideally similar)
when one half of the * equality ' judgments at every point has
been added on to this curve (Fig. 3). Now it is always in-

sisted that when the judgments are made on but one pair of
stimuli, the number taken must be very large indeed if the
measure of precision so derived is to be worth anything,
it is seldom if ever suggested, as was done in the second para-
graph of this paper, that by the method of least squares the
measure of precision may be calculated from more pairs of
stimuli and proportionately fewer judgments on each pair. And
the results in the two cases will be to all intents and purpoiet
identical. Now this would be applying the m. of r.w.c, to the
full data of Fig. i : which shows that the distinction between

ED WIN B. HOL T.

the methods of m.e. and r.w.c. is not one of the meagreness or
fullness of the data to be evaluated, that is, is not a question of
the number of pairs of stimuli used. But it is a question of the
degree of accuracy aimed at. The m. of r.w.c. gives a rela-
tively fine measure of precision, of which the mean error, prob-
able error, and other quantities of the m. of m.e. are the bares
and roughest indications. But so far as these last signifiy any-
thing, they suggest approximately the steepness of that same
curve (Fig. 3) of which the ' Pracisionsmass ' of the m. of r.w.c.
is a relatively accurate measure. The difference between the
two mathematical methods is thus purely one of degree of
accuracy, and it is a mere accident of technique that the m. of
m.e. cannot be used with judgments on only one pair of stimuli,
nor the m. of r.w.c. for the determination of not perceptible
differences. In fact the latter is possible if an experimenter
should choose to neglect a part of the wrong cases, taking only
those which form the curve of ' equal ' judgments (Fig. i), and
deriving the steepness of this curve by means of Fechner's
table. For this would be a permissible variation of the m. of
r.w.c., comparable to the several variations in the m. of m.e.
which have been used and recommeded.

It may well be questioned whether the m. of r.w.c. is not
a much more accurate procedure that the conditions of experi-
mentation ever justify ; or whether the application of least
squares would not be a case of penny-wise after pound-foolish,
in any sort of physiological work whatsoever. But we are not
here concerned with the mathematical minutiae of method, —
only with the general classification. The new tent poles are to
carry all the old canvas. Indeed if mathematical details were
here in question, it would be our first duty to examine and if
possible to justify the fundamental assumption of both methods,
i. e., that the theory of errors may be applied to curves which
never are and by all psycho-physic laws never can be, truly
symmetrical.

To survey our results once more, the four psycho-physic
' methods ' resolve themselves into the four problems, of finding
the threshold of sensation, and the thresholds of not perceptible,
just perceptible, and equal over-perceptible, difference; and

CLA SSIFICA T/ON OP PS YCHO-PH YS/C MB THODS. 357

then the real two methods of mean error and right and wrong
cases. The four original * methods ' are curiously illogical
classes, which have come about in the course of the develop-
ment of psycho-physics. They are historical relics. The first
two * methods ' are only problems ; the third is a problem and a
method ; while the fourth is a method but not a problem.

We have seen that in all psycho-physical experimentation
there are two stages of the work — the getting of data, and the
evaluating of them. We have already considered, so far as it
lies in our purpose, the second, purely mathematical stage. It
is instructive, and in part will account for the retention of the
false methodology, to note how the experimenter has to ap-
proach the first stage of his work, the getting of data. Sup-
pose that he wants to study by means of the ordinary olfactom-
eter and one of the four ' methods ' the discrimination for
odors. He cannot use the * method of mean gradations ' because
few if any subjects are able to identify and hold in mind an
over-perceptible difference between odors. So the experi-
menter thinks that he must choose another ' method ' ; the fact
it that he must choose another problem, that is, he must not
hope to measure the olfactory discrimination for over-percept-
ible differences. Similarly he will believe himself deterred
from the « method of mean error,' because he has been taught
that this « method ' involves the adjustment of the stimuli by the
subject himself : and of course an odor-tube cannot be adjusted
back and forth like a monochord. He may think of doing
himself, as experimenter, the adjusting, continuing each time,
although this is contrary to the school directions, until the sub-
ject declares a just not perceptible difference. This is possible
to do, but hardly advisable ; since it happens by an accident of
physics that the olfactometer is more suitably and naturally
adjusted from less olfactory stimulus to more, rather in the
opposite direction. It is an accident, then, that the experi-
menter cannot well study just not perceptible differences, but
must choose the remaining problem, the « method ' of just per-
ceptible differences.

It may be said that this necessity of casting about for a
problem (n.p.d.tj.p.d.% e.o.p.d.) suited to the accidental pccu-

358 EDWIN B. HOLT.

liarities of the sense-organ to be studied and the apparatus
at one's disposal, has made the spurious distinction between
' methods ' seem real and practical to experimenters who seem
not to have noticed that the distinction is in problem and not in
method.

Thus the first stage of psycho-physical experimentation,
the getting of data, is practically not determined by the experi-
menter, but by relatively accidental circumstances, — peculiari-
ties of the sense-organ chosen and of the apparatus available.
And in determining what data can be gotten, these accidents
determine at the same time what problem can be studied, that
is, whether n.p.d., j.p.d, or e.o.p.d. The word 'accident'
may be objected to. Yet it is fair to call accidental the circum-
stance, for instance, that the n.p.d. cannot be studied in the
active muscle-sense. Such seemingly chance and irrelevant
factors are almost numberless. An important one of them,
although it influences the general problem less directly, is the
possibility of individual or massed stimulation of end-organs.
By an anatomical accident the olfactory end-organs cannot be
stimulated individually ; so that what is an interesting problem
of the dermal senses, the relative thresholds of different indi-
vidual end-organs, cannot be studied in the sense of smell. It
is apparently an accident, though an interesting one, that some
sense-organs, as the olfactory, gustatory, or dermal when
stimulated singly, do not afford us clear sensations of intervals,
i. e., of over-perceptible differences. Hence their power of
discrimination must be studied in other respects ; the problem of
e.o.p.d. is debarred. Once again, if the experimenter happens
to have two tuning-forks and nothing else, he will necessarily
adapt his problem to the m. of r.-w.c. with which he will have
to evaluate his data. Whereas if he had a sonometer, he would
more naturally let the subject adjust for himself and give judg-
ments of n.p.d. The case in which the investigator has but
two different stimuli to apply (and these must be nearly alike)
is the one case in which truly the method as -well as the problem
is determined by irrelevant circumstances. Otherwise the
method is chosen (m. of m.e. or m. of r.w.c.) according to the
degree of accuracy which is desired.

CLASSIFICATION OF PSYCHO-PHYSIC METHODS. 359

Aside from the accidents which determine the problem, there
is another kind which influences only the technique or at best
bears but remotely on the problem. Such a factor, for instance,
is the matter of simultaneity or succession of stimuli. Since the
olfactory end-organs have to be stimulated all at once, any dis-
crimination which is studied has to be a successive one (apart
from the very doubtful case which some would claim, of simul-
taneous bilateral stimulation). This circumstance would still
leave as possible any of the three problems ; but it helps to re-
strict more precisely the technique to be adopted. There are
countless other accidents of a similar sort.

Therefore in first approaching a psycho-physical problem,
and in trying to get it realized in some arrangement of appa-
ratus, the experimenter finds that very little is left to his free
choice. Sometimes indeed he may choose no more than barely
the sense which he studies. Two kinds of accidental circum-
stances, as we have seen, restrict his course, although in prac-
tice it is scarcely necessary to distinguish between them. In
order to accept the inevitable, and set up his apparatus with the
least waste of time and thought, the experimenter must run
through the possible limitations and find out definitely what ones
actually confront him. The possibilities are fairly well included
under the following categories, although the list aims merely at
being serviceable but not exhaustive.

1. Comparison — (a) simultaneous, (6) successive.

2. Comparison — (a) direct (immediate), (6) mediate.

3. Comparison between — (a) two stimuli, (6) more than two
stimuli.

4. Variation — (a) random, (V) progressive.

5. Variation — (a) continuous, (6) discontinuous (step- wise).

6. Apparatus operated by — (a) the experimenter, (£) the
observer.

7. Actual relation of the stimuli — (a) known to the observer,
(b) not known to the observer.

8. (If the sense to be studied and the above enumerated
conditions are such as still to leave the question open): Dis-
crimination of — (a) threshold of stimulation, (£) not perceptible
difference of stimulation, (c) just perceptible difference of stimu-
lation, (d) equal over-perceptible differences of stimulation.

3^0 EDWIN B. HOLT.

These headings are all familiar to the psychologist and need
no elucidation. They are neither completely independent nor yet
mutually exclusive. Class i, for instance, is independent of 2 ;
but 3, #, excludes 4 and 5. After learning what of these alter-
natives are open, the experimenter will see his way of proceed-
ing rather precisely marked out. Herewith the first methodo-
logical stage is ended. Before coming to the second stage, of
mathematical evaluation, he has only to get his data.

In regard to this second stage we have already seen that
there are only two ways of evaluating data (although each
method allows some minor variation), the methods of mean
error and of right and wrong cases. It was not the purpose of
this paper to discuss the methods in detail, but only to analyze
the so-called four * methods ' and to classify the results. This
we have now done. It need only be noted once more, that the
choice between the two actual methods (except in the case of
only two stimuli being used) depends on nothing but the degree
of accuracy which is desired, that is, on the amount of labor
which the experimenter thinks proper to devote to the inquiry.
We may now pass to two of the other emendations of the tradi-
tional methodology, which have been offered.1

1 The writer greatly regrets that before this article was actually set up, he
had not seen the admirable work of G. E. Miiller, in the Ergebnisse der Physi-
ologic, 2ter Jahrgang, 1903, II Abth., SS. 266-516. Miiller finds four cases
(Falle) in which psycho-physic methods may be used. These are our /. of s.,
j.p.d., n.p.d. and e.o.p.d. Of 'methods' he finds three; the first, in which
the observer adjusts the variable stimulus in random sequence ; the second, in
which the experimenter does this, but in orderly increasing or decreasing pro-
gression ; and the third, in which the stimuli are not adjusted (old method of
right and wrong cases). There then remains the task (Aufgabe) of finding in
general two values, a mean and its variability. This may be done immediately
(our m. of m.e.), or by the mediation of formulae (our m. of r.zv.c.).

Of course this is in essentials far nearer to that which has been urged above
than is any other classification hitherto offered. To the present writer it still
seems, however, that the ' cases ' are problems of which accidental circum-
stances largely determine the choice ; that Miiller's ' methods ' are merely
three among a large number of equally important such accidents ; and lastly,
that the actual methods are Miiller's two Aufgaben, the treatment of the results
either with or without formulae. But only the second point is a difference in
principle, while the first and last are merely nominal.

CLA SS1FI CA TION OP PS YCHO-PH YSIC ME THODS. 36 1

THE CLASSIFICATION OF KULPE.*

Kiilpe's analysis of the methods is not fundamentally dif-
ferent from that of Wundt, but there are superficial differences
which are worthy of a brief consideration. For Kulpe there
are two groups of methods, that of minimal changes and that of
errors ; these two correspond to Wundt's gradation and telling-
off methods. The method of minimal changes has four * appli-
cations'; while the error methods are two — the method of
right and wrong cases and the method of mean error.

The method of minimal changes may be * applied ' to the
determination of threshold (Reizbestimmung), to the comparison
of stimuli (Reizvergleichung, the n.p.d. mentioned above), to
the determination of difference (Unterschiedsbestimmung or
j.p.d. above), and to the comparison of differences (Unter-
schiedsvergleichung or e.o.p.d. above). This virtually admits,
although Kiilpe seems unconscious of the fact, that these four
groups are not methods but problems, as has been argued in
this paper. Furthermore his classification is symmetrical and
consistent : /. e. —

1. Threshold — (a) of sensation, (b) of interval between
sensations.

2. Equality — (a) of sensations, (b) of intervals between
sensations.

Kiilpe well says that there is one method, which he calls the
* method of minimal changes, that applies to these four classes.
This ' method ' is essentially like the procedure described by
Wundt under the same name, and is subject to the objection
which was noted in the early part of this paper. This is, as
will be remembered, that whether the compared stimulus varies
continuously or step-wise, there is no point at which the change
in judgment from being always * greater ' or always • less ' to
being for the first time « equal,' or vice versa, is truly significant.
For let us suppose the compared stimulus to be decreasing
toward the standard, it may be considerably greater than this
and be judged « equal ' while when further decreased it will be
again judged « greater.' Or if it is increasing to the standard,
it will often be judged ' equal ' when considerably less than the

1 Kulpe, O., Grundriss der Psychologic, Leipzig, 1893, SS. S5-*»-

362 EDWIN B. HOLT.

standard, but again ' less ' when it has been increased nearly to
equality with the standard. To ignore these inconsistencies
and to interrupt the comparisons with the first judgment of
equality which is given (or inequality, as the case may be) is,
if the compared stimulus varies step-wise, to take a measure-
ment of discrimination which depends materially on the size of
the steps which have been arbitrarily chosen. If the compared
stimulus varies continuously, the case is a trifle better but not
much, since then the measure of discrimination is considerably
vitiated by fatigue and expectation (see above). Furthermore
it is often not possible to-make the compared stimulus vary con-
tinuously.

For these reasons Wundt virtually admits, as we have seen,
that here is no method ; and of his ' method of minimal changes '
he leaves, although merely by implication, only the problem of
the j.^.d. Kiilpe, however, accepts the informal procedure
disparaged by Wundt, and insists that it is a method. It has
been asserted elsewhere in this paper that every psycho-physical
method must involve a procedure which yields at least a part of
the data shown in Fig. i, and must then mathematically evalu-
ate from these some sort of a measure of discrimination. Now
what part of Fig. i does this procedure yield which Kiilpe
recommends ? And how are these data evaluated ? The
* method ' is to present to the subject a pair of stimuli, a standard
and a compared, for his judgment of * less,' ' equal ' or * greater.'
The compared stimulus varies in successive presentations not at
random but so as to approach or depart from the size of the
standard stimulus ; and it may do both, either above or below
this standard. Thus there are four modes in which the com-
pared stimulus can vary. Whether in a given experiment some
or all of these modes are used depends on the problem to which
this ' method of minimal changes ' is being applied.

Let us suppose, for example, that it is being applied to the
problem of j.p.d. All four modes are used. The compared
stimulus (c.s.) starts so much smaller than the standard stimulus
(5.5.) that it is always judged smaller ; and is gradually increased
until the first judgment of equality is gotten. Here the pro-
gressive comparisons are interrupted, although if c.s. were

CLA SSIFICA TfON OF PS YCHO-PH YS/C MS TI/ODS. 363

further brought up toward 5.5. it would very likely be judged
less once or twice more. Here is of course the weakness of
this method. Now c.s. is taken equal to *.$., and made gradu-
ally to decrease until the first judgment of less is given. So far
the experimenter has a series of judgments * less ' ' less ' et cet.
— 'equal, 'for c.s. increasing; and a series * equal' « equal' et
cet. — 4 less,' for c.s. decreasing. In both cases c.s. is smaller
than 5.5. But it can increase and decrease while larger than
5.5. Therefore two more similar series are gotten, for values
of c.s. above 5.5. Curiously enough, Kiilpe writes as if it were
enough to take the average of only these four values of c.s. at
which the first change in the judgment occurred.1 But he can

0
FIG. 4.

not mean so, for it appears later,2 that a mean variation may be
in question ; so that of course, as one would expect, the four
series are to be gotten many times, and an average of all taken
for the actual measure of discrimination. The reader will tee
at once that if plotted these raw data would look like Fig. 4.

The relation to Fig. I (the dotted lines) is clear enough.
Where the differences between c.s. and 5.5. are so very small or
so very large as always to be correctly told, the curves run
smoothly. They become jagged where the chance errors inter-
rupt the various series of ' less,' « less,' et cet. — , / equal/

1 Kiilpe, op. cit. SS. 60-62.
»/M</., SS. 66.

364 EDWIN B. HOLT.

* equal ' et cet. — , or ' greater,' * greater ' et cet. — , with a
change in the judgment uttered. Now, as we have already
seen, it is just because the first change in judgment depends
more on the size of the step variations of c.s., on fatigue or ex-
pectation, than on the fineness of discrimination in question that
this method, as even Wundt grants, is not a proper method. It
selects for further mathematical evaluation a certain portion of
the data of Fig. i ; but the principle of selection is largely,
though not utterly, independent of the power of discrimination
which is being studied. And whereas in the m. of r.iv.c. when
only two stimuli are used, the selection although arbitrary yet
leaves enough of the essential parts of the curves ; the principle
of selection here used effectually excludes just these indispen-
sable parts, namely the parts which show the characteristic
steepness of the curves.

Therefore in the earlier part of this paper the name of
method was justly denied to this procedure, since it is abso-
lutely inadmissible. And it was insisted that the only allowable
and yet mathematically convenient method was to take data
giving the complete curves of Fig. i, and then to express ap-
proximately the steepness of those curves (transformed perhaps
as in Fig. 3) by means of a mean error, probable error, or by
some such readily obtainable quantity. Kiilpe's method of
minimal changes is simply under no circumstances allowable.
The merit of this part of his classification lies wholly in his
four-fold ' application ' of the * method of minimal changes.'

The other part of his classification needs but brief considera-
tion. As to his ' method of right and wrong cases,' he well
says : l That it ' is capable of as manifold application as the
method of minimal changes ' ; but then he goes on to describe
the method so wholly in the traditional way that one has the
impression that the use of only two stimuli is as essential to this
method as is Gauss' equation itself.

Kulpe's treatment of the * method of mean error ' adds noth-
ing to that of Wundt, except the statement that while the
method has been used only for the n.fi.d., it could be used for
the e.o.-p.d. as well. But he declares (S. 78) that it can be used

1 Ibid., S. 70.

CL A SSI PICA TION OF PS YCHO-PH YS/C ME THODS. 365

only when experimenter and subject are one person, and when
the stimulus can vary continuously. Both of these statement*
we have seen to be untrue.

THE CLASSIFICATION OF EBBINGHAUS. l
• The other classification which will concern us in this paper
is that of Ebbinghaus, given after his short and admirable ac-
count of the traditional four methods. Ebbinghaus says (S.
75) that, ' The goal of psycho-physic method is the determina-
tion of those stimulation values which are the outward causes
of equal-seeming psychic values (that is, of equal intervals be-
tween sensations),' of course, then, the threshold of stimulation
is not a problem of psycho-physics ; and in fact the author makes
no mention of it.

Now as to the size of the intervals between sensations, it
can be either just perceptible or over-perceptible, and if the
latter it can be as large as one chooses. Herewith our prob-
lems oij.p.d. and e.o.p.d. are recognized, although still called
methods ; while the problem of n.-p.d. is ignored. Secondly,
as to the way of judging, this may be so chosen that the sub-
ject (S. 75) ' has in mind the idea of an equal interval and
varies the outward stimulus until the sensation which it gives
corresponds to this idea : or, one can present repeatedly for
judgment a given pair of stimuli and let the subject give his
judgment in terms of certain replies previously agreed on,'
such as * less,' ' equal,' and * greater.' The first procedure is
called the * finding a stimulus ' to correspond to a given judg-
ment; the second, the ' finding of judgment* to correspond to
a given pair of stimuli. Lastly, in every case the final result
must be the average of many observations ; and since each ob-
servation will vary from this average, the mean of the variations
must always be given (mean deviation), in order to show the
reliability of the total average.

This scheme is very original and suggestive ; but inade-
quate. It recognizes but one true method, the one which we
have called above the m. of m.e. ; and but two problems (which
Ebbinghaus still calls methods) to which this one method can

1 Ebbinghaus, H., GrundzUge der Psychologic, Leipzig, 1902, SS.

366 EDWIN B. HOLT.

be applied; those are our problems of j.p.d. and e.o.fi.d. The
problem of threshold of stimulation is dismissed as not lying
within the field of psycho-physics. The expediency of so
narrow a definition of psycho-physics is doubtful ; for it should
seem better to account the problem of psycho-physics the meas-
ured correlation in general of stimulus and sensation ; and thus
to include the problem of threshold, which after all must be
studied if at all by the m. of m.e. or that of r.-w.c. Also the
problem of n.p.d. finds no recognition because, as the author
says (S. 78) : 'The third method, finally, that of mean error [that
is the third traditional method, by which he refers to the prob-
lem of n.p.d.], affords nothing directly which can be utilized
as a measure of sensation, since it operates not with a sensa-
tion interval, but with the disappearance of an interval'; i. e.,
it operates with a not perceptible interval, which is seemingly
not to the point. And again he says (S. 69) : ' The determina-
tion of this mean error [made in judgments of n.p.d.~\ through
the range of a given sense has of course its use, but clearly
these values are something quite different from equal sensation
intervals in the above mentioned sense [that is, of just percep-
tible, and equal over-perceptible, intervals] , and the process of
getting them is no true measuring of sensation but a process
having some relation to such a measuring.' Ebbinghaus gives
elsewhere his grounds for this view, in that the n.p.d. never
had any value except on the assumption that it bears a definite
and fixed relation to thej.p.d. from the same standard stimulus.
This assumption has not been shown to be valid, and Ebbing-
haus seems to account it very speculative if indeed not certainly
false. He may be quite right. The point is one more of
mathematical technique than of classification. If he once
granted the validity of the n.p.d., Ebbinghaus would doubtless
class it with \hej.p.d. and e.o.p.d.

The analysis given in this paper found the third so-called
* method ' of tradition to consist in a problem and a method.
Ebbinghaus discards the problem but retains the method, which
is our m. of mean error. This is in fact his one method, and
it must be used in every one of the four procedures into which
he resolves the traditional ' methods.' These are, once more :

CL A SS/F/CA TtON OF PS YCHO-PH YSIC ME THODS. 367

1. Just perceptible interval with finding of stimulus.

2. Just perceptible interval with finding of judgment.

3. Over-perceptible interval with finding of stimulus.

4. Over-perceptible interval with finding of judgment.
Thus in every psycho-physical procedure an average of the

observations must be taken, with their mean variation or error.
These two quantities are the measure of discrimination which
is desired.

But what of the * Pracisionsmass ' as yielded by Gauss's
equation and Fechner's table of integrals? Ebbinghaus mini-
mizes in his system this method of evaluation because, it should
seem, he accounts it a very laborious process to be used only
when all the observations are based on but two stimuli : the
more cumbersome mathematical evaluation being used solely in
order to make up for the poverty of the material of observation.
The author does not admit, apparently, that here is an evalu-
ation method of relatively great accuracy, which can be applied
as well to observations on many pairs of stimuli as to those on
only one pair.

We have seen that from the first and second traditional
methods Ebbinghaus analyzes out his two categories of just
perceptible and over-perceptible, difference. Both of these
* methods ' proceed by '•finding the stimulus J whereas the fourth
' method ' (of right and wrong cases) proceeds by '•finding the
judgment ' ; hence these two new categories. Every psycho-
physical procedure uses one of the last two, together with one
of the first two categories. Thus arise the four methods of
Ebbinghaus, given on the preceding page. But now all four
of these methods have to use the little germ of truth which lay
in the traditional third « method ' (of mean error) : this was the
use of the average and its mean error. Wherefore the final
form of Ebbinghaus's four methods is as follows :

1. Just perceptible interval with finding of stimulus ; average
of all observations with m.e.

2. Just perceptible interval with finding of judgment; aver-
age of all observations with m.e.

3. Over-perceptible interval with finding of stimulus ; aver-
age of all observations with m.e.

ED WIN B. HOL T.

4. Over-perceptible interval with finding of judgment ; aver-
age of all observations with m.e.

Thus this system, which is somewhat simpler than the tradi-
tional four, is gotten by the omission of two problems, those of
the threshold of stimulation and of the n.p.d. ; and by the
omission of one method, that of right and wrong cases. It is
curious that aside from the relatively special and insignificant
expedient for evaluating judgments on only two stimuli, Eb-
binghaus finds nothing in the traditional method of right and
wrong cases save the suggestion to fit the judgment to the
stimulus instead of the stimulus to the judgment, as in minimal
changes.

Although this classification leaves out so much that it becomes
inadequate to the subject, it probably has the distinguished
merit of being the first radical and strictly logical recasting of
the methodology. It is well worthy of study, and will be found
to be possibly, the clearest and best form for teaching the meth-
ods of psycho-physics to elementary students.

SUMMARY.

The four traditional methods of psycho-physics are found
to be an illogical scheme of a subject which if analyzed re-
solves itself into the following system.

I. Four problems as to the relation between stimulus and
sensation, which admit of quantitative treatment.

(a) The threshold of stimulation (/. ofs.}.

(d) The not perceptible difference of stimulation (n.p.d.).

(c) The just perceptible difference of stimulation (j.p.d.').

(d) The equal over-perceptible difference of stimulation
(e.o.p.d.).

II. A great diversity of procedure by which data on these
problems are gotten (see above), and of which the one thing
essential is that judgments shall be obtained which group them-
selves like a part or like the whole of the judgments in Fig. i,
and so that the steepness of at least one of the curves of Fig. I
is implicity contained in the judgments. This variety of pro-
cedure is not a variety of choice open to the experimenter.
The procedure used in any case depends on relatively acci-

CLASS I PICA TION OF PS YCHO-PH YSIC MB THODS. 369

dental characteristics of the sense-organ chosen for study ; and
for this reason they are not susceptible of rigorously logical classi-
fication. Furthermore it generally happens that these different
accidents, so far from being a source of freedom, actually re-
trict the experimenter in his choice of problem, so that he is
obliged to study a particular one of the four problems.

III. Two real methods by which the judgments can be
evaluated into a measure of the discrimination of the sense
studied : that is, by which the steepness of one or all of the
curves in Fig. i can be approximately or accurately expressed.
These may be called, out of respect to tradition.

(a) The method of mean error (m. of m.e.).

(b) The method of right and wrong cases (m. of r.w.c.).
The difference between these is one of accuracy. The former

method measures the precision roughly by taking an average of
the observed values and their mean error, or probable error, et
cet. The latter method is more refined and uses the equation
of Gauss and the integral tables of Fechner to obtain a ' Pra-
cisionsmass.' The significance of either of these measures lies
in its being an index of the steepness of one or all of the curves
of Fig. i.

Either of the two methods may be used with any of the four
problems. There are thus eight alternatives. Having chosen
the sense-organ which he will study, the experimenter finds his
mode of procedure limited in many respects by accidental pecu-
liarities of this sense and of the apparatus at his disposal. He
may then choose what of the four problems he will study, or
he may find even this determined by the accidental peculiarities.
He is, however, free to choose his method of evaluation : and
the basis of his choice is nothing but the degree of accuracy
which he desires in his results or, what generally comes out to
the same thing, the amount of labor which he is willing to spend
on them. The one exception is the case in which some cir-
cumstance limits him to the use of only two stimuli, for then i
limits him also to the m. of r. w. c.1

1 The MSS. of this articlewas received April 17, 1904.— ED.

STUDIES IN THE INFLUENCE OF ABNORMAL
POSITION UPON THE MOTOR IMPULSE.

BY DR. CHARLES THEODORE BURNETT.

I. THE JAPANESE ILLUSION AND THE MIRROR ILLUSIONS.

The experiments that form the basis of the following report
are the first of a series designed to open a new approach to the
psycho-physics of the motor impulse, by way of the modifica-
tions that occur in the control of a limb when placed in unusual
positions. The particular investigations of this paper are con-
cerned with the direction of the impulse as shown in the control
of the fingers when the hands are placed in unusual positions ;
concerned, in other words, with the ability to move a given fin-
ger at command.

We shall consider first the Japanese illusion. It occurs
when, with arms crossed, the hands are clasped thumbs down,
and are turned thumbward till they point up. If an onlooker,
pointing to one of the fingers, asks the man thus situated to
move it, the latter is frequently unable to do so, moving, if any-
thing, some other finger. Of the experimental conditions, it
need be said only that the hands were unclasped after every
movement in many series ; and that either the wrists and neigh-
boring parts of the arms were concealed by a cloth wrapped
about them, or the observer was covered with a sort of apron
fastened about the neck and having an opening with edges
drawn together by an elastic cord. Through this opening the
clasped fingers could be thrust while most of the remainder of
the hand was concealed. The purpose of these precautions was
to preserve as long as necessary an illusion that yields pretty
readily to experience of the situation. In the first form of this
experiment the finger to be moved was indicated visually to the
observer by pointing, whereupon the latter was to make the
movement as quickly as possible. No attempt was made to
eliminate the possible influence of the crossing of hands,
whether right over left or left over right. But here and through-
37°

ABNORMAL POSITION AND MOTOR IMPULSE. 371

out, except in a single instance noted in its place, each finger
(including thumbs) was as often required to move as is any other.
The order of choice was irregular.

TABLE I.

JAPANESE ILLUSION.

i
&

S5 M

*\
o »•

HW

Symmet.
Opposite.

1

Next

Finger
Opposite

m

*£«

Mi»ce11.

OppodU.

MiKell
Same.

i

«

§

Baldwin.

9 Oct.1

40

22

IS

5

I

I

16 "

60

M

5

5

4

7

7

23 "

20

I

I

i

Emerson.

28 "

60

29

16

2

4

5

2

U

>S

4 Nov.

80

35

19

I

8

7

8

rj

Kleinknecht.

3 Dec.

40

19

15

4

8

it

Miller.

28 Oct.

60

23

14

I

6-

2

3

20

4 Nov.

60

15

12

i

2

2

13

Rouse.

16 Oct.

40

20

13

4

I

2

3

17

23 "

40

19

16

3

'9

30 "

60

25

21

I

3

3

22

Rowland.

28 "

40

14

7

2

4

I

5

9

4 Nov.

60

26

16

4

6

4

22

Totals.

620

240

154

9

37

35

2

3

57

ilT

RESULTS.

1. The disturbance in the direction of the motor impulse is
rather large, as shown by the proportion of errors to the num-
ber of experiments.

2. In some cases adjustment to the abnormal position is not
long in occurring. Miller and especially Baldwin show this.
The latter, in a short test made after those recorded, showed
entire readjustment.

3. The erroneous movements far more frequently occur in
the finger symmetrically opposite than in any other. Following
at a long distance are erroneous movements in the next fingers
opposite and the next fingers on the same side. Errors of
other types are few and scattering.

4. The errors occur far more frequently when the move-
ment is to be made with the left hand than when it is to be
made with the right. No observer shows a contrary tendency,

though some exhibit none.

1 Omitted from totals because fingers were not equally employed.

* In these columns throughout the tables are recorded the number of fail-
ures for the hand in question.

372

CHARLES T. BURNETT.

5. With hands in normal position, palms up, there are
practically no errors.

6. When a finger is touched as well as pointed out, there is
almost never an error.

7. When the eyes of the observers were closed and the fin-
ger to be moved was indicated by naming it, the following re-
sults were obtained :

No. Exp.

Errors Symmet. Opp.

Errors Miscell.

Baldwin.

40

o

5

Emerson.

30

0

o

Kleinknecht.

3°

0

4

Miller.

60

9

4

Rouse.

40

8

5

Rowland.

40

7

0

Here is a great reduction in the illusion. That this is not
due in all cases merely to a growing familiarity with the situa-
tion is shown by the results of Emerson. Work in connection
with this illusion previous to the present test had not occurred
for four weeks. The day after this test the old conditions were
restored and the illusion was back as strong as ever. The
results of Miller show about as much illusion as in one of the
sets of experiments recorded against him in Table I. Hence it
seems possible that his confusion lay in his kinsesthetic knowl-
edge of where his fingers were located. This confusion nearly
disappeared for him when the hands were laid palm downward
on the table pointing away from the body, while the other con-
ditions of auditory stimulus and closed eyes were maintained.
The presence of a weakened illusion with Rouse is perhaps
connected with the fact that he visualized his hands. This
Miller did not do.

CONCLUSIONS.

1. It appears from (6) and (7) above that the illusion lies in
the visual, not in the kinaesthetic, experience of abnormal posi-
tion, though one observer presents a possible exception.

2. The large excess of erroneous movements made with the
ringer symmetrically opposite shows how large a factor in the
direction of the motor impulse is the visual peculiarity of a given
finger. The motor current appropriate to that peculiarity is

ABNORMAL POSITION AND MOTOR IMPULSE. J73

started ; but the element contributed by visual position divert* it
to the wrong hand.

3. Why, it may be asked, does the visual experience of ab-
normal position divert the current far more frequently to a fin-
ger on the opposite hand than to another on the same hand? A
glance at the hands in the position appropriate to the illusion
will show that the roots of the fingers lie on the side opposite
to the arm to which they belong ; that the right-hand fingers
point from the left to the right, and the left-hand fingers from
the right to the left. This is just the reverse of what is true
when the hands are clasped in the usual way. Going upon the
basis of procedure in the normal situation, the observer in the
unusual position moves the finger that really lies on the side on
which the given finger appears visually to lie. This process of
reasoning is, of course, wholly in the mind of the experimenter.
For the observers the process is so mechanical that they are
obliged to consider seriously when asked how they obey a given
command. The usual reply is that they simply see what is wanted
and then do it. The movement appears to follow directly upon
the visual cue. It is not to be denied that the observers feel in
some measure confused in this unusual position and occasion-
ally feel almost unable to move any finger. The attitude of
hasty attention that favors so many geometrical-optical illusions
seems to be the best one in the present instance. The confu-
sion soon yields far enough to permit a movement that is not
merely spasmodic.

4. Why any correct movements at all? They become pos-
sible by a new adjustment to the new position — a recognition that
the right-hand fingers point from the left and the left-hand fin-
gers from the right. Some effort may be required to substitute
the new visual cue for the old, and, when effort fails, habit
steps into control. The new adjustment may be but partly suc-
cessful and a wrong finger moved on the correct side. The
mistakes of this sort give the second maximum of errors.

5. Is there any psychological account to be given of the sec-
ond focus of errors in Table I., viz., in movements of the fingers
next to the correct one whether on the same or the opposite
hand? This is possible if in some way it could be shown that

374

CHARLES T. BURNETT.

the two fingers resembled each other. The middle and ring
fingers resemble each other more than do any other two adjacent
members, and the thumb and forefinger least of all. Here is
the way in which the errors were distributed among adjacent
pairs :

Thumb and fore-finger = 19
Fore- and middle-finger = 19

Middle and ring = 25
Ring and little = 10

There is no ground here for basing the error wholly on mutual
resemblance, though to this it may at times be due. We seem
driven to a purely physiological account.

6. That the second greatest tendency to error should involve
moving a finger next to the correct one, while yet this tendency
cannot be due in general to resemblances, suggests that what
would be the habitual course of the motor impulse is preventing
somehow a wider divergence in its actual course. It does not
appear otherwise why the errors should not be more widely dis-
tributed.

7. The source of superiority in control of the right over that
of the left hand does not at once appear. For movements so
simple in the normal position such a difference does not exist.

TABLE II.

JAPANESE ILLUSION.
Left hand crossed over right.

No.
Exper.

Errors.

Symmet.
Opposite.

Next
Finger
Opposite.

Next
Finger
Same.

Right.1

Left.1

Emerson.

40

23

15

8

14

9

Kleinknecht.

40

14

14

6

8

Rouse.

40

5

3

2

5

Rowland.

40

16

II

3

2

2

14

Totals.

160

58

43

13

2

Right hand over left.

Emerson.

40

19

14

5

6

13

Kleinknecht.

40

17

16

i

3

14

Rouse.

40

12

10

i

i

5

7

Rowland.

40

18

14

3

i

5

13

Totals.

1 60

66

54

10

2

1 Totals henceforth are not recorded in these columns because of the diver-
gence among observers.

ABNORMAL POSITION AND MOTOR IMPULSE. 375

One objective factor not thus far controlled might be involved
in this result. The abnormal position studied here can be
obtained by crossing right hand over left or left over right.
There is frequently a difference in strain in the two wrists ; and
the hand and wrist of more intense sensation might possibly be
under better control. So much is at least suggested by the les-
sening of error when the control was of the auditory-kimesthetic
type. Or we might indeed find the reverse to be true. Table
II. gives us the results of experiments similar to the preceding,
but designed to show the effects, if any, of the method of cross-
ing. Baldwin is omitted in this test because the illusion had
nearly disappeared for him.

RESULTS.

1. There are still many more failures in case of a com-
manded movement with the left hand than with the right. There
is but one observer whose results suggest any influence of the
method of crossing. No simple relation is apparent between
the presence or absence of a feeling of strain, as reported by
the observers, and this particular tendency to error. So the
cause must still be sought.

2. The distribution of errors is like that in Table I, except
that all scattering errors have disappeared and very few are
found in the next finger on the same side.

MIRROR EXPERIMENTS.

In the following sets of experiments the abnormal position
was attained by the use of a mirror, occasionally of two. The
mirror space inverts the spaces of the real world in a direction
perpendicular to the plane of the mirror ; so that fingers in front
appear to be in the rear, and those to the right lie apparently on
the left ; and vice versa. A direct view of the fingers was pre-
vented by a broad collar of cardboard. After a few of these
experiments had been made it was thought best, to the end of
preserving the illusions in force, that the observers either close
their eyes or look away after noting the finger to be moved, and
then complete the movement. They were forbidden, however,
to develop any new sources of information after closing the

376 CHARLES T. BURNETT.

eyes. The general conduct of the experiments was as before
except that the thumbs were not used, since in some positions
they could not be conveniently interlocked with neighboring
members.

The results are so arranged in the tables as to show the
extent to which the errors follow the mirror reversal. To illus-
trate— when the hands are clasped palms up and the line of the
interlocked fingers is perpendicular to the plane of the mirror,
the forefingers which are really farthest from the body will in
the mirror space be nearer the real body, while the little fingers,
which are really nearer, will in the mirror space be farther
away. If now the observer be directed by pointing to move a
forefinger and he thereupon move the little finger or ring finger,
that error would show that the movement followed upon the
visual cue, the mirror space being regarded not otherwise than
as real space. An erroneous movement of fore- or middle-
finger for either ring- or little-finger will be classified thus ; not
so fore- for middle-finger or vice versa, nor ring- for little-
finger. In other words, the eight fingers, being interlocked, are
divided by a median line into two sets. The finger wrongly
moved must not lie in the same half with the finger pointed out,
if the error is to be classed as following the mirror reversal. If
it does lie in the same half, one cannot say that the error is not
due to the same cause. But by arbitrarily limiting the evidence
to the more striking cases, a preponderance of these will make
our conclusions much stronger.

By way of introduction we may notice here the character of
the errors occurring in the attempt to trace with a pencil the
outlines of figures that cannot be seen directly but only as
reflected in a mirror. Henri l reports such experiments. The
present results confirm his in all essential respects, (i) When
asked to trace the outlines of a rectangle whose side was
parallel to the plane of the mirror, all seven observers succeeded
easily, though in four a false start in the opposite direction was
noted when they began to trace the lines perpendicular to the
mirror. This is the space relation that the mirror reverses, and

1 'Revue generale sur le sens musculaire.' V. Henri. Annde Psych., V.,
pp. 504-8.

ABNORMAL POSITION AND MOTOR INPULSE. 377

the wrong movement thus conformed to the visual cue. (2)
The tracing of the diagonals in this position was almost, if not
quite, an impossibility for four observers, movements being
made at right angles to the one desired, i. *., in a direction con-
forming to that of the reflected line. For the other three observers
the movement was easy enough except at high speeds, where
an occasional error similar to the foregoing betrayed the ten-
dency usually held in check by the successful adjustment to the
new conditions. The reaction seems to involve the association
of a new kinaesthetic complex with a given visual impression as
soon as the reflected image shows that the movement is being
made in the right direction. These two types of reaction sug-
gest two types of brain function — the one where the organic
paths already formed chiefly determine the direction of the
motor impulse, and the presence of an element common to the
new and the old is sufficient to draft the entire current into the
old channels ; while in the other type all the new elements con-
tribute in determining the direction of the motor impulse. (3)
When the corner of the rectangle was toward the mirror, the diffi-
culty in drawing sides and diagonals respectively was reversed ;
but in kind was like the earlier error. (4) If a more complicated
figure, such as a six-pointed star, be set for outlining, the diffi-
culty increases, though in the case of one or two observers all

TABLE III.
MIRROR FRONT. SUPINATION. FINGERS CROSSED IN PALMS.

Following

Mirror.

k

Errors.

fig

II

t^TJ
j

* &**

Miscell.
Same.

Symmet
Opposite

-65

" = i
£l

Miscell.
Opposite

J

a

o

Baldwin.

48

2.S

22

I

2

II

j-

Emerson.

72*

19

10

4

«7

16

Kleinknecht.

24

2

2

I

I

Miller.

721

45

30

4

8

2

I

22

•3

Rouse.

40

14

21

5

7

I

»9

Rowland.

881

26

7

i

4

10

4

10

Totals.

344

165

99

ii

31

19

5

1 Results of several days combined. Tendency in the separate series the
same as that in total except in Rowland's failures to right and left. The ex-
cess of right-hand failures is due to the results obtained at a single sitting.

378

CHARLES T. BURNETT.

the new adjustments desired throughout this experiment were
made with ease.

TABLE IV.

MIRROR FRONT. PRONATION. FINGERS CROSSED OVER BACKS OF HANDS.

Following Mirror.

Next

No.
Exper.

Errors.

Same
Hand.

Opposite
Hand.

Finger
Same.

Symmet
Opposite.

Finger
Opposite.

Right.

I,eft.

Baldwin.

40

9

6

2

I

4

5

Bmerson.

40

2

I

I

2

Kleinknecht.

40

3

2

I

I

2

Rouse.

40

8

I

5

2

4

4

Rowland.

40

II

I

2

i

7

i

10

Totals.

200

33

10

2

10

4

7

Of the special conditions governing the experiments of
Tables III. and IV. it need only be said that in both cases the
line of the fingers was kept perpendicular to the plane of the
mirror, so far as comfort would allow. The fingers were so
clasped in the work of Table IV. that the left forefinger always
came next to the body ; while for the experiments of Table III.
the left little finger occupied that place, except in a part of the
tests with Emerson and Rowland.

RESULTS.

1. The disturbance in the direction of the motor impulse is
very markedly shown in Table III., though one observer is
almost unaffected.

2. The influence of the visual factor appears in the fact that
more than two thirds of the errors follow the mirror reversal.

3. There is no prominent tendency toward an excess of
failures in one hand over the other. For most of the observers
it is quite absent.

4. Nearly one fifth of the errors consists in a movement of a
finger of the opposite hand. This is not due to any inversion
effected by the mirror, so far as one can see.

5 . The bulk of all the errors not directly accounted for by
the mirror reversal consists in the wrong movement of the sym-
metrically opposite finger and of the next finger on the same side.

6. Under the conditions of Table IV. the illusion has greatly
decreased. It is to be especially noted that the causes operative

ABNORMAL POSITION AND MOTOR IMPULSE. 379

in the former case to produce errors that the mirror reversal
could not directly account for are now much more effective.
The mirror errors are about one third the total, while in Table III.
they are more than two thirds.

7. The errors, barring those of a single observer, show no
tendency to concentration in either hand.

TABLE V.

Two MIRRORS, IN FRONT AND BELOW. SUPINATION. FINGERS CLASPED
OVER BACKS OF HANDS.

Following Mirror.

Next

Next

No.
Exper.

Errors.

Same
Hand.

Opposite
Hand.

Kinder

Same.

Finger
Opposite.

Svm iiir t .
Opposite.

Right.

Left.

Baldwin.

40

8

6

I

I

3

5

Emerson.

40

16

15

I

10

6

Kleinknecht.

40

16

ii

5

8

8

Rouse.

40

21

10

2

i

I

7

5

16

Rowland.

40

12

4

I

6

I

2

10

Totals.

200

73

46

3

M

2

8

TABLE VI.

Two MIRRORS, IN FRONT AND BELOW. PRONATION. FINGERS CLASPED

IN PALMS.

No.
Exper.

Errors.

Following Mirror.

Next
Finger

Next
Ham

^\ rv. !:i ' t
Opposite.

Right.

Left.

Same

Opposite

Hand.

Hand.

ppos

Baldwin.

40

16

10

I

I

I

3

6

zo

Emerson.

40

II

10

I

5

6

Kleinknecht.

40

8

5

I

2

6

a

Rouse.

40

26

6

7

3

10

7

19

Rowland.

40

9

4

5

4

5

Totals.

200

70

35

8

5

I

21

8. What is the cause for the great difference in the amount
of illusion between Tables III. and IV.? A suggestion readily
occurring would attribute it to the greater ease of recognizing
the fingers as individuals when they are clasped over the backs
of the hands. In support of this view may be cited the results
of some experiments performed on Rouse. The conditions
differed from those of Table IV. in this, that the fingers were
covered with paper rolls that largely concealed their individual

38° CHARLES T. BURNETT.

characteristics. In the same number of experiments his errors
were three times as many ; and more than two thirds of these
followed the mirror reversal. And these results were obtained
a week after the former, so that the former results do not appear
to have been due to practice.

But we shall find in Tables V. and VI. evidence to show us
that the positions of pronation or supination can importantly
modify the illusion ; and so to these factors in the present case
we shall have to allow some influence.

For the experiments of Table V. two mirrors were used, at
right angles to each other ; one flat, the other perpendicular to
the median plane of the observer. The fingers were clasped
in the manner indicated by the tables and directed downward,
so that the observer looking into the upright mirror could see a
reflection of the image of the flat mirror. A cloth over the top
of the upright mirror prevented a direct reflection of the hands
in it. The image as seen by the observer reversed the real
position of pronation or supination and also, as in the preceding
experiments, the halves of each hand. In all essentials the con-
ditions of Table V. resemble those of Table IV., the conditions
of Table VI. those of Table III., except in pronation and supi-
nation.

RESULTS.

1. The total amount of errors is greater in the position of
supination than of pronation. In other words, the pronated
hand appears to be under better control. The results of Miller
have to be excluded from Table III. in order to make a justifi-
able comparison. Specifically stated, the errors for Table V.,
(supination) are more than double those in Table IV. (pronation).
A comparison of Tables III. and VI. yields similar results.

2. Again the errors find a second center in Table V., in the
fingers next to the one indicated and on the same side ; while
in Table VI., this second center is rather in the finger symmet-
rically opposite.

3. The tendency of the movement to follow the visual cue
is still evident.

ABNORMAL POSITION AND MOTOR IMPULSE. 381

TABLE VII.
A. MIRROR FRONT. LEFT PALM UP. RIGHT DOWN.

.

Following Mirror.

.

fc

,_

i

I

11
Ii

Opposite
Hand.

Symmet
Opposite.

I

N

sl

X

I!

1

i

Baldwin.
Bos well.

48i

28

16

20

I

5

5

5
3

2

i

i

»3
5

%

Emerson.

40

17

II

2

2

6

ii

Holt.

24

13

12

I

a

ii

Kleinknecht.
Rouse.

24

4

\S

3
22

I

I
10

2

i
17

j

Rowland.

48

9

3

6

3

Miller.

48

26

16

4

4

2

9

'7

Totals.

352

172

109

I

17

29

ia

a

B. As ABOVE, EXCEPT RIGHT PALM UP, LEFT DOWN.

Baldwin.

48

i,5

8

i

2

2

2

ii

4

Bos well.

72'

37

20

i

5

II

ii

7

Emerson.

40

13

ii

2

9

4

Holt.

24

15

13

i

I

7

8

Kleinknecht.

24

5

i

2

2

4

i

Rouse.

48

31

21

4

4

2

>9

12

Rowland.

48

29

II

i

9

3

3

2

M

15

Miller.

48

21

9

6

4

2

13

8

Totals.

352

166

94

3

29

29

9

a

TABLE VIII.

A. MIRROR FRONT. DIRECTION OF FINGERS OPPOSITE. LEFT PALM UP,

RIGHT DOWN.

i

M

W
i

w

Following Mirror.

Next Finger
Same.

Miscell.
Same.

Next Finger
Opposite.

Miscell.
Opposite.

1

5

*1

&n

Opposite
Hand.

Symmet.
Opposite.

Baldwin.

40

25

23

I

I

8

«7

Emerson.

40

18

17

I

8

10

Kleinknecht.

40

7

4

3

7

Rouse.

40

22

13

I

3

4

I

8

M

Rowland.

40

18

12

I

i

2

a

3

»3

Totals.

200

00

60

3

5

6

7

B. As ABOVE, EXCEPT RIGHT PALM UP, LEFT DOWN.

Baldwin.

40

16

13

I

3

I

7

Emerson.

40

19

16

2

I

8

ii

Kleinknecht.

40

16

14

2

S

U

Rouse.

40

26

ii

12

3

II

IS

Rowland.

40

21

IS

2

3

Z

II

10

Totals.

200

98

69

2

18

5

4

1 These results are combined from the work of two day«, agreeing in tendency.

382

CHARLES T. BURNETT.

TABLE IX.

A. MIRROR FRONT. CAPS ON PAI,M-DOWN FINGERS. LEFT PAI.M UP,

RIGHT DOWN.

Following Mirror.

£

u

V

1

i

e"°

S
"8 a

•^ Q>

G+i

S'<«

a u

s a

*. °l

?S
*?

n.tJ
S8
*> ft

l!

i

s

o

fe

w

<» «

as

aa

0.0.
t«o

xtfl

1

SJ;

M ft
.0

^§

M

Baldwin.

40

22

15

2

3

2

8

14

Emerson.

40

8

8

I

7

Kleinknecht.

40

16

12

4

16

Rouse.

40

24

19

3

2

13

ii

Rowland.

40

9

8

I

5

4

Totals.

2OO

79

62

5

IO

2

B. As ABOVE, EXCEPT RIGHT PAI,M UP, LEFT DOWN.

Baldwin.

40*

22

18

3

i

12

IO

Emerson.

401

*3

ii

2

9

4

Kleinknecht.

40

13

7

3

3

4

9

Rouse.

40

24

18

i

4

i

8

16

Rowland.

40

23

20

i

2

9

14

Totals.

200

95

74

4

13

3

i

TABLE X.
A. MIRROR FRONT. CAPS ON Au, FINGERS. LEFT PAI,M UP, RIGHT DOWN.

ii

Following Mirror.

fc

H .

B

p.

H

jj

8fi

ii

II

be .

sa

l!

a.ti
S8

D'oJ

"5

5j

0

W

ftW

aS.

3

S"3

^0

s§

5

^

o

C/3Q

*

!zi

Baldwin.

40

18

16

I

i

7

ii

Emerson.

40

18

17

i

8

IO

Kleinknecht.

40

13

13

ii

2

Rouse.

40

18

15

2

i

3

15

Rowland.

40

14

9

2

3

i

13

Totals.

2OO

81

70

3

5

3

B. As ABOVE, EXCEPT RIGHT PAI,M UP, LEFT DOWN.

Baldwin.

40

20

17

2

i

9

II

Emerson.

40

19

19

15

4

Kleinknecht.

40

28

20

8

10

18

Rouse.

40

27

24

I

2

12

15

Rowland.

40

18

14

i

3

12

6

Totals.

200

112

94

i

6

II

1 Results obtained at two sittings but accordant.

ABNORMAL POSITION AND MOTOR IMPULSE. 383

A further test of the influence of pronation and supination,
as well as of the visual position of the members, was devised in
these new experiments.

The conditions belonging to them are the following : The
line of the crossed fingers is again perpendicular to the mirror
plane ; but the clasped hands are one in the position of pro-
nation, the other in that of supination. The differences among
these four sets are the result of an attempt to eliminate the fac-
tors that might be responsible for the tendency to mass failures
in a given hand. So in Table VIII. care was taken that the
fingers of the supinated hand should not be allowed to curl up,
as they are inclined to do ; but should maintain their direction
as steadily as do the fingers of the other hand. In Table IX.
the attempt was made to check the one-sidedness that might
well grow out of the greater ease in recognizing fingers whose
backs are in view, by covering those fingers with caps made in
the form of paper tubes. And these coverings were extended
to the fingers of both hands in Table X., as equalizing most
fairly the conditions for both. Here also the effort was made
to maintain the opposition in direction of the fingers. Finally,
the same number of experiments was performed with each hand
in a given position.

RESULTS.

1. There appears at first sight to be no simple relation
between the conditions studied and the tendency to mass failures
in one hand. Looking further, however, we find that while

frequently there is no such tendency , yet when it does occur,
the drift is to the supinated hand. Cf . Table XI. One observer
is a definite and consist exception.

2. This must be at least relatively independent of ease in
recognizing the fingers, since it occurs even when the caps are
on both hands.

3. Tables VII. and VIII. show a massing of erroneous move-
ments on the symmetrically opposite finger, as well as on the
next fingers of both the same and the opposite sides. This
tendency to a confusion of hands cannot be accounted for as ft
case of mirror reversal. In Tables IX. and X., there is no such
drift upon the symmetrically opposite finger, but the next fingers
on the same side are chiefly favored.

CHARLES T. BURNETT.

4. In general, the distribution and the significance of the
errors here agree with Tables III. and IV. A test of the ringers
in this position under the condition of direct vision showed prac-
tically complete control.

TABLE XI.
SUMMARY OF TABLES VII.-X.

Table VII.

Table VIII.

Table IX.

Table X.

"S

'a

'a

"a

i

d

V

a

•E

d

i
|

i

6

B

a

3

*j

B

a

3

a

a

'2

(4

1

2

a

2

,_J

a

o

o

o

o

0

O

o

0

Baldwin.

13

15

8

17

1. up

8

14

1. up

7

II

ii

4

r. up

9

7

12

10

9

II

Emerson.

6
9

ii
4

1. "
r. "

8
8

IO

ii

I

9

7
4

1. "

r. "

8
15

IO

4

r. up

Kleinknecht.

i

4

3

i

7

2

14

r. down
1. "

16

4

9

r. down
1. "

ii

IO

2

18

r. down
1. "

Rouse.

17
19

18

12

r.

8
ii

15

1. up

13
8

ii
16

1. down

3

12

15

15

1. up

Rowland.

3
14

15
15

1.

3
ii

15

IO

1. "

5
9

4
14

1. "

I
12

13
6

1. "
r. "

Bos well.

5

26

1.

21

16

r.

Holt.

2

ii

1.

7

Miller.

9

17

1.

13

r.

Against the significant differences in Table XL is indicated
the hand that was supinated. All the observers, except Klein-
knecht agree in concentrating failures, if anywhere, in the hand
whose palm is up. There are but two exceptions in the twenty-
two cases. Kleinknecht is just as constant in the opposite
direction and furnishes a larger number of significant cases than
does any other observer. Why the cause operating in the other
observers should produce intermittent effects does not so far
appear.

The conditions prevailing in these new experiments were
calculated to increase yet more the influence of the abnormal
visual position of the fingers. Two mirrors were set together at
an angle of about 90°. The observer sat over against the apex
of the angle thus formed, and his clasped hands lay in the
region embraced by the angle of the mirrors. The manner of
clasping the hands is shown in the table. The fingers were

ABNORMAL POSITION AND MOTOR IMPULSE. 385

TABLE XII.

A. Two MIRRORS, RIGHT-LEFT, CAPS ON AU. PINCERS, L«rr PAU* Ur,

RIGHT DOWN.

I

i

1

Following Mirror.

i,
r

1

t

Front-back and
Right-left.

Front-
back
Only

Right-left Only.

Symmet.
Opposite.

Miscell.

N. xt
Finger

. !;,;,..«;(..

Mtocell.
Opposite.

Baldwin.
Emerson.
Kleinkuecht.
Rouse.
Rowland.

40
40
40
40
40

32
38
40
36
32

II
2O

9
16
10

I

4
5

3

I

3

16
15

20

14

15

7
I
I

a
3

ao
16
la

ao
ao
ao
ao

Totals.

200

I78

66

10

7

So

9

6

B. As ABOVE, EXCEPT RIGHT PAI.M UP, LEFT DOWN.

Baldwin.

40

29

9

I

ii

7

X

17

xa

Emerson.

40

40

20

3

2

15

ao

20

Kleinknecht.

40

39

18

3

I

16

X

20

19

Rouse.

40

39

22

3

3

9

2

»9

ao

Rowland.

40

34

17

i

ii

4

I

15

«9

Totals.

200

181

86

ii

28

51

3

a

disguised with the usual caps. The observer looked into the
right mirror to see the reflection of the image as originally
given in the left mirror. The second mirror gave a right-left
as well as a front-back reversal of the real position of the fin-
gers. The hands were so placed that but little could be seen

of any primary images.

RESULTS.

1. The very large proportion of errors shows the strength of
the illusion.

2. This amount is so great that there is little chance to mass
errors in either hand. The three cases where there is such a
tendency conform to the chief type in Table XI.

3. The predominance of wrong movements is in the sym-
metrically opposite finger and those fingers in the opposite hand
that lie next to the indicated finger. These are exactly the
places where one would expect the wrong movement to be
made. Where a new adjustment is made for the front-back
reversal, the observer knows where on his hand the finger lie*
that he would move, but he mistakes the hand. Where a new

386

CHARLES T. BURNETT.

adjustment is affected for neither reversal, the observer knows
where in a given half of the hand the movement should be
made, but he confuses both halves and hands.

TABLE XIII.

A. MIRROR FRONT. BACK OF I/EFT HAND AGAINST PALM OF RIGHT.
CAPS ON ALL FINGERS.

Following Mirror.

Next

No.
Exper.

Errors.

Next
Finger
Opposite.

Symmet.
Opposite.

Miscell.
Opposite.

Finger
Same.

Same.

Right.

Left.

Baldwin.

40

8

3

4

I

2

6

Emerson.

40

9

5

I

3

9

Kleinknecht.

40

17

9

I

I

6

10

7

Rouse.

40

12

3

2

4

3

5

7

Rowland.

40

21

15

2

3

I

I

20

Totals.

200

67

35

6

8

17

I

B. As ABOVE, EXCEPT REVERSED RELATION OF HANDS.

Baldwin.

40

9

3

4

2

5

4

Emerson.

40

14

12

2

2

12

Kleinknecht.

40

19

12

4

3

5

14

Rouse.

40

16

5

4

2

5

8

8

Rowland.

40

15

8

6

i

3

12

Totals.

200

73

40

14

2

15

2

TABLE XIV.
A. MIRROR / 20° RIGHT.

Following Mirror.

Next

No.
Exper.

Errors.

Symmet.
Opposite.

Next
Finger
Opposite.

Miscell.
Opposite.

Finger
Same.

Miscell.
Same.

Right.

Left.

Baldwin.

40

21

9

3

8

I

II

10

Emerson.

40

23

9

8

6

15

8

Kleinknecht.

40

22

18

2

2

17

5

Rouse.

40

37

34

I

I

I

19

18

Rowland.

40

23

14

5

4

5

18

Totals.

200

126

84

19

I

21

I

B. MIRROR /_ 20° LEFT.

Baldwin.

40

24

20

3

i

10

14

Emerson.

40

21

18

i

2

12

9

Kleinknecht.

40

15

13

2

IO

5

Rouse.

40

18

17

I

13

5

Rowland.

40

20

7

10

i

2

6

14

Totals.

200

98

75

14

i

8

i

ABNORMAL POSITION AND MOTOR IMPULSE. 387

4. The difference in the amount of errors for the front-back
and the right-left illusion indicates that adjustment to the former
is much more easily effected. Our practical use of mirrors
helps us to overcome the first illusion. The second sort of ex-
perience is relatively novel.

5. The number of errors that cannot be directly accounted
for by the influence of the visual position is nearly negligible.

6. The right-left illusion is stronger when the left palm is
up; the front-back illusion, when the right palm is up.

The conditions of the experiments in these tables were ar-
ranged to show the general principle of visual control, hitherto
copiously illustrated, in yet further ways. For Table XII I., the
hands are placed back against palm, the fingers interlocked,
with the little fingers on the outside, and the line of the fingers
parallel to the plane of the mirror set up in front. In the ex-
periment of the other table the fingers were clasped palm up,
the line parallel to the median plane of the body. The mirror
was placed at an angle of about 20° with the median plane.
This angle was made as small as possible consistent with a con-
venient view on the part of the observer. The arrangements in
both these cases were to reverse in appearance the position of
the hands with reference to each other.

It should be said of the first set of experiments that the posi-
tion chosen was so difficult a one that it was nearly impossible
to keep each set of fingers in lines parallel to each other and to
the mirror. Such displacements tended to produce reversals
among the fingers of a single hand. This probably accounts,
in part at least, for the erroneous movements made with the
correct hand, though these are certainly not in excess of sim-
ilar errors in Table XIV., where such an explanation is not
possible.

The caps were used in Table XIII. because the clasped fin-
gers did not symmetrically correspond, and the resulting differ-
entiation, if seen, might lessen the illusion. This reason did not

hold in Table XIV.

RESULTS.

1. The expected illusion occurs in both cases.

2. The heaping of errors in Table XIV. on the finger sym-

388

CHARLES T. BURNETT.

metrically opposite, and in the other table upon those fingers of
the opposite hand that lie next to the indicated finger, is due in
both cases to the same cause, viz., their occurrence in parts of
the opposite hand spatially corresponding to the indicated finger.
3. The noticeable tendency in Table XIV. to a movement
of the next fingers, either on the same or the opposite side, con-
firms earlier results ; and we have already seen (conclusion 5
under Table I.) that the error cannot be set down wholly to re-
semblance.

TABLE XV.
SUMMARY.

Supination.

Pronation.

Table I.

Table II.

Table III.

Table V.

Table XIII.

Table XIV.

Table IV.

Table VI.

tj

j

£

+1

2

d

£

^J

A

-J

~

-w

a

<£i

*j

j

M

V

M

<U

M

V

bo

4J

be

S

h/i

"<D

M

M

*flj

M

Hf

M

a

14

Hi

a

HI

a

Hi

a

a

Baldwin.

7

8

II

14

^

S

2

6

ii

10

4

«.

6

TO

Emerson.

22

42

14

9

17

16

10

6

9

15

8

2

5

6

Kleinknecht.

8

ii

7

9

I

i

8

8

10

7

11

5

i

2

6

2

Rouse.

6

58

5

«

IQ

5

16

5

7

iq

18

4

4

ij

19

Rowland.

9

31

2

14

16

10

2

10

1

20

5

18

1

10

4

5

Miller.

5

53

22

23

Baldwin.

5

4

10

14

Emerson.

6

13

2

12

12

q

Kleinknecht.

3

14

5

14

10

5

Rouse.

5

7

8

8

13

5

Rowland.

5

13

3

12

6

14

Table XV. presents a summary view of the failures to make
the correct movement as these appear in the right and the left
hands. All the tables are included where both hands agree in
pronation or supination. This particular condition, as it is
found in the unsymmetrical relation of the hands, has already
been discussed in connection with Table XL Here the results
for Tables I., II., XIII. and XIV. fall into one group, as being
concerned with an illusion that tended to throw the movement
over to the opposite hand ; while the remaining results were
obtained where the illusion tended to divert the movement to the
opposite side of the same hand. Tables III., V., XIII. and
XIV. are concerned with positions of supination, and Tables
IV. and VI. with pronation. For Tables I. and II. the position
is a combination of both.

ABNORMAL POSITION AND MOTOR IMPULSE.

389

RESULTS.

1. The tendency to mass errors, where it occurs at all,
shows a drift toward the left. Of twenty-six instances, twenty
are of this type and six of the opposite type. Four of the latter
are confined to one table (Table XIV.).

2. The existence of such a tendency seems to be connected
with the illusion that throws the erroneous movement over to
the opposite hand. Twenty-one instances occur in Tables I.,
II., XIII. and XIV., where the illusion is of this type. The
remaining five are scattered through the other four tables.

TABLE XVI.
COMPILED FROM EXPERIMENTS EMBODIED IN TABLES VII. AND X.

Erroneous Move.

Failure*.

2

3

4

5

*

3

4

s

Baldwin.

Right.
Left.

4
6

I
II

21
24

5

7

12

IS

13
17

2

8

6
6

Emerson.

R.
L.

2

4

21
24

IO

13

IS
IS

17
18

5
i

2
I

Kleinknecht.

R.

L.

10

9

10

15

6

2

i

7

4

10

i
4

la

10

«3

10

Rouse.

R.

L.

IO

18

8
16

II

7

13

10

12

16

5
5

10

19

7
19

Rowland.

R.
L.

IS

21

17
6

4
3

2

3

5

IO

4

12

M

10

.6

The numbers at the heads of the columns indicate the fingers in order,
beginning with the forefinger.

The interesting questions naturally occur whether there is
any tendency (i) to make more erroneous movements with one
finger than with another ; and (2) whether more failures occur
similarly. To answer these questions the eight hundred experi-
ments of Tables VII. and X. were worked over to discover the
distribution of errors and failures among the fingers. Table
XVI. presents the details.

RESULTS.

1 . The several observers do show a preference among the
fingers in erroneous movements and also a massing of failures
but they disagree with each other.

2. The right and left hands show a somewhat remark;
agreement in distribution for any one observer. Out of the

39°

CHARLES T. BURNETT.

forty cases, there appear to be but five where the relative dis-
tribution in the two hands is markedly different.

TABLE XVII.
To vSnow DRIFT OF ERRORS TOWARD THUMB OR LITTLE FINGER.

Table III.

Table V.

Table VII.

Table VIII.

Table IX.

Table X.

Table XII.

I.

V.

I.

V.

I.

V.

I.

V.

I.

V.

I.

V.

I.

V.

Kleinknecht.
Rouse.
Rowland.

I
20
13

I
13

3

15
12
10

i

2

I

4
46

22

4
15
13

19
17
27

4
16
8

17
30
23

9
14
9

27
25
25

14
17
4

40

24
29

12

13
10

Totals.

34

17

37

4

72

32

63

28

70

32

77

35

93

35

Emerson.
Baldwin.

12

2

17

21

5
6

ii

2

IO

13

16
23

I
14

34
25

4
19

17
23

5
6

32
29

16

18

22
23

Totals.

14

38

ii

13

23

39

15

59

23

40

ii

61

34

45

Of the Roman numerals at the heads of the columns I. means thumb and
V. little finger.

Table XVII. presents a new analysis of the results of the
tables summarized therein. All the erroneous movements that
were not made with the symmetrically opposite finger were
classified on the principle of their occurrence either thumbward
or toward the little finger from the indicated finger or its sym-
metrically opposite fellow. The observers were distributed so
evenly between the two classes that they are separated in the
table into two groups. Those of the former tables are included
in this survey that showed the largest amount of errors falling
elsewhere than on the finger symmetrically opposite.

RESULTS.

1. The observers fall into two opposing groups, each show-
ing a very consistent tendency of its special type, and one a
very large one.

2. Considering each observer separately, we find that in one
case only is there a direct contradiction of type, while in but
three cases is neutrality almost or quite complete.

In what direction are we to look for an explanation of the
facts that have come forward in the course of these experiments?
In the first place, there is the fact of the existence of an illusion
connected with an abnormal position of the members. We

A B NORM 'A L POS1 T1ON A ND MO TOR IMPULSE. 39 1

found this to be due in nearly every case to the abnormal visual
factors, since their removal destroyed the illusion. A single
observer in the Japanese illusion seemed to show that abnormal
kinaesthetic factors were involved in producing it. We have to
do here with a special case of neural habit. Visual cues and,
more rarely, kinassthetic cues have become in practice the well-
defined guides of movement, to such an extent, indeed, that
when these become untrustworthy through a change of condi-
tions, it is only by effort, more or less, that the movement nor-
mally connected with them is prevented from occurring.

This principle seems to be illustrated yet further in our re-
sults. It appears that adjustments seeming equally easy to
both hands in normal positions are less easy for the left than
for the right when the positions are abnormal, as in our experi-
ments, though the hands agree in position ; and it appears
further that for the supinated hand the adjustment is also more
difficult. In other words, the neural habits underlying our
practice in the control of our movements are primarily adjusted
to a given space relationship of members ; while plasticity is
greater for the right hand than for the left, and for either hand
pronated than supinated, though in the latter case we must not
forget that for one observer just the reverse was true. The
foregoing difference between the right and the left hands seems
to be in line with the greater ease in control of the right that
we find in many normal movements, though in the one we have
investigated that difference had disappeared, yet only to re-
appear, as reversion to an earlier type, under the condition of
abnormal position. This greater adjustibility in one half of the
brain than in the other half we can view as related to practice.
A similar account is possibly justifiable for the better control of
the pronated hand, though we have still to dispose of our con-
sistent exception. One is tempted to formulate a hypothesis
along the familiar lines of the « sensory ' and * motor ' types,
thereby saving our main principle in this case. For example,
let us make the following suppositions : (i) Less vivid sensa-
tions represent our limbs in consciousness when they are norm-
ally than when abnormally disposed. (2) In the character of
the motor discharge either the nature of the incoming currents

39 2 CHARLES T. BURNETT.

or the situation of the centers may be prepotent. If the nature
of the incoming current prevails, then the less familiar the
situation the better the adjustment, and vice versa ; but if the
situation of the centers prevails, then the more familiar the out-
ward situation, the more correct the response. The former is
the ' sensory ' type, in which must be classified the single
observer whose control is best over the supinated hand ; while
the rest of the observers belong to the latter or * motor ' type.

The tendency of the erroneous movement to be drawn
toward either the thumb or the little finger, according to the
type, may be due to the more habitual employment of the mem-
bers that lie on a given side. The difficulty with this view is
that one would expect all erroneous movements to be drawn
thumb ward, since that side is probably in all but rare cases the
stronger. Individual tendencies to favor or fail in a given
finger have probably a share in the explanation accorded to the
foregoing fact.

We have reason to believe that resemblance plays some part
in the drift of erroneous movements toward the finger sym-
metrically opposite ; but the amount of this error when the fin-
gers are disguised with caps suggests the existence of an addi-
tional factor, perhaps purely physiological. In this direction
points also the prominence of the fingers next to this and to the
indicated finger in wrong movements ; for we found in the dis-
cussion of Table I. that resemblance as a complete account of
this case was out of the question. The precise nature of this
additional factor is obscure to the writer.

There is further obscurity about the connection between the
prevalence of failures in a given hand and the presence of an
illusion that tends to throw the movement over to the opposite
hand. The strength of the evidence for such a connection we
saw in Table XV.

SUMMARY.

1. The influence of abnormal position upon the motor im-
pulse, under the conditions of these experiments, is to change
its direction in certain well-defined ways (cf. all tables).

2. There is a strong tendency to move the finger that really
is where the indicated finger appears to be (cf. all tables).

ABNORMAL POSITION AND MOTOR IMPULSE. 393

3. That visual factors control the movement is shown by the
disappearance of the illusion when touch is added to vision, or
where vision is excluded and the stimulus is auditory. Its
failure to disappear in the latter case for one observer shows
that occasionally abnormal kinaesthetic factors can rise to im-
portance (cf. discussion under Table I.).

4. There is a greater tendency to a wrong direction of the
impulse if the indicated movement is to be made (i) with the
left hand (Table XV.), and (2) with the supinated hand (Table
XL). A single observer out of eight is pretty consistently of
the opposite type in (2).

5. This tendency to mass failures in a given hand is not due
to the greater difficulty of recognizing as individuals the fingers
of that hand. Cf. Tables VII.-X.

6. In the case of the Japanese Illusion, it is not due to a
greater strain on one wrist than on the other. Cf. Table II.

7. The prevalence of failures in the right or the left hand
seems to depend upon the conditions favoring that form of the
illusion that throws the movement over to the other hand (Table
XV.).

8. Individual observers are inclined to favor particular
fingers in erroneous movements and to fail more frequently in
control of one finger than of another ; but among themselves
the observers are very divergent (Table XVI.).

9. There are subordinate tendencies to move : (i) The
fingers next to the indicated finger on the same hand ; (2) the
symmetrically opposite finger, and (3) the fingers next to the
latter (cf. all tables).

10. The tendencies described in (i) and (3) above are not
due to the resemblances between the correct and the wrong
finger. An examination of the results in Table I. showed that
the middle and ring fingers, which resemble each other most of
all, were not mistaken for each other with more significant fre-
quency than the thumb and forefinger.

11. There is a further tendency for wrong movements to be
drawn toward the thumb side of the hand, in the case of three
observers, and toward the little finger in the other two (Table
XVII.).

394 CHARLES T. BURNETT.

12. The existence of the illusion is based on the law of
neural habit. Our habitual dependence upon the visual cue in
controlling our movements leads us astray when that cue no
longer truly represents the actual situation. Failures are more
frequent in the left hand because finer adjustments are less
habitual to it. For that reason they are more frequent in the
position of supination. The condition of the centers is pre-
potent in determining the reaction. In the exceptional type in
which failures occur more frequently in pronation, the reaction
may be viewed as determined chiefly by the incoming currents
Here the less familiar the situation, the more vivid the accom-
panying sensations and the better the adjustment. In the former,
the more familiar the situation, the more correct the response.
For the other facts in this summary, I can give no explanation.

The observers taking part in the work were students in the
Harvard Psychological Laboratory, one being an instructor.
Of the number, two were women and six were men. I acknowl-
edge most heartily their cooperation, as well as that of Professor
Miinsterberg, to whom I owe the suggestion of the problem.1

1The MSS. of this article was received April 14, 1904. — Ed.

DISCUSSION.
MIND AND BODY— THE DYNAMIC VIEW.

It requires a certain temerity to reopen the perennial problem pre-
sented by the apparent dualism of mind and body. It might appear
that the last word worth saying had long since been said. It is, how-
ever, indisputable that the point of view of psychology, and, to some
extent, of philosophy also, is changing. At least its language is
changing and this change is distinctly favorable to a new statement, if
not a solution of this problem. Accordingly, a number of valuable
contributions to the literature of this subject have appeared within the
last few months and the evidence that a monistic construction is de-
sired by nearly all is cumulative. As Professor Moore says: u *Life*
experience is one inclusive activity of which consciousness and habit —
the psychical and the physical — are, to the last analysis, constituent
functions."1

The present tendency on the part of the physical sciences to escape
from the shackles of a material hypothesis offers a 'psychological
moment ' for philosophy to capture the entire forces of both com-
batants.

In advance attention must be called to the fact that there is no
dualism in any one science, neither can there be. Biology has no
body-soul controversy; neither has psychology, as such. It is only
when we attempt at the same time to use both sets of criteria that
dualism arises. The psychological subjective-objective dualism is a
polarizing of what is and always must be a single activity into two
aspects, it does not create a pair of incommensurables. It follows
that this inquiry very naturally assumes the form indicated in the
article entitled ' Mind and Body,' by J. Mark Baldwin.1

"The distinction between phenomena of mind and body, consid-
ered as distinct types of presented phenomenal change, requires the
use of two distinct categories of construction, the genetic and the
agenetic. Physical science it is which interprets the agenetic.
explaining concept of cause is illustrated only and always in transfor-
mations of energy. On the other hand, is the special realm denomi-

1 Univ. Chicago Contrib. to Philos., Vol. III., i.
* Princeton Contributions to Psychology^ III., a.

39 S

396 MIND AND BODY— THE DYNAMIC VIEW.

nated ' subjective.' " (The author adds that ' life processes are really
genetic,' an admission which will greatly influence our attitude toward
the distinction between genetic and agenetic as here defined.)

The problem is formally set in the following inquiry : ' Can we
hold each set of phenomena to its own legitimate construction, and at
the same time, reach a comprehensive conception of the concomitance
of mind and body under which the scientific formulas appropriate to
each may be given full value?' {Ibid., p. 38.) This question be-
comes more pertinent if this author is correct (as we believe him to
be) in saying ' that the present forms of the interaction theory involve
a confusion of categories, due to the failure to maintain a consistent
level of mental development.' {Ibid., p. 39.)

" Philosophy asks: How can we think reality in one thought?
In terms of our present discussion, how can body and mind, being
what we have come to think them to be, live hospitably housed to-
gether in one phenomenal group of facts ? " These questions are such
as to arrest our fullest attention and awaken our keenest interest.
This statement of the problem is most helpful and necessary to further
progress, but the answer given in this place is tentative and explora-
tory. That a single and simple solution is ultimately expected is in-
dicated by the italicised phrase : ' All this means that the world is,
after all, one and that the categories of mental construction, derived
in a process of evolution by actual treatment of the world, cannot
Jinally reflect processes in essential contradiction with each other.'

This is, in fact, the criterion of congruousness, which is the last
appeal and unanswerable argument of monism. The universe is an
organism and contradictory categories could not have developed under
a law of evolution. It is quite disappointing, therefore, especially
after an appeal to an ' all-comprehensive and completely full experi-
ence ' as the content of ' aesthonomic idealism ' to learn that ' psycho-
logical parallelism then is, from the point of view of science, our posi-
tive catch,' even though there is ' hope for a theory of correlation of
these characters which will yield a higher adaptation in the whole
realm of science.' This is the more disappointing in that the one-
sided and unsatisfactory nature of a simple scientific solution has just
been insisted on. But Professor Baldwin modestly refuses to expose
to view the statement of the metaphysical solution designated as y£s-
thonomic Idealism and we are left with one foot on biological foun-
dations and the other on psychological conclusions but with the door
of hope open before us. It was inevitable that others should take
advantage of this fresh statement of the problem to attempt this next

DISCUSSION. 397

step which is to land us with both feet upon some monistic construction.

It is, at any rate, certain that the correlation sought cannot be in
either of the partial realms. Neither biology nor psychology, as such,
can hope to afford a solution which involves both of them. The unity
must be sought in a field large enough to include both.

Nevertheless, it is important for our purpose that we should get
the formulated results of both to be carried up into the higher sphere.
In order to secure this material a brief survey of these contiguous
fields will be necessary. It must be noted in advance that the net re-
sult in each of these cases is of one kind; there are no incommensur-
ables or incompatibles in either sphere. These appear only when the
ultimate data of biology on the one hand, and psychology on the other,
are attempted to be compared (and this attempt is made in terms of
one or the other of these sciences) that incompatibility appears. The
suggestion is obvious that the incompatibility arises from the methods
and not from the content — or, in other words, from the impossibility
of attempting psychological structures with biological tools, and vice
versa.

We may also anticipate our conclusion in so far as to call atten-
tion to the way in which the problem set for us by Professor Baldwin
is disposed of by the so-called ' functional school ' of psychologists
who save us the trouble of further discussion by denying the existence
of any problem. But it is notorious that, a quarrel once on, it is a
work of supererogation to show that there is nothing to quarrel about
It is when the quarrel is over that the proof of its futility is balm to
our wounds.

The most concise and intelligible statement of this functional solu-
tion which the writer now recalls is that given by Professor Bawdcn
in THE PHILOSOPHICAL REVIEW, XIII., 3, May, 1903. *» Mind, as
here viewed, is the totality of the functioning of matter (in so far as
function may be said to imply end or purpose). The psychical is the
meaning of the physical." " Mind is simply a collective idea for all
the psychic functions of an organism — and the psychic functions are
coextensive with the growth of an organism. Mind is not an entity
behind the process of consciousness, it is that process itself. Mind is
just as truly a growth as any other living thing." *• It can be a growth
only if of the nature of a process. Mental life is a continual synthetic
construction. It is simply a name for the orderly continuous function-
ing of an organism under conditions of tension in adaptation" (p. 308).

Professor Bawden uses for the theory thus stated the title »
tional Theory of Parallelism," to which the present writer objects on

398 MIND AND BODY— THE DYNAMIC VIEW.

several grounds, two of which may be mentioned. First, there is an
implied recognition of a material substrate — of a something of which
the mental activity is a ' function.' Second, the theory is not one of
parallelism except as one returns to the artificial dualism of isolated
sciences. Or, to make the criticism general, the view point is that of
psychology while the subject is germain to metaphysics. That this
writer has himself recognized and pointed out the remedy for these
supposed defects may be gathered from his article in Vol. I., No. 3,
of The Journal of Philosophy, Psychology and Scientific Methods.
" Under the name of energy, motion is now regarded as itself the
essence of reality, and the idea of brute, lump matter drops away.
In place of a static we get a dynamic theory of the nature of reality "
(p. 63). Professor Bawden also points out the paradox insisted on
by Professor Baldwin. "The solution of this apparent paradox lies
in seeing that consciousness, taken apart from the organism which is
conscious, is not an entity or thing or even a process ; it is simply a
meaning or significance. * * * After abstracting the psychical by defi-
nition, from the physical, there still cling to our psychological state-
ments of the nature of consciousness traces of our conceptions of ma-
terial objects. * * * Any thinking or speaking is a polarizing into
two aspects in thought of what is an undivided unity for action.
This, of course, is a methodological not an ontological dualism ; hence,
it is paradoxical only for him who forgets its methodological origin."

But these are passages by the way, and we may return to our own
survey. As we have already seen, the difficulties in the historic at-
tempts are due. in a very large part, to the attempt to combine in one
discussion the methods and data of two or more diverse methods of
investigation. Usually the biologist, who essays to discuss the relation
of mind and body, is unable to complete his analysis as a biologist
simply ; he cannot forget that he is also a person, with experiences of
his own which he feels sure are also repeated in the lives of the objec-
tive units he is discussing. He cannot divorce his biological discussion
from its psychological interpretation.

This is, of course, implied in the very nature of the topic, for any
discussion of the relation of mind and body implies the use of the
tools or methods, as well as the data of two sciences, and the question
at issue is just the inquiry whether these data are commensurable and
whether these methods and tools can be employed in the same discus-
sion. As a biologist I cannot consistently inquire as to the relations
between mind and body nor can I, as psychologist, properly discuss
the body, except as an image presented to sense. The question

DISCUSS/ON. 399

reduces to this : Is it possible for the sciences of subjective and objec-
tive phenomena, respectively, to present to philosophy the result* or
interpretations of their research in common terms to that the unifica-
tion (the real business of philosophy) can be completed.

First as to biology. One of its results is the recognition of living
individuals. This is no easy matter nor can the discrimination be
considered complete. Colonies and social groups imply lateral con-
nection which appears in various forms throughout the series and the
existence of which we must suspect in cases which by their nature
prevent us from definitely recognizing it. Individual men are such
units and biology busies itself in recording the complicated synthesis
and coordinations of energy displayed therein. Reciprocal communi-
cation between part and part, mutual reaction of function upon function
demonstrates a ' vital ' relation of unity. No new force is discovered
and, of course, no other than a physical force could be recognized if
many existed. This may be claimed as matter of definition, for any
phenomenon recognized by physical science would be ipse facto
physical.

But there has been talk of a vital force. Such a term could only
be a name for a coordination or a bond. Such a relation is a truth —
a truth of the highest importance, and may well be worthy of a dis-
tinct name — but it is not a fact of the same order as heat, light or
weight.

The recognition of a living unit is a fact of the same kind as the
formation of the judgment of ' substance ' or « object.' * A living ob-
ject ' is such a constant group of coordinated experiences as not only
persists in established relations but proves adaptable to changes in the
environment by reactions thereto without destroying the essential
coherance of these experiences. A living thing is a construct similar
to any other thing. One would not say that the inanimate object WM
created by cohesion, though that may be a name for a part of the ob-
served coherence of attributes. Neither shall we gain by saying that
the animate body is created or maintained by a vital force. Any
given object, e. g., any given man has his own individual formula de-
scriptive of the totality of the reactions (or shall we say the trajectory
or career) . Not that we could express this formula by any means
but such a formula could be conceived as possible.

Now our investigation of the individual man results in our deter-
mining certain partial elements in this all-inclusive formula. We get
a little idea of the energic phases resulting in circulation, respiration,
innervation, etc. Sometimes we are fortunate enough to be able to

400 MIND AND BODY— THE DYNAMIC VIEW.

subsume several minor formulae under one more general or more
inclusive. We never doubt that the possibility exists of a synthesis
which would show all these coordinated in one career. Of course it
is soon discovered that many individuals are wrapped up in any one
subject and that units of a higher order (species, etc.) can be formed
— unities which are formulae for a vastly more complex coordination
yet presenting themselves to us in such wise that we are often able to
approximate nearer to a total formula or statement of the career than
is possible in case of the individual.

Now as biologists we observe the acts of the free individual and
discover fundamentally no difference in kind between the secretion of
bile, the peristalsis of the digestive organs and the most complicated
free motions of prehension, locomotion, etc. There is biologically
no difference between the act of the phagocytes preying on bacteria
in the tissues and the Indian hunter in pursuit of bear and the Wall-
Street broker preying on simple-minded citizens — each of these acts
is beautifully adaptive. So far as we know, the image on the retina
is as real an ' occasion ' for the prehensile phenomenon that follows
as the carbondioxide stimulus on the respiratory center is of the respir-
atory spasms which result.

We can biologically observe that the liver secretes bile ; we can
equally observe that action in the vicinity of the fissure of Rolando is
followed by adaptive motions in the muscles of the limbs and that a
stimulus in Broca's region is followed by reaction of the vocal organs.
But it would be entirely incompetent for the biologist to say that brain
action produces thought. Adaptive reaction is no proof of mentality
as usually understood.

However, we are all born psychologists and, even though we deny
the soft impeachment, we cannot escape this congenital peculiarity.
We feel and sometimes we fancy that we think. We may now-a-days
be a little afraid to admit volition but we still feel quite sure that other
people are responsible for at least part of their actions.

These same physical phenomena, reported to our biological obser-
vation in terms of visual, tactual, auditory, and other reactions, are
reported by the subject in terms of something which he alone can
possess, viz., a subjective reaction, let us say a pain. But let us sup-
pose that the subject of our study is also a trained observer. He
might report to us as biologists the conditions of his own body as
observed by him, that is, as he feels it, sees it, hears its vital move-
ments, etc., and this information, if reliable, would become a part of
our biological formula just as it would if we ourselves or some inde-

DISCUSS/ON. 401

pendent observer had recorded it. In addition, this subject might
report data which we could by no means know anything about, t. g.,
a pain, or peculiar sensation, and he might locate it with reference to the
previous data. This is also valid biological material — this information
is so important that frequently a surgeon will not hesitate in bringing a
life into jeopardy by an operation upon such testimony alone. He,
at least, has no doubt that that particular sense of tenderness and pain
indicates a modification of the normal biological processes* in, let us
say, the appendix vermiformis. But he does not make the mistake
of trying to excise the pain — he is a consistent biologist and to him
the pain is diagnostic simply. Even the so-called empiricists in
medicine do not commit that mistake (except verbally). That is
the pet sin of current psychology alone. To the biologist the reported
pain is as objective a phenomenon as the tympanic reaction to palpa-
tation or the cessation of peristalsis.

The reported ' mental ' reactions of a higher type, with all the
adaptive interrelations, fit into his formula for the life so long as they
are descriptive data only. From his own experience (as psycholo-
gist) he may clothe these reports in a garment of reality, for he has
felt the like, but, as a biologist, they are just other forms of reaction,
like the contraction of a muscle. The experience of joy or a minor
pleasure is connected with circulatory, muscular and nervous activi-
ties, and one is a fact to be catalogued like the others. So it appears
that the whole field of descriptive physiological psychology is a purely
biological science and is to be cultivated with the same tools as any
other department of biology. A great deal of unrealized hope and of
futile effort might, perhaps, have been saved by an adequate realiza-
tion of this classification. Whatsoever a man (biologist) soweth,
that shall he also reap.

But meanwhile we must give the psychic its due. None of these
biological achievements would have been possible but for the subjec-
tive reaction which has not only made it possible to perceive and to
assemble data, but on the accuracy and adequacy of whose forms the
possibility of all classification depends. It is not merely that the
objective world reveals itself to us, but we have created this objective
world in accordance with forms inherent in our subjectivity. It is not
merely that our personal experience has Btamped each elementary
reaction with the certificate of reality without which it would be
valueless, but the very form of the apprehension of the external world
has been the product of the form of our subjectivity.

It appears, therefore, that so long as we persistently abstracted the

402 MIND AND BODY— THE DYNAMIC VIEW.

content of experience and the organization of it from the act of receiv-
ing and organizing the matter seemed simple, but when we ask our-
selves, as sometimes we must, how it happens that we react as we do
to the external world and not equally and indifferently otherwise, the
difficulties of the problem appear.

Psychology may now examine the problem and attempt a solution
from its own point of view. We now have to do with experiences as
avowedly ours, i. «., immediate realities. We have a multitude of
presentations differing in mode. This difference we can never under-
stand, we can only feel it. No Weber's law or periodic formula will
explain why we feel light, taste, pain, etc. These are the data out of
which all that we know is to be formed. There is nothing else.
But a succession of different modes would never give us the contrast-
ing perception of difference vs. identity on which all our psychological
development rests. Here the old psychology demands its own, claim-
ing that such recognition of difference (to put it simply) between pres-
entations of sense in sequence implies a tertium quid — a soul — in
which the comparison must be made. Just as, it is claimed, we can-
not determine whether one figure is identical with another until it is
measured by or in a third thing, so we cannot detect difference until the
two compared elements are brought mutually into relations to another.

To this it may be replied that the ultimate test in geometry is
superposition. In last analysis the demonstrations reduce to applica-
tions of this law of superposition. This analogy, if of any value,
tends rather to the other conclusion that the perception of difference
arises from the reaction between two presentations (or their several
energic grounds) superposed in such wise that the overlapping or non-
agreeing part forms a new percept. Yet here too we imply a con-
tinuum. It is not a conscious continuum. There must be a somewhat
persisting through a greater or less span of time which not only some-
how preserves some counterpart of one impression, but receives a new
one in such wise that the new one is different from what it would have
been but for its predecessor. Things are going on that are not re-
ported in consciousness — things which determine the mode of con-
sciousness at this moment, and which preserve the effects of the energy
involved in some preceding form of consciousness.

We have the curious anomaly then of living in a sphere (psychic)
the grounds of which are indubitably in something else. This some-
thing else has been called the soul. The little rivulet of conscious-
ness on the wave of which rides present experience is all that is open
to examination. We strive to ascertain whether relations (cause and

DISCUSSION. 403

effect, shall we say) can be discovered between elements in this wave
of consciousness and others in other portions of the stream. But how
do we now know anything even of the existence of these other events?
Evidently the ground of their reproduction lies in the structure (/. *.,
activities) of this tertium quid or soul. It appears entirely incor-
rect to speak of relations between successive acts of consciousness—
the relations are between the total acts of which consciousness is one
of the ' meanings ' or modes. There is then no such thing, strictly
speaking, as association of ideas. Is consciousness then but a feeble
reflection of an inaccessible light and are such relations as we discover
between successive flickers of the reflection dependent for their expla-
nation on the reactions of the hidden light? Something like this,
apparently.

This deeper light maybe studied only through these imperfect, in-
termittent, one-sided, reflections — how imperfect only the trained psy-
chologist can fully appreciate. And yet (lest we forget) these flicker-
ing reflections constitute our psychic \\ic,Jide current definitions. To
say that they can by any means directly influence our inner light is
absurd. No more could we kill our enemy by stabbing his shadow or
feed our friend by offerings before his statue. Yet undoubtedly objec-
tive events do affect the psychic manifestations. This process might
be illustrated by the actor who shoots the apple from the head of his un-
seen assistant by aiming with aid of a mirror, or by the Japanese fleet
securing accurate aim at Port Arthur by wireless messages from vessels
at a different angle.

We do not seek to communicate directly with our friend's thought
but we strive to send our message through eye or ear to that some-
what from whence the thought arises. Here is undoubtedly a formal
expression of some sort of parallelism but it can hardly be called a
psycho-physical parallelism. Physically we did not find any reason
for assuming anything psychic at all. Why should we say that this
psychogenetic somewhat is physical?

But perhaps it is not wholly clear that the conscious process doe*
not react on the body. Let us look at it in another way. I feel fear
and because I feel fear I react in a certain way. Not at all. This
statement is contradictory to all that we know of animal activity.
feel fear because certain activities are coordinated in a peculiar manner,
or rather, certain coordinations or equilibriated forms having been
induced, I feel fear. Fear may be but one of the expressions of that
coordination, and there are others, some of which issue in running
away, screaming, etc. Fear is the reflection, shall we say, of a con-

404 MIND AND BODY— THE DYNAMIC VIEW,

flagration having many phases? The fact that I feel fear is not the
' cause ' of my running away.

I communicate the occasion for my fear, ' a burglar,' to my
neighbor. Did I communicate my fear to him ? Not in the least.
Neither did I communicate running away to him. The great wave
dashes upon a rock and passes onward in a hundred eddies, but the
sound that is produced at the same time did not produce the eddies.
(Let us not push this figure too far.)

Psychology may construct a geometry for the relations between
the various experiences and rest content that the expression corresponds
to valid relations existing in the unknown ground of consciousness.
But these elementary experiences are only immediate data — our only
way of knowing this ' ground ' — the rest are only formula? for arrang-
ing them. Judgment is such a formulating activity but is not it deter-
mined by something inhering in the same ground? Is there any ex-
ternal reason why we should formulate the concept ' substance,' for
instance, or does such formulation express but a phase of the consti-
tution of the ' ground ' ? It would appear that the mechanism for
testing truth as much as that in which ' reality ' inheres, is something
back of consciousness or of which consciousness is only one expres-
sion. The form in which my judgments are cast is a fact to be dealt
with as much as the existence of mode itself, and each act of com-
parison or identification has a certain mode or feeling tone which
stamps it as ' ours' rather than another's, and thus adds ' reality' feel-
ing to the fact of thought though it in no way vouches for the ' truth '
of its content.

It becomes apparent then that both biology and psychology become
conscious of limitations and so are aware that there are facts outside
of their boundaries which are nevertheless necessary to the full un-
derstanding of the living individual. Biology assembles observations
of the behavior of the individual. No one observer is able com-
pletely to observe and so part of the information is reported by others
and among the others there may be even the subject of observation
himself.

The facts assembled by his own effort and that of his fellow
laborers and even, to a certain point, by the observed individual are
of the same kind, but the last mentioned is able also to report phe-
nomena inaccessible to the others, yet these unique data fall into con-
gruous relations with the others and supplement or confirm data of the
direct or objective sort. Their validity it is foolish to deny and they
become part of the biologist's material (pain, animal behavior, etc.).

DISCUSSION. 405

The method of securing this information does not trouble the
biologist who remembers that all of his data without exception were
derived by inference from psychic acts or modes of experience.
Psychic and physiological data come to us over the same route. It it
when we seek to interpret these that we find it necessary to resort to
a most complicated contrivance in our own mental activities for out*
ward projection in one case and inward reference in the other. We
are informed by the genetic psychologists that there is a stage prior to
this polarization of experience in the development of the individual.
If this be so we have really encountered nothing so far justifying us
in setting up so fundamental a distinction as that between mind and
body. The most we can say is that we discover in ourselves a differ-
ence between simple psychic acts («. *., immediate experiences) and
the arrangements, relations, and inferences we are forced to make of
them apparently as a result of some orderly or organic mechanism
underlying or including the power to experience. Two things remain
unknown and unknowable from the standpoint of both biology and
psychology, viz., the reason for the modes of simple experience and
for the forms of judgment based on them.

The problem is now appealed by both parties to a higher court.
The trouble has been lack of jurisdiction in each case. It cannot be
said that either department has found justification for separating body
and soul. Each has recognized its limitations and, at first blush has
been inclined to lay all the blame for the ' other ' it discovers or postu-
lates upon the rival science.

The trouble all along has been that the judge is also particeps
criminis and the biologist can no more divest himself of psychologi-
cal infirmities than the psychologist can forget that he is also human
and so biological.

Metaphysics is therefore called upon to reconcile the residual and
unassimilated results of both. Biology asserts that its field is a unit
and everything harmonious so long as it does not consider the source
of its information, but the moment that question is raised, it is forced
to admit that all it has in the way of data is a mass of inferences or
judgments the form or validity of which it can in no wise explain,
and that these judgment are based on immediate experience in various
modes, the differences between which are as unexplained as is the
nature of consciousness itself. Biology therefore relinquishes this
problem to psychology with some asperity to make of as much as
possible. (It may be confessed that it is not very much that is
of it.)

406 MIND AND BODY— THE DYNAMIC VIEW.

Psychology catalogues experiences and names the forms of judg-
ments and diagrams the observed relations, polarizing them into sub-
jective and objective without finding any inherent difference between
them and discovers that there is no direct relation between one experi-
ence and the next. As one feeling does not cause another there must
be some kind of organic nexus behind experience. One thought does
not call up another any more that the secretion of bile to-day produces
a similar act to-morrow, both sets of phenomena are ' explained ' as
related to some organism or continuum. Psychology is prone to sus-
pect biology and to think that a brain is the thing back of thought in
which all psychological manifestations are bound together. When
convinced of the futility of this suggestion it gives up the quest, sim-
ply concluding that the bodily phenomena are ' parallel ' to the mental.
This is nothing but a polite way of confessing defeat, or of keeping
out of the quarrel.

One common element may be recognized in the midst of the ob-
scurity of this discussion, viz., forms of activity. It is not the fact of
energy but its mode that presents to science its multifarious material.

So when asked to arbitrate this dispute metaphysics offers some
such result as is briefly given in the sequel.

But first a word as to the nature of energy. Of energy, in the
nature of the case, nothing can be known except as expressed in the
form of activity. Nothing is to be gained, therefore, by postulating
matter or other entity, different from or behind activity, as a cause or
ground of activity. As stated above, to us energy is known and can
only be known by its form or mode. Behavior is the thing. Energy
is the term representing the fact (all facts known or possible) con-
cerning behavior. Dynamic realism definitively abandons the search
for the unknown ground of behavior and claims that for any human
philosophy the activity itself is the ultimate. It especially declines to
be deceived by any analogy requiring us to know what by nature and
definition must ever remain unknown, viz., matter, a something
itself incapable of action, but the ground of all action.

But energetic form may be viewed in two ways. Otherwise ex-
pressed, all activity in a world of reaction expresses itself in two
classes of modes, one which we may call intrinsic, the other extrinsic.
This is a direct result of a law, which is clear enough from the phys-
ical side but has hardly been sufficiently appreciated in philosophy ;
namely, that activity is meaningless without resistance. Any expres-
sion of energy in a universe is dual in its manifesta .ion. We could
perhaps imagine, or at least, speak about unimpeded energy or ' pure

DISCUSSfON. ; ;

spontaneity,' which would possess only an intrinsic mode. Its mean-
ing would be for itself alone. No such manifestation of energy it
possible. Physically, action and reaction are constantly associated
and equal. A single or isolated force is impossible. In metaphysics,
reality is the reaction of objective and subjective — the 4 affirmation
of attribute.' Morally, the solution of the problem of good and evil,
from this point of view, is that the real good is a doing or striving,
and the evil is the condition of such strife ; this is good in the making
but evil if unvanquished. (See Paulsen's System of Ethics.) Meta-
physically speaking, every being in every phase of its career hat a
double meaning — a meaning for itself and a meaning for the uni-
verse. Illustrations are apt to be misleading or unconvincing, but let
us use a psycho-geometrical analogy. We may suppose that a certain
type of being is represented by an elliptical orbit or trajectory. This
activity will impress itself upon adjacent (in Lotze's sense) energic
modes and the form, extent, and result of this activity will depend on
the nature or mode of the activity in question (here represented by an
elliptical trajectory). The resulting readjustment may be supposed to
extend indefinitely. The universe as a whole is different from what
it would have been but for this particular energetic manifestation.
This is the extrinsic side. Now this being is known to the observer,
not by what it is, but by its extrinsic effects, by the impress it makes
on the universe, or, more particularly, on the immediate environment
of the observer.

But there is another way in which our ellipse must be viewed. A»
a result of its activity upon the world, the world has reacted upon it.
The trajectory is thereafter a different kind of ellipse for having re-
acted with the rest of the universe. Its intrinsic nature has altered.
Its locus formula would have to be rewritten. The inner meaning is
constantly changing. The next time a reaction takes place the effect
will be different from that of the former activity.

Now suppose, as we must, that certain sorts of trajectories or
modes (not to say all of them) express this intrinsic form in terms
analogous to consciousness. This psychic mode is the intrinsic mean-
ing corresponding to the given locus formula.

A still further suggestion could be hazarded : It might be sup-
posed that a certain degree of complexity would be necessary in order
to reach any particular type of conscious expression. Then, il
were complicated systems of equilibriated energy (say human bodies)
which were subject to cyclical or rhythmical variations, it is possible
for the equilibriated unit to drop from a state of extreme complexity.

408 MIND AND BODY— THE DYNAMIC VIEW.

with an intrinsic mode of consciousness, into one not' intrinsically
capable of consciousness in any given form. Later on, in another
phase, the activity could again rise above the ' dead-line ' into that
phase whose intrinsic form is psychic. In the interval below the
' dead-line ' we say the subject sleeps. What the ' genetic modes ' of
the equilibriated unit might be no one can tell till he himself experi-
ences them.1

But how does it happen that we feel our conscious life as a con-
tinuum ? So far as our feeling it is concerned the question does not
need to be asked, for we have no mechanism for recognizing the
hiatus, but there is that behind which bridges the hiatus yet to be
accounted for. It might be said that the intrinsic form varies sympa-
thetically in response to every influence and retains such segments of
past experience as serve to connect all in a present unity of experience.

The ground for our confidence in the general correctness of the
data of mind is to be found, especially from the evolutionary point of
view, in the belief that all these forms of energy have been evolved by
interaction and that the influence of one part is justly and adequately
expressed in every other part. This is what we mean in metaphysics
by describing the universe as an organism. On this basis alone a
monistic interpretation is possible.

The view just expressed cannot be called parallelistic except by
doing violence to the usual form of statement of parallelism and,

1 Perhaps the most apt physical illustration of the idea of psychical equilib-
rium advocated by the present writer may be gained by the study of the gyro-
scope. I am not aware that the mystery of what Foucault called the ' fixity
of the plane of rotation ' and what Tait and Thomson describe as ' gyroscopic
domination ' has ever been adequately explained but we may easily convince
ourselves that composite motions of revolution may be so adjusted as to acquire
a high degree of independence of external influences (such as gravitation) and
to present great resistance to impacts from without. Such a system becomes
gyrocentric.

The formula given for the estimation of the angular velocity, etc., of the
gyroscope is sufficiently complex and we can only faintly imagine the difficul-
ties in the way of constructing a formula covering all phases of gyroscopic
interaction — of wheels within wheels. But when one contemplates the com-
plexities which must characterize the gyrocentric activities cooperating to pro-
duce the type of equilibrium required to produce a thought imagination is quite
at fault.

There can be no doubt that the concentric equilibrium produced is capable
of offering a very high resistance to external impacts in some directions while
being, like the gyroscope, exceedingly sensitive in its responses to influences in
other directions. In other words, the nature of the response is directly a func-
tion of the form of the equilibriated forces.

DISCUSSION. 409

similarly, it can be classed with » identity ' systems only at considerable
hazard of misconception. We prefer to speak of it simply a* dynamic.

In details it is very hard to present this view in such • way as to
give to it the same pleasing objectivity which accompanies the idea of
a material brain grinding out thought as a mill grinds out flour. If
we admit that the complicated equilibriated organism of our being
developed under the law of evolution it need not surprise us that the
reaction corresponding to sensation of redness is an invariable count-
erpart of some particular orderly happening in what we call the
objective world, nor yet need we consider it impossible that, under the
same law, that peculiar conscious reaction which we call a judgment
of « substance ' (always some particular substance) corresponds with
coordination having a constant value as 'representing an objective
thing. So on indefinitely. The most complicated coordinations of
our mental life have a meaning which expresses a real (evolutionary)
correspondence with other things in the universe (objective realities
not otherwise known to us). Even the much discussed concept of
' freedom ' must have its value — it is somehow true. However much
its philosophical interpretation may trouble us, if we are consistent
evolutionists and fully grasp the meaning of the word ' dynamic,' we
must accept its practical implications as genuine.1

C. L. HBRRICK.

SOCORRO, NEW MEXICO.

1 The MSS. of this article was received March 28, 1904. — ED.

Psychological review

P7

v.n

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET

UNIVERSITY OF TORONTO LIBRARY