Skip to main content

Full text of "Lectures on human and animal psychology"

See other formats










Translated from the Second German Edition 







FIRST EDITION, October, 1894. 
SECOND EDITION, June, 1896. 
THIRD EDITION, November 190 
FOURTH EDITION, August, 1907. 


THE present volume is the first of Professor Wundt's 
writings to be made generally accessible to the English- 
speaking public. Its comparatively popular and intro- 
ductory character will, it is hoped, render it especially accept* 
able both to those beginning the study of psychology, to whom 
the technicalities of the author's Grundziige would present very 
considerable difficulties, and to workers in other departments of 
science who may desire some knowledge of the methods and 
results of the new psychological movement. 

The translators have endeavoured throughout to retain the 
oral form of the original Lectures. They have aimed, at the 
same time, to furnish a literal, as distinguished from a verbal, 
rendering of the German text. In view of the confusion which 
still obtains in English psychological terminology, they have 
attempted a precise use of words even at the occasional cost of 
literary effect. No word or phrase, however, has been em- 
ployed which has not already received the sanction of well-known 
psychological writers. 




TlHE Translators have made but few changes, and those of 
minor importance, in the text of this edition. The 
system of nomenclature adopted in the first issue of the 
work has met with general approval, and therefore remains 
unchanged. A few words and phrases have been altered, here 
and there, in the interests of greater clearness and precision ; 
and an index of names and subjects has been added. 

I5th April, 1896. 


WHEN I was asked some years since by the publisher of 
this work to undertake its revision, I felt some hesita- 
tion in complying with his request. The first edition 
of the Lectures appeared thirty years ago; and during that 
time there had not only been a great advance in experimental 
psychology, but my own scientific knowledge and convictions 
had been considerably increased and modified. Thirty years 
ago the science was no more than a programme for the future. 
Except in psychophysics, where Fechner had just broken 
ground, everything remained still to do ; and distrust and sus- 
picion met the investigator at every turn. As for myself, I had 
had but little experience in the difficult work of psychological 
analysis, which the gradual development of the experimental 
methods has done so much to further ; and set about my task 
with more zeal than discretion. So that for years before the 
appearance of the first edition of my Physiological Psychology, 
in which I took up the same problem with more modesty and 
caution, I had learned to look upon the Lectures as wild oats 
of my youthful days, which I would gladly have forgotten. 
But, unfortunately, hypotheses and views represented in them 
would every now and again be confused with or counted among 
my more mature convictions. 

That I have resolved to undertake a second edition despite 
these somewhat discouraging facts, and in preference to the 
more grateful task of writing a new work of similar character^ 
is due in the main to two circumstances. In the first place, I 

vi Author s Preface to Second Edition 

thought that, though the original volumes were defective both in 
general execution and in many points of detail, still a certain 
number of their chapters might stand unchanged, while I 
should perhaps be unable to attain again the freshness and 
force which characterised the first expression of my views. 
Secondly, every statement in the former edition about which I 
had modified or abandoned my original opinion seemed to lay 
upon me an obligation which I would fain discharge to the 
limit of my ability. Nevertheless, I will not omit in this place 
the express declaration that I no longer recognise as mine any 
view formulated in the earlier edition which is not admitted into 
the present. The elimination of everything that more recent 
inquiry had superseded has considerably diminished the size of 
the work. It has also suffered curtailment by the total exclu- 
sion of the discussions of social psychology which occupied a 
large portion of the second volume of the original book. It has 
been a matter of principle with me to restrict the contents of 
the Lectures to the individual psychology of man and the 
animals. As a matter of fact, the amount of material which 
social psychology has at its disposal is now so great, and the 
position of the science with regard to the points discussed has 
undergone so radical a change, that revision of the old chapters 
would necessarily mean rewriting. But within the prescribed 
limits, I have adhered to my former plan of not attempting any 
completeness of exposition, but rather of taking full advantage 
of the lecture form and confining myself to the treatment of 
topics which I thought especially characteristic of the spirit 
and trend of modern psychology. At the same time, it seemed 
permissible to make the work in some sense supplementary to 
my other writings by devoting few words to subjects which I 
have elsewhere discussed in detail, and giving more attention to 
topics which are less prominent, particularly in my Physiologi- 
cal Psychology, Thus I have based the discussion of Weber's 
law entirely upon the method of just noticeable differences, 
although this is the most imperfect of the measurement-methods 
and would hardly now be employed in investigations which 

Author s Preface to Second Edition vii 

made any claim to scientific accuracy. Again, in developing 
the theory of spatial localisation I have retained my previous 
plan of elucidating its much-misunderstood fundamental con- 
ceptions, and of the sensations attaching to movement have 
only dealt with muscle-sensations, although the role of certain 
complexes of pressure-sensations in the surrounding parts is 
really not less important. The reader who desires a deeper 
insight into psychology will, I hope, not omit to refer in such 
cases to my more systematic work, which is more especially 
devoted to the investigation of the physiological correlates of 
psychical processes. 

The first edition of these Lectures was principally based upon 
Fechner's Psychophysik and my own Beitrcige zur Theorie der 
SinneswahrneJunungen, which appeared between 1838 and 1862. 
The lectures dealing with these subjects have undergone the 
least alteration in the second edition. I may perhaps be also 
allowed to state that the treatment of the problem of the 
causality of will in Lecture XXIX. stands for the most part 
precisely as it did in my previous exposition. The following 
lectures of the second edition present portions of the older work 
in revised form: I. (I., II., of the former edition), II. (VII.), Ill 
(VIII.), IV. (IX.), VIII. (XIV.), IX. (XV.), X., XI. (XVI, 
XVII.), XII. (XXI.), XIII. (XXII.), XXIX. (LV., LVI.) ; 
entirely rewritten are V. (XL), VI. (X.), VII. (XIII.), XIV. 
(XXX.), XXV. (XXXI.), XXVI. (LI., LII.), XXVIII. (XLII.) ; 
new are : XV., XVI., XVII., XVIII., XIX., XX., XXL, 
XXII., XXIIL, XXIV., XXVII., XXX. Very little of the 
lectures of the first edition not quoted here has been included in 
the present volume. 


LEIPZIG, April, 1892. 




Philosophical Anticipations of Psychology, 1-5. Spiritualism 
and Materialism, 5-7. Methods and Aids of Psychological 
Investigation, 7-11. 


Analysis of Mental Processes, 12-15. Idea and Sensation, 15-17. 
Intensity and Quality of Sensation, 17, 18. Measurement of 
the Intensity of Sensation, 18-32. 


Estimation of the Intensity of Sensation, 33-39. Mathematical 
Expression of the Law of Sensation-intensities, 39-42. Signi- 
ficance of Negative Sensation-values, 42-47. Unit of Stimulus 
and Unit of Sensation, 47-49. 


A Just Noticeable Sensation, 50-56. Upper and Lower Limit of 
Weber's Law, 56-59. Psychological Interpretation of the 
Law, 59-63. 


Quality of Sensation, 64-66. Tone-sensations ; Beats, 67-73. 
Clang-colour, 73-75. Simultaneous Clangs, 75-77. Noises, 
77, 78. Measurement of Differences of Tone-sensations, 78-81. 
The Tonal Scale, 81-84. Relation to Weber's Law, 84-86. 


Light-sensations ; Sensations of Colour and Brightness, 87, 88. 
Analysis and Mixture of Colours, 88-95. The Three Primary 
Colours, 96-98. Leonardo's Four Principal Colours, 99-103. 
Theory of Light-sensations, 103-107. 


Relation of the Senses of Sight and Hearing, 108. Positive and 
Negative After-images, 108-110. Mechanical and Chemical 
Senses, no, in. Phenomena of Contrast, 111-118. General 
Law of Relativity, 118, 119. 


Reflex Movements, 120-126. Purposiveness of the Reflex, 
126-128. Development of the Reflexes of Touch and Sight, 


Muscle-sensations : their Influence upon Localisation, 134-138. 
Connexion of Sensations of Movement with other Sensations, 
138-142. Influence of Practice upon Reflex Movements, 

x Contents 


143-145. Arrangement in Space a Process of Association, 



' Influence of Ocular Movement on Spatial Vision, 149-156- Geo 
metrical Optical Illusions, 156, I 5 7- Spatial Perceptions of the 
Sense of Touch, 157-160. Accidental and Congenita Blind- 
ness 161-163. Why are not Visual Objects Inverted? 163-167. 
Concluding Remarks upon the Theory of Space-construction, 



LECT T Te E Sepion H of Visual' Ideas ; Influence of Boundary Lines, 
170-172! Ideas of Depth, 172-176 Relations between 
Apparent Magnitude and Distance of Objects, 176- 

. 182-194 

LECT Bm R ocul" V ; Difference of the Two Retinal Images, , 82-186. 
The Stereoscope ; Simplest Stereoscopic Experiments, 186-190. 
Theory of Stereoscopic Vision, 190-194. 


LEC C U ombination E of N Similar Stereoscopic Images, I95-I97. Idea- 
tional Change in Stereoscopic Combination, 197- *99- - c " 

tion and Lustre ; Theory of Lustre, 199-205. - Phenomena of 
Suppression in Binocular Vision, 205-209. 


. Sense-feelings, 2 \ 3-2: 7- Common 

Feeling and the other Total Feelings, 217-221. Relation of 
Feeling to Idea, 221, 222. 


ng to Will'; Impulse and .Desire ^24. 
Development of Will, 224-228. Simple and Complex Vo 
tary Acts, 228-233. Psychological Elements in Voluntary 
Action, 233, 234. 

oonsousn 235-39. -Condition of Ideas * 
Consciousness, 239-243- Perception and Apperception ; Clear- 
ness and Distinctness of Ideas, 244-24?- . Phenomena accom- 
panying Apperception, 247,248. Attention, 249,250. 
consciousness, 250,251. 

Passive and Active 


^2-255. Attention and Will; Fluctuations of Attention, 
256-258. Range of Consciousness ; Formation and Division 
of Rhythmical Series, 258-265. 


e m P or Ideas, 266,267. Personal Diffcrenc^ of 

Astronomers : Eye and Ear Method, 267-271- **** 
Registration, 272-276. Reaction-time, 276-278. Temporal 
Determination of Mental Processes, 278-281. 


of .deas, S 8 Z - 2 83. ' AssocU.ion of ^ 

Simultaneous Association; Complication, 283-286. Assimi- 
lation, 286-290. Theory of Simultaneous Association, 290-294. 


LECTURE TWENTIETH ......... 295-307 

Successive Associations, 295-297. Association by Similarity 
and Contiguity, 297. Cognition and Recognition as Simple 
Forms of Association, 298-303. Theory of Successive Associa- 
tions, 304-306. Indirect Association, 306, 307. 

LECTURE TWENTY- FIRST ........ 308-322 

Concepts and Judgments, 308-311. Distinguishing Marks of 
Intellectual Processes, 312-314. Development of the Intellec- 
tual Functions, 314-316. Mental Derangement, 316-32'.!. 

LECTURE TWENTY-SECOND ........ 323-339 

Dreams, 323-327. Sleep-walking, 327, 328. Hypnotism and 
Suggestion, 328-335. Auto-suggestion and Post-hypnotic In- 
fluence, 335-337. Errors of the ' Hypnotism-psychology,' 

LECTURE TWENTY-THIRD ........ 340-352 

Problems of Animal Psychology ; Deficiencies of the Science, 
340-345. Methodological Rules, 345, 346. Acts of Cognition 
and Recognition among Animals, 347-350. Association among 
the Lower Animals, 350-352. 

LECTURE TWENTY-FOURTH ........ 353-366 

Mentality of the Higher Animals, 353-357. Animal Play, 357, 
358. Alleged Formation of Judgments and Concepts, 358-3618. 
General Significance of Association, 362-364. Man and the 
Animals, 364-366. 

LECTURE TWENTY-FIFTH ........ 367-380 

Connexion of Affective States in Consciousness, 367-369. Sen- 
sory Accompaniments of Compound Feelings, 370,371. 
Emotion, 371-377. Intellectual Feelings, 378-380. 

LECTURE TWENTY-SIXTH ........ 381-394 

Expression of the Emotions, 381-385. Impulsive and Volitional 
Action, 385-388. Instinctive Action, 388, 389. Theories of 
Instinct, 389-394. 


Instinctive Action in Man, 395-397. Acquired Instincts, 397-399. 
Connate Instincts, 399-401. Practice, Imitation, and Inherit- 
ance, 402-406. Relation of Animal to Human Instinct, 

LECTURE TWENTY-EIGHTH ........ 411-422 

Social Instincts ; Temporary Associations and Friendships of 
Animals, 411. Animal Marriage, 412, 413. Animal Societies 
and States, 414-422. 

LECTURE TWENTY-NINTH ........ 423-436 

Voluntary Action, 423, 424. The Causality of Will, 424-428. 
Relation of the Individual to the General Will, 428-432. 
Character as the Ultimate Cause of Will, 432-436. 

LECTURE THIRTIETH ......... 437-454 

Concluding Remarks ; The Question of Immortality, 437-440. 

The Principle of Psychophysical Parallelism, 440-445. Old 

and New Phrenology, 445-448. The Empirical Significance 

of the Principle of Parallelism, 448-451. The Nature of Mind, 

- 451-454- 




PSYCHOLOGY, even in our own day, shows more clearly 
than any other experiential science traces of the con- 
flict of philosophical systems. We may regret this influence 
in the interest of psychological investigation, because it has 
been the chief obstacle in the way of an impartial examination 
of mental life. But in the light of history we see that it was 
inevitable. Natural science has gradually taken shape from a 
natural philosophy which paved the way for it, and the effects 
of which may still be recognised in current scientific theory. 
That these effects are more fundamental and more permanent 
in the case of psychology is intelligible when we consider the 
problem which is set before it Psychology has to investigate 
that which we call internal experience, i.e., our own sensation 
and feeling, our thought and volition, in contradistinction to the 
objects of external experience, which form the subject matter 
of natural science. Man himself, not as he appears from with- 
out, but as he is in his own immediate experience, is the real 
problem of psychology. Whatever else is included in the circle 
of psychological discussion, the mental life of animals, the 
common ideas and actions of mankind which spring from simi- 
larity of mental nature, and the mental achievements of the 
individual or of society, all this has reference to the one 
original problem, however much our understanding of mental 
life be widened and deepened by the consideration of it. But 
the questions with which psychology thus comes into contact 
are at the same time problems for philosophy. And philosophy 


Lectures on Human and Animal Psychology 

had made various attempts to solve them long before psycho- 
logy as an experiential science had come into being. 

The psychology of to-day, then, neither wishes to deny to 
philosophy its right to occupy itself with these matters, nor is 
able to dispute the close connection of philosophical and psy- 
chological problems. But in one respect it has undergone a 
radical change of standpoint. It refuses to regard psychological 
investigation as in any sense dependent upon foregone meta- 
physical conclusions. It would rather reverse the relation of 
psychology to philosophy, just as empirical natural science long 
ago reversed its relation to natural philosophy, in so far, that 
is, as it rejected all philosophic speculations which were not 
based upon experience. Instead of a psychology founded upon 
philosophical presuppositions, we require a philosophy to whose 
speculations value is ascribed only so long as they pay regard 
at every step to the facts of psychological, as well as to those 
of scientific, experience. 

It will, therefore, be a matter of principle for us in these lec- 
tures to stand apart from the strife of philosophic systems. But 
since the thought of to-day is subjected on all sides to the in- 
fluence of a philosophic past which counts its years by thou- 
sands, and since the concepts and general notions under which 
an undifferentiated philosophy arranged the facts of mental life 
have become part of the general educated consciousness, and 
have never ceased to hinder the unprejudiced consideration of 
things as they are, it is our bounden duty to characterise and 
justify the standpoint which we propose to adopt. We will, 
therefore, first of all glance for a moment at the history of 
philosophy before the appearance of psychology. 

In the beginnings of reflective thought, the perception of the 
external world preponderates over the internal experience of 
idea and thought, of feeling and will. The earliest psychology 
is therefore Materialism : the mind is air, or fire, or ether, 
always some form of matter, however attenuated this matter 
may become in the effort to dematerialise it. Plato was the 
first among the Greeks to separate mind from body. Mind he 
regarded as the ruling principle of the body. And this separa- 
tion paved the way for the future one-sided dualism which con- 
sidered sensible existence as the obscuring and debasing of an 

Philosophical Anticipations of Psychology 3 

ideal, purely mental being. Aristotle, who combined with the 
gift of speculation a marvellous keenness of observation, at- 
tempted to harmonise these opposites by regarding mind as the 
principle which vitalises and informs matter. He saw the 
direct operation of mental powers in the forms of animals, in 
the expression of the human figure at rest and in movement, 
even in the processes of growth and nutrition. And he gene- 
ralised all this in his conclusion that mind is the creator of all 
organic form, working upon matter as the sculptor works on 
marble. Life and mentality were for him identical terms ; 
even the vegetable world was on his theory endowed with 
mind. But, apart from this, Aristotle penetrated more deeply 
than any of his predecessors into the facts of mental experience. 
In his work upon the mind, the first in which psychology was 
ever treated as an independent science, he sharply separates 
from one another the fundamental mental activities ; and, so 
far as the knowledge of his time allowed, sets forth their causal 

The Middle Ages were wholly dominated by the Aristotelian 
psychology, and more especially by its basal proposition that 
mind is the principle of life. But with the dawn of the modern 
period begins in psychology, as elsewhere, the return to 
Platonism. Another influence combined with this to displace 
Aristotelianism ; namely, the development of modern natural 
science and the mechanical metaphysics which this develop- 
ment brought with it. The result of these influences was the 
origin of two psychological schools, which have disputed with 
one another down to the present day, Spiritualism and 
Materialism. It is a curious fact that the thought of a single 
man has been of primary importance in the development of 
both these standpoints. Descartes, the mathematician and 
philosopher, had defined mind, in opposition to Aristotle, as 
exclusively thinking substance; and following Plato, he ascribed 
to it an original existence apart from the body, whence it has 
received in permanent possession all those ideas which transcend 
the bounds of sensible experience. This mind, in itself unspatial, 
he connected with the body at one point in the brain, where it 
was affected by processes in the external world, and in its turn 
exercised influence upon the body. 

4 Lectures on Hitman and Animal Psychology 

Later Spiritualism has not extended its views far beyond 
these limits. It is true that Leibniz, whose doctrine of monads 
regarded all existence as an ascending series of mental forces, 
attempted to substitute for the Cartesian mind-substance a more 
general principle, approximating once more to the Aristotelian 
concept of mind. But his successor Christian Wolff returned to 
the Cartesian dualism. Wolff is the originator of the so-called 
theory of mental faculties, which has influenced psychology 
down to the present day. This theory, based upon a superficial 
classification of mental processes, was couched in terms of a 
number of general notions, memory, imagination, sensibility, 
understanding, etc., which it regarded as simple and funda- 
mental forces of mind. It was left for Herbart, one of the 
acutest thinkers of our century, to give a convincing proof of the 
utter emptiness of this ' theory.' Herbart is at the same time 
the last great representative of that modern Spiritualism which 
began with Descartes. For the works of Kant and of the other 
philosophers who came after him, Fichte, Schelling, and Hegel, 
belong to a different sphere. In Herbart we still find the 
concept of a simple mind-substance, which Descartes introduced 
into modern philosophy, but pushed to its extreme logical 
conclusion, and at the same time modified by the first principles 
of Leibniz' monadology. And the consistency of this final 
representative of speculative psychology makes it all the more 
plain that any attempt to derive the facts of mental life from 
the notion of a simple mind and its relation to other existences 
different from or similar to itself must be vain and fruitless. 
Think what lasting service Herbart might have done psychology, 
endowed as he was in exceptional measure with the power of 
analysing subjective perception, had he not expended the best 
part of his ingenuity in the elaboration of that wholly imaginary 
mechanics of ideation, to which his metaphysical presuppositions 
led him. Still, just because he carried the concept of a simple 
mind-substance to its logical conclusion, we may perhaps ascribe 
to his psychology, besides its positive merits, this negative value, 
that it showed as clearly as could be the barrenness of 
Spiritualism. All that is permanent in Herbart's psychological 
works we owe to his capacity of accurate observation of mental 
fact ; all that is untenable and mistaken proceeds from his 

Spiritualism and Materialism 5 

metaphysical concept of mind and the secondary hypotheses 
which it compelled him to set up. So that the achievements 
of this great Spiritualist show most plainly that the path which 
he travelled, apart from all the contradictions into which it led 
him, cannot ever be the right road for psychology. ' This notion 
of a simple mental substance was not reached by analysis of 
mental phenomena, but was superimposed upon them from 
without. To assure the pre-existence and immortality of the 
soul, and (secondarily) to conform in the most direct way with 
the logical principle that the complex presupposes the simple, 
it seemed necessary to posit an indestructible and therefore 
absolutely simple and unalterable mind-atom. It was then the 
business of psychological experience to reconcile itself with this 
idea as best it might. 


When Descartes denied mind to animals, on the ground that 
the essence of mind consists in thought, and man is the only 
thinking being, he cou'd have little imagined that this proposi- 
tion would do as much as the strictly mechanical views which 
he represented in natural philosophy to further the doctrines 
which are the direct opposite of the Spiritualism which he 
taught, the doctrines of modern Materialism. If animals are 
natural automata, and if all the phenomena which general 
belief refers to sensation, feeling, and will are the result of purely 
mechanical conditions, why should not the same explanation 
hold of man ? This was the obvious inference which the 
Materialism of the seventeenth and eighteenth centuries drew 
from Descartes' principles. 

The nai've Materialism with which philosophy began had 
simply ascribed some kind of corporeality to mental existence. 
But this modern Materialism took as its first principles 
physiological hypotheses ; thought, sense, and idea are 
physiological functions of certain organs within the nervous 
system. Observation of the facts of consciousness is of no avail 
until these are derived from chemical and physical processes. 
Thought is simply a result of brain activity. Since this activity 
ceases when circulation is arrested and life departs, thought is 

6 hires on Hitman and Animal Psychology 

nothing more than a function of the substances of which the 
brain is composed. 

More particularly were the scientific investigators and 
physicians of the time inclined, by the character of their pursuits, 
to accept this explanation of mental life in terms of what seemed 
to them intelligible scientific facts. The Materialism of to-day 
has made no great advance in this or in any other direction upon 
the views promulgated in the last century, e.g. by de la Mettrie, 
and developed by Helvetius, Holbach, and others. But this 
equating of mental process and brain function, which makes 
psychology a department of cerebral physiology, and therefore 
a part of a general atomic mechanics, sins against the very 
first rule of scientific logic, that only those connections of facts 
may be regarded as causal which obtain between generically 
similar phenomena. Our feelings, thoughts, and volitions cannot 
be made objects of sensible perception. We can hear the word 
which expresses the thought, we can see the man who has 
thought it, we can dissect the brain in which it arose ; but the 
word, the man, and the brain are not the thought. And the 
blood which circulates in the brain, the chemical changes which 
take place there, are wholly different from the act of thought 

Materialism, it is true, does not assert that these are the 
thought, but that they form it. As the liver secretes bile, as the 
muscle exerts motor force, so do blood and brain, heat and 
electrolysis, produce idea and thought. But surely there is no 
small difference between the two cases. We can prove that bile 
arises in the liver by chemical processes which we are able, in 
part at least, to follow out in detail. We can show, too, that 
movement is produced in muscles by definite processes, which 
are again the immediate result of chemical transformation. But 
cerebral processes give us no shadow of indication as to how our 
mental life comes into being. For the two series of phenomena 
are not comparable. We can conceive how one motion may be 
transformed into another, perhaps also how one sensation or 
feeling is transformed into a second. But no system of cosmic 
mechanics can make plain to us how a motion can pass over into 
a sensation or feeling. 

At the same time modern Materialism pointed out a more 

Methods and Aids of Psychological Investigation 7 

legitimate method of research. There are numerous experiences 
which put beyond all doubt the connection of physiological 
cerebral function on the one hand and of mental activity on the 
other. And to investigate this connection by means of experi- 
ment and observation is assuredly a task worth undertaking. 
But we do not find that Alaterialism, even in this connection, has 
made a single noteworthy contribution to our positive knowledge. 
It has been content to set up baseless hypotheses regarding the 
dependence of mental function upon physical process ; or it has 
been concerned to refer the nature of mental forces to some 
known physical agency. No analogy has been too halting, no 
hypothesis too visionary, for its purpose. It was for some time 
a matter of dispute whether the mental force had more resem- 
blance to light or to electricity. Only on one point was there 
general agreement, that it was not ponderable. 

In our day the conflict between Materialism and Spiritualism, 
which was raging in the middle of the century, has almost worn 
itself out. It has left behind it nothing of value for science ; and 
that will not surprise any one who is acquainted with its details. 
For the clash of opinion was centred once more round the old 
point : in the questions concerning mind, the seat of mind, and 
its connection with body. Materialism had made the very same 
mistake which we have charged to the spiritualistic philosophy. 
Instead of plunging boldly into the phenomena which are pre- 
sented to our observation and investigating the uniformities of 
their relation, it busied itself with metaphysical questions, an 
answer to which, if we may expect it at all, can only be based 
upon an absolutely impartial consideration of experience, which 
refuses to be bound at the outset by any metaphysical 


We find, then, that Materialism and Spiritualism, which set 
out from such different postulates, converge in their final result. 
The most obvious reason of this is their common methodological 
error. The belief that it was possible to establish a science of 
mental experience in terms of speculation, and the thought that 
a chemical and physical investigation of the brain must be the 

8 Lectures on Human ana Animal Psychology 

first step towards a scientific psychology led alike to mistakes 
in method. The doctrine of mind must be primarily regarded 
as an experiential science. Were this otherwise, we should not 
be able so much as to state a psychological problem. The 
standpoint of exclusive speculation is, therefore, as unjustifiable 
in psychology as it is in any science. But more than this, so 
soon as we take our stand upon the ground of experience, we 
have to begin our science, not with the investigation of those 
experiences which refer primarily only to objects more or less 
closely connected with mind, but with the direct examination of 
mind itself, that is, of the phenomena from which its existence 
was long ago inferred, and which formed the original incentive 
to psychological study. The history of the science shows us 
that mind and the principal mental functions were distinguished 
before there was any idea that these functions were connectec 
with the brain. It was not any doubt as to the purpose of this 
organ which led to the abstraction which lies at the foundation 
of the doctrine of mind, but simply observation of mental 
phenomena. Sense, feeling, idea, and will seemed to be related 
activities ; and they appeared, further, to be bound together by 
the unity of self-consciousness. The mental processes began, 
therefore, to be looked upon as the actions of a single being. 
But since these actions were found again to be intimately 
connected with bodily functions, there necessarily arose the 
question of assigning to mind a seat within the body, whether 
in the heart, or the brain, or any other organ. It was reserved 
for later investigation to show that the brain is the sole organ 
which really stands in close connection with the mental life. 

But if it be sensation, feeling, idea, and will which led in 
the first instance to the assumption of mind, the only natural 
method of psychological investigation will be that which begins 
with just these facts. First of all we must understand their 
empirical nature, and then go on to reflect upon them. For it 
is experience and reflection which constitute each and every 
science. Experience comes first ; it gives us our bricks : reflec- 
tion is the mortar, which holds the bricks together. We cannot 
build without both. Reflection apart from experience and ex- 
perience without reflection are alike powerless. It is therefore 
essential for scientific progress that the sphere of experience be 

Methods and Aids of Psychological Investigation 9 

enlarged, and new instruments of reflection from time to time 

But how is it possible to extend our experience of sensations, 
feelings, and thoughts ? Did not mankind feel and think 
thousands of years ago, as it feels and thinks to-day ? It does, 
indeed, seem as though our observation of what goes on in the 
mind could never extend beyond the circle to which our own 
consciousness confines it. But appearances are deceptive. Long 
ago the step was taken which raised the science of psychology 
above the level of this its first beginning, and extended its 
horizon almost indefinitely. History, dealing with the expe- 
rience of all times, has furnished us with a picture in the large 
of the character, the impulses, and the passions of mankind. 
More especially is it the study of language and linguistic de- 
velopment, of mythology and the history of religion and custom, 
which has approached more and more closely, as historical 
knowledge has increased, to the standpoint of psychological 

The belief that our observation is confined to the brief span 
of our individual life, with its scanty experience, was one of the 
greatest obstacles to psychological progress in the days of the 
earlier empiricism. And the opening up of the rich mines of 
experience to which social psychology gives us access, for the 
extension of our own subjective perceptions, is an event of 
importance and of promise for the whole circle of the mental 
sciences. Nor is that all. A second fact, of still greater import 
*br the solution of the simplest and therefore, most general 
psychological problems, is the attempt that has been made to 
discover new methods of observation. One new method has 
been found ; it is that of experiment, which, though it revolu- 
tionized the natural sciences, had not up to quite recent times 
found application in psychology. When the scientific inves- 
tigator is inquiring into the causes of a phenomenon, he does 
not confine himself to the investigation of things as they are 
given in ordinary perception. That would never take him to 
his goal, though he had at his command the experiences of all 
time. Thunderstorms have been recorded, indeed carefully 
described, since the first beginnings of history ; but what a 
storm was could not be explained until the phenomena of 

io Lectures on Human and Animal Psychology 

electricity had become familiar, until electrical machines had 
been constructed and experiments made with them. Then the 
matter was easy. For when once the effects of a storm had 
been observed and compared with the effect of an electric 
spark, the inference was plain that the discharge of the machine 
; vas simply a storm in miniature. What the observation of a 
thousand years had left unexplained was understood in the 
light of a single experiment. Even astronomy, a science which 
we might think must of its very nature be confined to observa- 
tion, is in its more recent development founded in a certain 
sense upon experiment. So long as mere observations were 
taken, the general opinion that the earth was fixed, and that 
the sun and stars moved round it, could not be overthrown. It 
is true that there were many phenomena which made against 
this belief; but simple observation could not furnish means for 
the attainment of a better explanation. Then came Copernicus, 
with the thought : ' Suppose I stand upon the sun ! ' and 
henceforth it was the earth that moved, and not the sun ; the 
contradictions of the old theory disappeared, and the new 
system of the universe had come into being. But it was an 
experiment that had led to this, though an experiment of 
thought. Observation still tells us that the earth is fixed, and 
the sun moving ; and if the opposite view is to become clear, 
we must just repeat the Copernican experiment, and take our 
stand upon the sun. 

It is experiment, then, that has been the source of the 
decided advance in natural science, and brought about such 
revolutions in our scientific views. Let us now apply experi- 
ment to the science of mind. We must remember that in every 
department of investigation the experimental method takes on 
an especial form, according to the nature of the facts investi- 
gated. In psychology we find that only those mental pheno- 
mena which are directly accessible to physical influences can 
be made the subject matter of experiment. W T e cannot experi- 
ment upon mind itself, but only upon its outworks, the organs 
of sense and movement which are functionally related to mental 
processes. So that every psychological experiment is at the 
same time physiological, just as there are physical processes 
corresponding to the mental processes of sensation, idea, and 

Methods and Aids of Psychological Investigation 1 1 

will. This is, of course, no reason for denying to experiment 
the character of a psychological method. It is simply due to 
the general conditions of our mental life, one aspect of which is 
its constant connection with the body. 

The following lectures are intended as an introduction to 
psychology. They do not attempt any exhaustive exposition 
of the methods and results of experimental psychological inves- 
tigation. That would have to assume previous knowledge 
which cannot here be presupposed. Neither shall we include 
in the range of our discussion the facts of social psychology, 
whose contents is extensive enough to demand an independent 
treatise. We shall confine ourselves to the mental life of the 
individual ; and within those limits it will be the human mind 
to which we shall for the most part devote ourselves. At the 
same time it appears desirable, for the right understanding of 
individual mental development, that we should now and again 
institute a brief comparison with the mental life of animals. 





SO soon as ever the dawn of knowledge had broken upon 
us through the portals of the senses, we began to compare 
objects to reflect upon them. The first work of thought was to 
set things in their places, to transform the chaos of sense- 
impressions into an intelligible cosmos. But after everything 
.else has been arranged, there still remains something which has 
as yet no place -ourselves : our feeling, willing, and thinking; 
so that the question arises: how can our own mental life t 
made the subject of investigation like the objects of this external 
world of things about us? And yet-can such a question I 
asked ? Is it not really self-contradictory ? It is as though we 
required that the tone should hear itself, or the ray < 

sensed by itself. 

It is indeed, true that here, as we enter upon the study < 
psychology, a peculiar difficulty presents itself. If we try 1 
observe our mental activities, the observer and the observed 
object are one and the same. But the most important condi 
of a trustworthy observation is always thought to consist in t 
mutual independence of object and observer. Nevertheless, w<; 
should be overhasty if we disputed the possibility of psy- 
cholocncal observation in general because of this unavoidabl 
limitation of the science. Only so much is true : that i 
peculiarities of the object, in this case as in others, imply spe, 
conditions of its observation. These can be stated in two rules 
First so long as we confine ourselves to introspection, wit 
calling in any assistance from outside, mental processes may r 
6e observed directly while they are taking place. We mus 

Analysis of Mental Processes- 13; 

limit ourselves to analysing them, so far as possible, from the 
effects which they leave behind in our memory. Secondly : 
wherever it is possible, we must endeavour so to control our 
mental processes by means of objective stimulation of the 
external organs (particularly of the sense-organs, with the 
physiological functioning of which definite psychoses are 
regularly connected) that the disturbing influence which the 
condition of observation tends to exercise upon them is 
counteracted. This control is given by experiment. Not only 
does experiment, here as elsewhere, enable us to produce a 
phenomenon, and to regulate its conditions, at our pleasure : it 
possesses in psychology an especial importance, in that it alone 
renders self-observation possible during the course of a mental 

Let us now seek, in accordance with the first rule which we 
established, to recall the general impression which any particular 
mental experience has left upon us. This impression will always 
be that of a composite process. Some parts of it, images of 
external objects, we designate Ideas ; others, the pleasurable or 
painful reactions of our own mind upon these ideas, Feelings \ 
others, again, we term Efforts, or Impulses, or Volitions. It is 
certainly true that these elements of mental life never occur 
separately, but always in connection with, always in dependence 
upon, one another. Nevertheless, it seems absolutely necessary, 
at the beginning of a psychological investigation, to follow the 
example of discrimination already set by language, and to 
separate out the most important factors of this complex inner 
life and subject each of them in turn to a special analysis. 

Now, if these elements are all interconnected and inter- 
dependent, it is clear that, other things equal, we might begin 
the analysis which we contemplate with any one of them. 
Nevertheless, external reasons render it hardly possible to choose 
any other method of procedure than that of commencing with 
an investigation of ideas. We conceive of an idea as the image 
of some external object. We can, therefore, transfer to these 
images of external objects the abstraction which we always 
make in the case of the logical notions of the objects ; we 
can consider them just as though the feelings, impulses, and 
volitions, which in fact invariably accompany them, did not 

14 Lectures on Human and Animal Psychology 

exist. On the other hand, in the case of these feelings and 
impulses themselves, it is impossible to carry out an abstraction 
of the kind, because we are not in a position even to describe 
them without constant reference to the ideas with which they 
are associated. Granted that this results merely from the fact 
that all our designations took their origin from distinctions made 
between objects of the external world, and were only applied 
to our inner experiences at a comparatively late date, still 
it remains true that this general trend of the development of 
our knowledge necessarily determines the manner in which 
psychology sets to work to analyse those inner experiences. 

By an idea, then, we shall understand that mental state or 
mental process which we refer to something outside of our- 
selves, whether this attribute of externality be thought of as 
directly applicable in the present, or as applied to an object 
which has been directly given us in the past, or even as applied 
to an object which is only possible, and not actual. Under 
ideas, therefore, we include (i) sense-perceptions, which 
depend upon direct excitation of the organs of sense ; (2) 
memories of such sense-perceptions; and (3) images of fancy, be 
these what they may. The terminology adopted in many 
Psychologies, according to which the images of memory and 
fancy are alone designated ' ideas,' while the direct effects of 
sense-impression are termed exclusively ' perceptions,' we must 
judge to be unjustifiable and misleading. It lends colour to 
the view that there is some essential psychological difference 
between these two kinds of mental process, whereas such a 
difference is nowhere discoverable. Even the reflection upon 
which the distinction is based, the thought that images of 
memory and fancy do not correspond to objects actually pre- 
sented to us, is not universally valid. And, in the same way, 
sense-perceptions may very well be themselves taken for 
illusions of sense. So that the characteristics, by means of 
which two kinds of ideas are distinguished, can never be more 
than secondary, while the distinction itself cannot always be 
satisfactorily carried through. 

An idea, in the general sense in which we are here using the 
word, is always something composite. A visual image is made 
up of spatially distinguishable parts ; a sound is constituted of 

Idea and Sensation I 5 

clangs, while it is also conceived of as coming to us in a certain 
direction, i.e., is associated with spatial ideas. Our first problem 
in analysing ideas, therefore, consists in the determination of 
their simplest constituent elements, and in the investigation of 
the psychological properties of these. We call the psycholo- 
gical elements of ideas Sensations. Thus we speak of the idea 
of a house, of a table, of the sun or moon, but of the sensations 
of blue, yellow, warm, cold, or of a tone of definite pitch. This 
use of the word 'sensation,' we must notice, like the use of ' idea ' 
in the general sense mentioned above, has only become current 
in recent psychology. In the earlier treatises, and still to some 
extent in popular writings and belles lettres, we find the word 
'sensation' employed with the same meaning as 'feeling.' 
Here, and in what follows, we shall consistently adhere to the 
definition just given, according to which sensations are merely 
the simplest and most elemental psychological constituents of 
the idea. 


But the analysis of ideas into sensations does not conclude 
the task which we have set ourselves, the analysis of those 
mental processes which are referable to external objects. For 
in every sensation, again, we distinguish two properties, one 
which we name its strength or intensity, and another which we 
call its quality. Neither can exist in the absence of the other. 
Every sensation, be it of sound, heat, cold, taste, or what not, 
is possessed at once of a certain intensity and a certain 
quality. But, as a general rule, the two' attributes can be 
varied independently of each other. We can sound a musical 
note, e.g., at first quite softly, and then, by gradually increasing 
its strength, pass it through all possible degrees of intensity, 
while its quality remains unaltered. Or we can strike different 
notes one after the other, and so obtain different qualities, 
while we still keep, if we will, one and the same intensity of 
tone throughout. Here quality has changed ; intensity re- 
mains constant. This possibility of varying the two consti- 
tuents of sensation independently of each other depends upon 
the fact that the motions in external nature, by the operation 
of which upon our sense-organs sensation in general was origi- 

1 6 Lectures on Human and Animal Psychology 

nally occasioned, present two aspects, either of which may also 
vary without affection of the other. 

The processes of motion which, by their operation upon our 
senses, give rise to sensations, we commonly denominate 
stimuli, or more particularly sense-stimuli. Accordingly, we 
generally understand by stimulus the external motion-process, 
which, after it has acted upon the sense-organ and been con- 
ducted by sensory nerves to the brain, is accompanied by the 
mental process of sensation. Thus we regard the sound-waves 
of the air or the light-waves set up in surrounding space as 
stimuli, corresponding to our sensations of sound and light. In 
the same way, those motion-processes which are aroused, by 
the agency of such external stimuli, in our sense-organs and in 
the brain, may also be regarded as processes of stimulation or 
as constituents of the entire stimulation-process. For the sake 
of clearness, we will call these last internal stimuli. If we seem 
always to have the external stimuli primarily in mind when we 
are speaking of the relation of ' stimulus' to sensation, this is 
only because they are the more easily accessible to objective 
investigation. But wherever we can show good reason for the 
belief that the peculiar form taken on by a stimulus-process in 
the sense-organs, the sensory nerves, and the sense-centres of 
the brain exercises a determining influence upon a particular 
sensation, we shall, of course, be constrained to take into con- 
sideration the character of the internal stimuli and the transfor- 
mations which occur in the conversion of an external into an 
internal stimulus. 

Now, in whichever of these two senses we employ the notion 
of ' stimulus,' we are able to vary both the intensity and the 
form of any stimulation-process. But the intensity of stimulus 
corresponds to the intensity of sensation, the form of stimulus 
to its quality. (Thus, in the case of sound and light, the inten- 
sity of the sensations corresponds to the extent or amplitude of 
vibration, their quality to its rapidity. The quality of tone we 
call pitch ; the quality of light, colour.) Although, therefore, 
intensity and quality of sensation do not exist independently of 
each other, yet psychological analysis is able to distinguish them 
for its own purposes. In doing this, it is only completing an 
abstraction which was begun when ideas were separated out 

Intensity and Quality of Sensation 17 

from the totality of mental life, and continued a step farther in 
the subdivision of ideas into elementary sensations. 


We begin, then, with an investigation of the intensity of sen- 
sations. And .we leave for the present out of account everything 
which has reference to their qualitative aspect. 

If we compare with each other two different sensations of the 
same modality, we are undoubtedly able to pass judgment re- 
garding their, intensities. Our judgment runs either: The sen- 
sations are of equal intensity, or : They are not of equal in- 
tensity. The midday sun we assert to be brighter than the 
moon, the roar of a cannon louder than the crack of a pistol, 
a hundredweight heavier than a pound. These comparative 
judgments are taken directly from sensation. We really state 
in them merely this : that the sensations which the sunshine, 
the cannon, and the hundredweight arouse in us are more in- 
tensive than the sensations which we have from the moon, a 
pistol-shot, or a pound-weight. There is therefore possible a 
quantitative comparison of sensations. We can say of two sen- 
sations that they are of equal intensity, or that this one is of a 
greater or less intensity than the other. There our measure- 
ment of sensation ordinarily rests. We are not able to say how 
much stronger or how much weaker one is than another. We 
cannot estimate in the least whether the sun is a hundred or a 
thousand times brighter than the moon, the cannon a hundred 
or a thousand times louder than the pistol. Our ordinary 
measurement of sensation tells us only of ' equality/ of a ' more,' 
or of a ' less,' never of a 'so much more ' or ' less.' And 
this natural measurement is, therefore, as good as none at all 
when an exact determination of intensity is required. Although, 
perhaps, we may be able to observe that, as a general rule, in- 
tensity of sensation increases and diminishes with intensity of 
stimulus, yet we have not the remotest idea whether the two 
vary in the same ratio, or whether one increases more slowly or 
more quickly than the other. In a word, we know nothing of 
the law of the dependence of sensation upon stimulus. If we 
are to discover this, we must necessarily begin . by finding a 
more exact measurement for sensation. We must be able to 


1 8 Lectures on Human and Animal Psychology 

say : a stimulus of the intensity I occasions a sensation of the 
intensity I, a stimulus of the intensity 2 a sensation of the 
intensity 2 or 3 or 4, and so on. But, to do this, we must know 
what it means to say that ' this sensation is twice/ or ' three 
times,' or ' four times as great as that.' 

No\v, we have said above that it is possible to strike a note 
first of all very gently, at an intensity at which it can only just 
be heard, and then gradually to increase this intensity, until we 
reach a point at which the note is as loud as it can be made. 
Between these upper and lower limits the tone-sensation has 
passed, not by leaps and bounds, but smoothly and uniformly, 
through all its possible intensities. And the same is true of 
other sense-impressions. From every sensation-quality we can 
construct a one-dimensional series of sensation-intensities, which 
pass over into one another without break or gap. In such a 
series we may, first of all, quantitatively distinguish every mem- 
ber from every other member ; we say that the one of two com- 
pared sensations is the stronger, the other the weaker. But 
more than that. We find no difficulty in stating, after succes- 
sive comparisons, that the difference of intensity in one case 
was greater than it was in another. 

Now, as the result of these very obvious considerations, there 
arise for psychological investigation two separate questions. 
The first is : what is the basis of this natural measurement of 
sensation-intensities, which enables us directly, without knowing 
anything about the external affection of our senses, quantita- 
tively to compare different sensations ? And the second, 
which, as soon as stated, becomes a problem in experimental 
psychology, runs : may not this crude and inaccurate natural 
measurement be transformed into an exact one ; so that, e.g., 
we might be able to state how much stronger or weaker a given 
sensation was than another with which we compared it? We 
will try to answer this second question first. 


At first sight the attempt to measure the intensity of sensa- 
tions may appear overbold. How can we hope to reach any 
result when no definite measure is contained in the sensation 
itself? But if we take a little time to consider how it is that 

Measurement of the Intensity of Sensation 19 

the measurement of magnitude in general is carried out, matters 
will begin to look more hopeful. 

For all measurement there is required a standard. And this 
standard can never be the measured object itself. Thus we 
may measure the time of an occurrence by a clock ; and what 
the clock shows us is a uniform motion. Or we measure longer 
periods of time by days, months, and years ; and these corre- 
spond to uniformly repeated changes in external nature. That 
is, we measure time by space. But to measure space, on the 
other hand, we employ time. The length of the road over 
which we have travelled we estimate by the time that the 
journey has taken. And when we mark the successive divi- 
sions upon a scale, we must do it in a time order. So that the 
original measurement-units of space and time always coincide : 
an hour is just as much an hour of space-experience as an hour 
of time-experience. Space gives us our only means of measur- 
ing time, and time our best means of measuring space. Never- 
theless, there is a noteworthy difference in the way in which 
each of these two measures depends upon the other. For 
space-measurement it is only necessary that time should be al- 
ready existent ; it is not requisite that we should possess an 
exact measure of time. When we are constructing a scale, we 
must mark in one unit after another ; but, that once done, we 
do not need in every particular measurement to compute the 
number of units which the scale embraces. We measure directly 
with the whole scale ; that is, we take all at once, simultane- 
ously, what was constructed gradually. To carry out the most 
exact spatial measurement we need have no more than the 
general notions 'earlier,' 'later,' 'simultaneous.' Then, when 
space has been measured, we come back to time, in order to 
divide it up by the help of our spatial measurements. 

All exact measurement is, therefore, spatial measurement. 
Times, forces, everything that can be considered as magnitude, 
we measure by a spatial standard. Now, when we talk of com- 
paring the intensities of sensations, we imply that sensations are 
magnitudes. And although a direct comparison of sensation- 
intensities does not enable us to do more than pronounce them 
' less ' or ' greater ' or ' equal,' that is in itself no obstacle in 
the way of obtaining an exact measurement. For at first we 

2O Lectures on Human and Animal Psychology 

possessed only the vague ideas of * earlier,' ' later,' and ' simul- 
taneous ' in the case of time ; and yet we are now able to 
measure with very great accuracy temporal differences, the mere 
cognition of which would have far transcended our original 
powers. Indeed, it is just the same with sensation as with time, 
and with all the other magnitudes which, like these two, are 
primarily mental magnitudes. Temporal and spatial magnitudes 
are alike distinguished in the first place only as ' equal,' ' greater,' 
or ' less.' We quickly arrive at an exact determination of the 
latter, since we are able to measure each new space-magnitude 
by magnitudes already known. But the measurement of mental 
magnitudes is apparently attended with greater difficulties. In 
this sphere it was until recently only the movement of. thought, 
time, which had been subjected to an exact measurement, by 
the substitution, for movement, of ideas in us, of movements of 
objects without us, and especially those movements with which 
the impression of uniform regularity was invariably connected. 
An exact means of measuring time cannot, then, be obtained 
from time alone ; we must call in the aid of movement. in space. 
In the same way, we shall never be able to discover a means of 
measuring sensation in sensation itself, but must take into con- 
sideration the relation of its magnitude to other measurable 
magnitudes. And there is no magnitude which presents itself 
more obviously for this purpose than that of the jstimuJus, from 
which the sensation arises. Indeed, the stimulus furnishes us 
not merely with the most obvious, but with our only possible, 
means of measuring sensation. There is no other magnitude 
which stands in any such direct relation to the magnitude of 

The only assistance which sensation itself renders us. in . this 
measurement is that of the ordinary distinction of sensation as 
of ' greater,' 'less,' and ' equal' intensity. Everything else must 
be derived from the measurement of stimulus. If two .sensations 
are of equal intensity, our first thought, is that the external 
stimuli are also of the same intensity in the two ..cases. But 
measurement of them shows not seldom that this surmise is 
wrong ; that stimuli of different intensity may occasion sensa- 
tions of equal intensity. A weak eye finds ordinary daylight so 
intense that it involuntarily closes ; but the normal eye displays 

Measurement of the Intensity of Sensation 2 1 

no such tendency, except when looking directly at the sun. If 
we fall into a swoon, or into deep sleep, we do not sense the 
prick of a needle which, in the waking state, would cause us 
acut6 pain. Indeed, facts of that kind have been observed from 
the beginning of time. This greater or less receptivity of the 
organism, in face of external stimuli, we call sensibility or excita- 
bility. We say that a weak eye is more excitable than a strong 
one ; that we are more sensitive awake than asleep. But we do 
not ordinarily think of measuring this excitability. And yet the 
measure is given at once, if we only ascertain the intensities of 
the stimuli which, on different occasions, give rise to a sensation 
of equal intensity. If the stimuli are of equal intensity in both 
cases, the excitability is the same ; if the stimulus in the first 
case were twice or three times as strong as it is in the second, the 
excitability in the former experiment was half or a third as great 
as it is in the latter. In short, excitability is inversely propor- 
tional to the intensity of the stimuli employed for the production 
of equally intensive'sensations. 

Already, then, we have gained one result, which is not un- 
iniportant for our proposed measurement We have discovered 
a method of eliminating the differences of excitability which 
may be found to exist in different individuals or in the same 
individual at different times. And we are thereby in a condition 
to propose and define a unit of excitability, such as has been 
universally accepted for time, supposing, of course, that its 
proposition shall prove to possess any real significance. 

A further basis of measurement is given with the increase and 
decrease of sensation-intensity. What we all know With regard 
to this is only that the intensity of sensation increases and de- 
creases with the intensity of stimulus. If the ' sound in bur 
ear"' increases, we know that the external sound has become 
louder, always provided that we have no reason to assume a 
change of sensibility in our sense'-organs. Originally this con- 
clusion regarding increase of the external stimulus was merely 
an inference from increased intensity of sensation. Not until 
we have made those physical processes which constitute the 
stimulus the object of separate investigation can we attain to 
the definite conviction that this conclusion was correct. But in 
pursuing such an investigation we dome to make stimulus inde- 

22 Lectures on Human and Animal Psychology 

pendent of sensation, and so are on the road towards the dis- 
covery of a valid measurement of stimulus. 

Now, if our entire knowledge were confined to this fact, that 
sensation increases and decreases with stimulus, we should not 
have gained very much. But there are facts of direct and un- 
assisted observation which tell us something, even if in the most 
creneral terms, of the law which governs the intensive relations 


of stimulus and sensation. 

Every one knows that in the stillness of night we hear things 
which are unperceived in the noise of day. The gentle ticking 
of the clock, the distant bustle of the streets, the creaking of the 
chairs in the room, impress themselves upon our ear. And 
every one knows that amid the confused hubbub of the market- 
place, or the roar of a railway-train, we may lose what our 
neighbour is saying to us, or even fail to hear our own voice. 
The stars which shine so brightly at night are invisible by day ; 
and although we can see the moon in the day-time, she is far 
paler than at night. Every one who has had to do with weights 
knows that if to a gramme in the hand we add a second gramme, 
the difference is clearly noticed ; but if we add it to a kilo- 
gramme, there knowledge of the increase. 

All these experiences are so common that we think them 
self-evident. Really, that is by no means the case. There 
cannot be the least doubt that the clock ticks just as loudly by 
day as by night. In the clamour of the street or amid the 
noise of the railway we speak, if anything, more loudly than is 
usual. Moon and stars do not vary in the intensity of their 
light. And no one will deny that a gramme weighs the same 
whether it is added to one gramme or to a thousand. 

The sound of the clock, the light of the stars, the pressure of 
the gramme weight, all these are sensation-stimuli, and stimuli 
whose intensity always remains the same. What, then, do 
these experiences teach us ? Evidently nothing else than this : 
that one and the same stimulus will be sensed as stronger or 
weaker, or not sensed at all, according to the circumstances 
under which it operates. But what kinds of change in the cir- 
cumstances are there, which can produce this alteration in sen- 
sation ? On considering the matter closely, we discover that 
the change is everywhere of one kind. The tick of the clock is 

Measurement of tke Intensity of Sensation 23 

a weak stimulus for our auditory nerves, which we hear plainly 
when it is given by itself, but not when it is added to a strong 
stimulus of rattling wheels and all the other turmoil. The 
light of the stars is a stimulus for the eye ; but if its stimulation 
is added to the strong stimulus of daylight, we do not notice it, 
although we sense it clearly when it is joined to the weak 
stimulus of twilight. The gramme weight is a stimulus for our 
skin which we sense when it is united to a present stimulus of 
equal strength, but which vanishes when it is combined with a 
stimulus of a thousand times its own intensity. 

We can, therefore, lay it down as a general rule that a stimu- 
lus, in order to be noticed, may be so much the smaller if the 
stimulus already present is weak, but must be so much the 
larger the stronger this pre-existing stimulation is. From this 
alone we can see, in a general way, how our apprehension of a 
stimulus depends upon the intensity of it. It is plain that this 
dependence is not quite so simple as might have been expected 
beforehand. The simplest relation would evidently be that we 
should estimate increase of sensation in direct proportion to 
increase of stimulus-intensity. So that if the sensation I should 
correspond to a stimulus of the intensity i, sensation 2 would 
correspond to intensity 2, and sensation 3 to intensity 3, and 
so on. But if this simplest of all relations prevailed, a stimu- 
lus added to a present strong stimulus would occasion as great 
an increase in sensation as if it were added to a present weak 
stimulus ; the light of the stars would make as large an addition 
to the daylight as to the night. This we know not to be the 
case ; the stars are invisible by day. The increase which they 
occasion in our sensation is not noticeable, whereas this increase 
is very considerable indeed in the twilight. So that this much 
is made out as regards our comparative measurement of sensa- 
tion-intensities, that they do not increase proportionally to the 
increase of stimulus, but more slowly. But when we attempt 
to decide what the relation which obtains actually is, everyday 
experiences do not suffice. We have need of exact and special 

However, before we apply ourselves to the task of making 
these measurements, it is necessary that we should be quite 
clear as to the meaning of the questions which are before us 

24 Lectures on Human and Animal Psychology 

and the importance of the answers which we may expect to find 
to them. If we increase two stimuli of different intensities, 
e.g., a gramme and a kilogramme, by the same unit, e.g., by 
the pressure of a gramme, we come upon the fact that the ad- 
dition to the smaller weight is quite plainly perceived, whereas 
the addition to the larger one is almost or altogether imper- 
ceptible. This fact may be interpreted a priori in two ways, 
(i) It may be that the addition made to the stronger stimulus 
produces absolutely a smaller increase in sensation than the 
same addition made to the weaker. (2) Or it may be that the 
sensation-increase is the same in both cases, but that the 
stronger stimulus requires a greater increase in sensation than 
the weaker, if the differences are to be equally clear in conscious- 
ness. If the first hypothesis is correct, the measurements which 
we are to make will have direct reference to the relation be- 
tween stimulus-increase and the corresponding sensation-in- 
crease ; if the second, then the law of which we are in search 
will refer only to our apprehension and comparative estimation 
of sensations, and not to these themselves. Now, without these 
activities of apprehension and comparison, it is impossible for 
us to formulate any judgment whatsoever concerning sensation- 
intensities, from which it follows that the results of our measure- 
ment of sensation must, in the first instance, be interpreted on 
the alternative hypothesis : that all that we can get at directly 
is the relation between alteration of stimulus and our appre- 
hension of this alteration. It was -with this in mind that I was 
careful to say above, not that a given stimulus-increase produces 
a smaller sensation-increase when added to a strong, than when 
added to a weak, stimulus, but that in our estimation this in- 
crease is smaller. If the absolute sensation-increase is smaller, 
that can only be due to the working of another law, that of 
the parallelism of our estimation of a sensation-increase and its 
actual magnitude. Now, obviously, an answer to the question 
of the validity of such a hypothesis as that can only be looked 
for at the conclusion of a detailed investigation of the relation 
existing between the intensities of stimulus and sensation. This 
is the investigation upon which we are now to embark. You 
will, perhaps, allow me, for the sake of brevity, to speak in what 
follows simply of ' sensation/ when I should more correctly say 

Measurement of 'tke Intensity of Sensation 25 

* apprehension ' or ' estimation of sensation.' But I shall do 
so with the repeated caution that this mode of expression is 
only provisional, and with the assurance that I shall not fail in 
a later lecture to enter fully upon the question whether the im- 
plicit assumption that our apprehension of alterations in sensa- 
tion-intensity runs parallel with the alterations themselves is 
correct, or whether it must ultimately give place to some other. 

This being understood, then, the problem immediately before 
us takes the following shape. We are to determine what in- 
crease of sensation corresponds to equal increases of stimulus, 
or, in other words, to discover what stimulus-increase corre- 
sponds to equal increases in sensation. 

How to execute these measurements is something which our 
everyday experiences suggest. A direct measurement of sensa- 
tion-intensities we saw to be impossible. It is only sensation- 
differences which we can take account of. Experience showed 
us what very unequal sensation-differences might correspond to 
equal differences of stimulus. In most cases we find that the 
same stimulus-difference would be sensed or not sensed accord- 
ing to circumstances ; that, e.g., a gramme is sensed when added 
to another gramme, but not when added to a kilogramme. We 
should think very much less of the statement that a gramme 
added to a gramme produced a considerable difference, added 
to a kilogramme a slight difference, in sensation. And the 
reason is not far to seek. It is difficult to say whether one 
sensation-difference is just smaller or just larger than another ; 
but we have generally no hesitation in calling two sensations 
equal. We are quite sure that the stars are invisible by day ; 
but we might be in doubt as to whether the full moon is brighter 
by night than in the day-time. Our inquiry will, therefore, lead 
to results most quickly, if we start out with some arbitrary 
stimulus-intensity, observe what sensation it arouses, and then 
see how long we can increase the stimulus without having the 
sensation seem to change. If we carry out such observations 
with stimuli of varying magnitude, we shall certainly be obliged 
to vary the stimulus-increase which is just capable of producing 
a difference in sensation. A light, to be just visible in the twi- 
light, need not be 'nearly so bright as starlight; it must be far 
brighter to be just perceptible by day. If now we institute 

26 Lectiires on Human and Animal Psychology 

these observations for all possible stimulus-intensities, and note 
for each intensity the magnitude of the stimulus-increase neces- 
sary to produce a just perceptible increase of sensation, we shall 
get a series of numerical values, in which is definitely and im- 
mediately expressed the law according to which sensation alters 
as stimulus is increased. 

Experiments by this method are especially easy to carry out 
upon the sensations of light, sound, and pressure. We will con- 
sider the last of these first, since they are the most simple. The 
experimenter lays his hand comfortably upon a table. The 
chosen weight is placed upon it. Then a very small weight is 
added to this, and the question put whether the observer, who, 
of course, must not look at his hand during the experiment, 
notices any difference. If the answer is negative, a somewhat 
larger weight is taken, and the same procedure is continued 
until the increment of weight is found, which is just large 
enough to be sensed clearly. When an experiment has been 
concluded with one standard weight, a second and third are 
taken, and so on, until the magnitude of the just necessary in- 
crement of weight has been determined for a sufficient number 
of standards. 

We find a surprisingly simple result. The addition to the 
original weight, which is just enough to produce a noticeable 
difference in sensation, always stands in the same proportion to 
it. Suppose, e.g., that we had found that the necessary addition 
to a gramme was a quarter of a gramme. Then if, instead of 
grammes, we took pennyweights or ounces or pounds, we should 
have to add a quarter of a pennyweight to the pennyweight, a 
quarter of an ounce to the ounce, a quarter of a pound to the 
pound, in order to obtain a just noticeable difference. Or, if 
we confine ourselves to grammes, we must add two and a half 
to ten, twenty-five to a hundred, two hundred and fifty to a 

These figures explain the familiar fact that the difference be- 
tween heavy weights, to be cognisable, must be larger than the 
difference between light ones. But they also give us the exact 
formulation of the law which governs the relation of sensation 
of pressure to force of pressure exerted. You can hold this law 
in mind by remembering a single number, the number ex- 

Measurement of the Intensity of Sensation 27 

pressing the proportion of the added weight to the standard. 
Experimental results show that this proportion is, on the 
average and approximately, that of 1:3. Whatever magnitude 
of pressure may be exerted upon the skin, we sense its increase 
or decrease so soon as the amount added to or subtracted from 
it is one-third of the original. 

Experiments of the same kind, but in greater number and 
with greater accuracy, have been made with lifted weights. 
Here, of course, the conditions are not so simple. When we 
lift a weight, we have not only a pressure-sensation in the hand 
which holds it, but also a sensation in the muscles of the arm 
which raise hand and weight together. This second sensibility 
is much finer than that of pressure proper. Indeed, it has been 
experimentally shown that if lifting is allowed, an addition of 
merely -^^ to the original weight produces a difference in sensa- 
tion. Our sensibility to weight with lifting is, therefore, some 
five times as great as our sensibility to weight which simply 
exerts pressure. And the law of the dependence of sensation- 
upon stimulus may be similarly expressed in terms of the sensa- 
tion of lifting, the fraction ^ being replaced by -j^ or -j^-. This 
proportion holds whether the weight is large or small, whether 
we are speaking of ounces, pounds, or grammes. It tells us- 
that there must be added to a hundred grammes six, to a thou- 
sand grammes sixty, to every standard weight y^j- of its own- 
amount, if a difference in sensation is to be apprehended. 

To determine the objective magnitude of weights, we employ 
the balance ; to measure accurately the objective intensity of 
light, we use a photometer, or light-measurer. This is in principle 
an instrument by means of which the brightness of a given light 
is measured by reference to, and 
expressed in units of another light 
of constant brightness. A very 
simple form of the photometer is 
that schematically outlined in 
Fig. i. A vertical rod, s, is fixed 
in front of a white screen, w. Be- 
hind the rod is placed the light n, 
the intensity of which is regarded 
as the unit of measurement. Beside n is set the light /, whose 

28 Lestur&s on Human and Animal Psychology 

intensity is to be measured. Both lights throw a shadow on the 
white screen. Neither shadow is as dark as it would be if there 
were present simply the one light which produces it ; each is 
illuminated by the other light, and the greater the intensity of 
this other light, the brighter will the shadow appear. Suppose 
that both shadows are equally bright ; that would mean that the 
intensities of the two lights are equal. But suppose, again, that 
the shadow cast by the normal light, the brightness unit, is 
darker than the other ; this means that the intensity of the light 
which is to be measured is less than unity. We can readily 
'determine how much less by moving the normal light somewhat 
farther back, for it is' a law of optics that the intensity of a light 
is inversely proportional to the square of the distance of the 
luminous body. If the light, which was standing at the distance 
of one metre from the white screen, is moved in a straight line 
to a distance of ten metres, the intensity of the light falling upon 
the screen is reduced from 100 to I ; at the distance of ten 
metres it is a hundred times less than it was at the distance of 
one metre. We can easily institute in this way a quantitative 
comparison of a light of unknown intensity with a given normal 
light. We have only to shift the two lights to such distances 
that the darkness of the shadows cast upon the screen appears to 
-be precisely the same ; then we 1 measure the distance of each 
light from the screen, and the inverse ratio of the squares of 
the two distances gives us the relation of the intensities of 
the lights. 

We can turn this same method to good account for the 
measurement of the dependence of light-sensations upon inten- 
sity of light-stimulus. The strong illumination of the shadowless 
part of the screen and the weak illumination of the shadows both 
give rise to light-sensations, which are of course the more 
different the darker the shadows. If we set out with two lights 
of equal intensities, situated at the same distance behind the rod 
(say, two exactly similar stearine candles), the two shadows will 
tie of the same intensity ; i.e., they are equally different from the 
bright background upon which they are thrown. If now we 
move one candle farther and farther away, the shadow cast by it 
becomes weaker, and its difference from the illuminated back- 
ground less; till finally a point is reached at which this difference 

Measurement of the Intensity of Sensation 29* 

vanishes altogether. By measuring first the distance of the 
stationary candle from the screen, and secondly the distance of 
the candle which has been moved so- far back that its shadow 
has just disappeared, we obtain the data necessary for the formu- 
lation of the law of the increase of light-sensation with increasing 
intensity of light-stimulus. So long as only the stationary 
candle was there, the total illumination of the screen was due to 
it. When the other candle is moved up from a distance its light 
adds something to the whole amount of illumination present- 
But this increase is at first unnoticeable ; the point where it be- 
comes noticeable is fixed by the appearance of a second shadow 
of the rod. The place which this shadow comes to occupy is, of 
course, illuminated by the nearer candle, and not by the more- 
distant one ; and as soon as the latter has approached near 
enough to produce a noticeable increase in the total illumination 
the shadow must appear ; it is an index, so to speak, pointing 
to an increase of illumination. And we now possess, in the 
inverse ratio of the squares of the distances of the two candles 
from the screen, the relation of those light-intensities which con- 
dition a just noticeable difference of light-sensation. Suppose, 
e.g., that the first candle was placed at a distance of one metre, 
and the second (which casts a just noticeable shadow) at a, 
distance of ten metres, then the light-intensities stand to one 
another as 100 : 1 ; or, in other words, the intensity of the first 
candle must be increased by one-hundredth, if its increase is to 
effect, an increase of sensation. We have here pursued exactly 
the same method as in our experiments with weights. There we 
added to a heavy weight a lighter one, which just noticeably 
increased the sensation of pressure ; here we add to a strong 
illumination a weaker ,qne, which just noticeably increases the 
light-sensation. It only remains to extend these observations to 
different stimulus-intensities, as was done in the experiments 
with weights. Just as we varied our normal weights, so must we 
vary the luminosity of the standard candle by known amounts. 
That is very easily done. It is only necessary to move the 
candle backwards or forwards, and to calculate its luminosity 
from the distance at which it stands from the illuminated screen. 
Experiments made in this way soon convince us that the dis- 
tances of trje two candles always bea? the same relation to one- 

30 Lectures on Hitman and Animal Psychology 

another. If the second candle had to be placed at a distance of 
ten metres when the first stood at one metre, it must be placed 
at a distance of ten feet when the latter stands at one foot, at 
twenty metres or twenty feet when the distance in the other 
case is two metres or two feet, from which it follows that light- 
intensities which condition a just noticeable difference of sensa- 
tion always preserve the same relation to one another. They 
stand to each other as I : 100, as 2 : 200, etc. But this is the law 
which we discovered in our experiments with weights, and the 
law can just as well here be expressed by the number defining 
the relation of the just noticeable increase of illumination to the 
original illumination. This number is approximately yjLy ; that 
is, every light-stimulus must be increased by y^, if its increase 
is to be sensed. 

It is not hard to institute similar experiments in the sphere 
of sound. The intensity of the sound produced by the fall of a 
body upon some underlying surface increases with the magni- 
tude of its weight and the height of its fall. If we always em- 
ploy the same body, we can vary the intensity of the sound at 
will by varying the height of fall. Intensity and height of fall 
are directly proportional to one another. A fall from twice or 
three times the standard height produces a sound twice or three 
times as loud as the normal sound. A good way of turning 
this principle to account for the investigation of sound-intensi- 
ties which do not differ very greatly from one another is indi- 
cated in the schematic representation of the sound-pendulum 

given in Fig. 2. We take 
two ivory balls, p and g, of 
exactly the same size, and 
suspended by cords of equal 
length. Between the balls is 
placed a block of hard wood, 
c. If one of the two balls is 
let fall from any chosen height 
against the block, the result- 
ing sound is directly propor- 
tional to the height of its fall, 

which can be measured by the angle through which the ball 
was raised from the position of rest. The angle is read off from 

FIG. 2. 

Measurement of the Intensity of Sensation 3 1 

a graduated circular scale placed behind the block. The height 
of fall of the ball p, e.g., is the distance ac ; for the ball q, the 
distance be. That is, the balls strike the block with the velocity 
which they would have possessed had they fallen vertically from 
the points a and b. If ac and be are made equal by moving both 
balls through the same angle, the two sounds are naturally of 
equal intensity ; but if they are different, the sounds are also of 
different loudnesses. As we pass by slow degrees from equality 
to larger and larger differences of height of fall, dropping the 
balls in quick succession, so that the sounds may be accurately- 
compared, we find that for some time there is no noticeable 
difference of sound, despite the difference in height of fall. Not 
until this difference has reached a certain magnitude does the 
difference of sound begin to be noticeable. At that point the 
height of fall is measured for both balls. The difference, of 
course, gives us the amount by which a standard sound-intensity, 
measured by the total height of fall, must be increased if we are 
to obtain a just noticeable difference of sensation. Suppose, e.g., 
that the first ball had fallen through ten centimetres and the 
second through eleven. That would mean that the standard 
sound-intensity must be increased by ^ before a difference 
could be sensed. By making similar measurements over a very 
large number of heights of fall, we shall learn whether this 
relation is constant when the sound-intensity is increased or 
diminished. Just the same is found to hold here as in the case 
of weights and light-intensities : the relation of stimulus-incre- 
ment to stimulus-intensity always remains the same. Every 
sound must be increased by about one-third for the production 
of a clear increase of sensation. 

We have found, then, that all the senses, whose stimuli we 
can subject to exact measurement, obey a uniform law. How- 
ever unequal may be the delicacy of their apprehension of sen- 
sation-differences, this law is valid for all : that the increase of 
stimulus necessary to produce an equally noticeable difference 
of sensation bears a constant ratio to the total stimulus-inten- 
sity. The figures which express this ratio in the several sense 
departments may be shown in tabular form as follows : 

32 Lectures on Hitman* and Animal' Psychology 

Light-sensation .. . . -^n$ 

Muscle .. . - TT. 

Pressure I i 

Sound J 

These figures are far from giving as exact a .measure as might 
be desired. But they are at least adapted to convey a general 
notion of the relative sensibility of the different senses. First 
of all stands the eye. Next comes muscle ; the muscular sen- 
sation affords an accurate measure of the differences of lifted 
weights. Last, and on an approximate equality, stand the ear 
and the skin. 

This important law, which gives in so simple a form the rela- 
tion of our apprehension of sensation to the stimulus which 
occasions it, was discovered by the physiologist Ernst Heinrich 
Weber, and has been called after him Weber's law. He, how- 
ever, examined its validity only in special cases. That the law 
holds for all departments of sense was proved by Gustav Theodor 
Fechner. Psychology owes to him the first comprehensive in- 
vestigation of sense, the foundation of am exact theory of sen- 




THE question might, with some show of reason, be raised 
as to whether the law which we have discovered is valid 
for our quantitative estimation of sensation-magnitudes in general, 
or whether it possesses only a more limited importance. For 
all that we have directly ascertained is this : in what proportion 
the just noticeable sensation-difference stands to the stimulus- 
increment which conditions it. But, as a matter of fact, it will 
be easily seen that the determination of this proportion is simply 
a special case in the determination of a more general relation 
of dependency. 

No one will doubt that it is possible to pass gradually by 
very small sensation-differences to very large ones. Suppose 
that we take a sensation which has increased by a just notice- 
able magnitude, and that we allow this second sensation to 
increase again by a just noticeable difference ; the difference 
between the first and third will be clearer than that between 
the first and the second. And if we proceed in this way, always 
increasing by a just noticeable increment, we shall finally arrive 
at a sensation-intensity which is very much greater indeed than 
that of the sensation from which we set out. And we shall 
have correspondingly reached a very considerable difference of 
stimulus-intensity. Had we passed directly from the weak 
stimulus to the strong, and therefore from the weak to the 
strong sensation, we should never have been able to gain any 
exact information as to the dependency of sensation upon 
stimulus. Taking steps of such length from sensation to sensa- 
tion, we should not have been able to decide whether the 

33 D 

34 Lectures on Human and Animal Psychology 

sensation had increased in the same proportion as the stimulus. 
A result which we could only have attained to with difficulty, 
if we had tried to alternate between large sensation-differences, 
comes out of itself if we gradually increase the stimuli in such 
a way as to pass invariably from one just noticeable sensation- 
difference to another. By how much any one sensation exceeds 
any other is just as difficult to determine from their immediate 
comparison as it would be to say how many more grains of 
wheat there are in one heap than in a second. If we want to 
know that, we must just set to work and count every single 
grain. And, similarly, if we wish to learn how much more 
intense a second sensation is than a first, our best method will 
be to analyse the sensations into those elements which are the 
equivalents of just noticeable differences. 

It is true that in following this method we can never compare 
more than one sensation with another. But if we have once 
established a sensation-unit, we can easily determine by com- 
parison with it the magnitude of any other sensation whatever. 
Let us assume that we have adopted, as the unit of cutaneous 
pressure-sensibility, the sensation occasioned by the pressure of 
i gramme. We have found that the relation in which sensa- 
tion increases with increase of stimulus is expressed in the case 
of pressure-sensations by the fraction ^ ; i.e., the external 
pressure must increase by -|- of its intensity, if it is to produce 
a just noticeable increase of the pressure-sensation. We can, 
therefore, just distinguish i-^- grammes from I gramme ; while 
we can only distinguish 2| from 2, or 3|, i.e., 4 grammes 
from 3, etc. Now if we regard all equally noticeable sensation- 
increments as equal magnitudes, then obviously the magnitude 
of the just noticeable sensation-increase occasioned by the 
pressure of I gramme is equal to the just noticeable increase 
of the sensation occasioned, e.g., by a pressure of 10 grammes. 
So that we may think of any increase of a sensation of whatever 
intensity as being entirely made up of a number more or less 
of just noticeable sensation-increments. We may assume that 
these begin at the point where the external stimulus just suffices 
to excite a sensation. Now, then, we are in a position to give 
quantitative expression to sensation-intensity, however great or 
small this may be. One sensation is twice, three times, or four 

Estimation of tJie Intensity of Sensation 35 

times as intensive as another, when it is made up of twice, three 
times or four times as great a number of equal sensation-in- 
crements. This system of measurement presupposes that we 
follow up sensation in its gradual increase. But that is the 
case, strictly speaking, in all measurement. All the measures 
which we possess consist of a series of measurement-units. 
The unit which we have chosen for sensation is the just notice- 
able increment. If a sensation is made up of four times as 
many units as another, then it is four times as great as that 
other ; just as a scale on which four inches are marked is four 
times as long as one which measures only one inch. If we 
merely estimated the relation of the two scales as regards length, 
our comparison would perhaps not be very accurate. An exact 
judgment is only possible by the application to each of the 
same measurement-unit. And it is precisely similar with sensa- 

The method of measuring sensations of various intensities by 
the addition of just noticeable differences would, however, be 
very cumbrous. We can plainly reach our end very much 
more quickly so soon as we have learned the law according to 
which sensation-increase is correlated with increase of stimulus. 
Having formulated such a law, we could predict that exactly 
so great an increase of stimulus would condition so great an 
increase of sensation. 

As a matter of fact, we possess a law of this kind. Weber's 
law tells us that a stimulus must always increase in a like ratio, 
if the corresponding increase of sensation is to be equally 
noticeable. So that, for practical purposes, any question of 
sensation-measurement may now be put in the form : by how 
many units, or by how many equally noticeable magnitudes 
will, on Weber's law, a given sensation be increased, if we 
increase the stimulus by a definite number of its units ? Or 
conversely : how great must a given stimulus be made, in order 
that the sensation may increase by a definite number of sensa- 
tion units ? Let us take pressure-sensations once more, for 
purposes of illustration. You will remember that the sensation 
occasioned by I gramme must be intensified by gramme for 
it to increase by I unit. Suppose now that we wished to learn 
how much the pressure must be intensified for the sensation to 

36 Lectures on Human and Animal Psychology 

FIG. 3. 

increase by 6 such units. We imagine 
the sensation-units arranged upon a scale. 
At the zero-point of this scale, which we 
will place for the moment arbitrarily at a 
stimulus of I gramme, we draw a perpendi- 
cular of any length to representthe gramme. 
In order now to represent the magnitude of 
pressure for a sensation increased by I unit, 
we must lengthen the perpendicular at I 
by of the perpendicular at o. 

Similarly at 2, we must lengthen the perpendicular I by ; 
at 3, the perpendicular 2 by ^, etc. Since the perpendiculars 
constantly increase, these incremental parts will also of course 
become larger ; we have to draw upon our scale lines of con- 
tinually increasing length. And it is plain that the magnitude 
of each of these lines stands to that of the perpendicular drawn 
at zero in the same relation in which the weight, occasioning 
the sensation-increase marked upon the scale, stands to the 
initial weight of one gramme. The question being, what weight 
has to be applied to produce a sensation-difference equal to 6 
sensation-units, we have only now to measure how much longer 
the perpendicular at 6 is than the perpendicular at o. 

If we connect the upper ends of the perpendiculars drawn 
upon our sensation-scale to represent stimulus-magnitudes, we 
obtain a curved line ascending more steeply as we approach the 
higher values of the scale. This curve obviously shows the de- 
pendence of our measurement of sensation-intensities upon the 
corresponding stimuli, not only for the points I, 2, 3, etc., but 
also for all points situated between these, e.g., for \\, \\. If we 
wish to discover what intensity of stimulus corresponds to some 
particular point lying between two unit values, we need only 
connect the point in question by a perpendicular with the curve 
representing the alteration of stimulus. The magnitude of the 
required stimulus is represented by the length of this perpen- 
dicular. The sensation-difference which corresponds to a point 
on the scale lying in this way between two unit values is, of 
course, not perceptible by us ; but it would be quite wrong to 
infer from this that it has no existence whatsoever. For we can 
only reach perceptible differences by heaping up, as it were, a 

Estimation of the Intensity of Sensation 37 

great number of imperceptible differences. It is mere chance 
that the just noticeable sensation-differences in our illustration 
fall exactly at the points I, 2, 3. If we were to take as our 
initial weight \ or gramme instead of I gramme, the whole 
scale would be shifted to the left, and the points where the 
numerals now stand would then fall between two numerals of 
this second scale. But the law of the variation of sensation- 
with stimulus-intensity would remain precisely as before. Our 
measurement on any scale is discrete, but the scale itself is con- 
tinuous. We cannot, you see, proceed from one weight to 
another so as to pass through all possible intermediate weights ; 
but we interpolate between 2 grammes -j^-, y^j., 10 1 00 , or perhaps 
even 10 ^ 00 of a gramme, if we wish to be exceedingly accurate 
in weighing. But no one would maintain that a weight of less 
than 1( jooo of a gramme is no weight at all. And just as there 
are differences of weight, which no balances can detect, so there 
are differences of sensation, which we are unable to cognise. 

Now there can be no doubt that the scale which we have been 
using to measure sensations, is not one particularly suited to its 
purpose. We started out from the simplest possible stimulus- 
magnitude, from the pressure of I gramme, our unit of weight. 
We made the zero-point of our scale correspond to this point, 
and proceeded to fill in our sensation-units to the right of it. 
But when we have done this, we have not put ourselves in a 
position to determine anything more than by how much we must 
increase the weight of a gramme in order to obtain a definite 
increase of sensation-units ; or how many sensation-units have 
been added to the pressure sensation of one gramme, when we 
are being stimulated by a weight of definitely greater magni- 
tude. We do not know in the least how great the sensation is 
which is occasioned by I gramme ; i.e., how many sensation- 
units are to be reckoned to the left of the zero point on one 
scale. The way to determine this is obviously to set out, not 
from a definite stimulus-unit, but from the unit of sensation ; 
and to measure onwards in terms of this, from the point where 
sensation begins. If, then, we wish our scale to be a natural one, 
we shall take the point at which sensation begins for our zero- 
point. But this is not at the same time the zero-point of 
stimulus. Some stimuli are so weak that they are not sensed at 


8 Lectures on Human and Animal Psychology 

all. In order to occasion a sensation, the stimulus must have 
attained a definite magnitude, which in each case is determined 
by the character of the sense organ. The case here is similar 
to that of sensation-differences. These are only perceived if 
the stimulus-differences are of a certain intensity. In the same 
way sensations in general are only perceived when the stimulus 
has attained a certain magnitude. It might, perhaps, be sup- 
posed that the two cases are not only similar, but identical 
that the intensity of the stimulus necessary to produce a sensa- 
tion at all is equal to the intensity of stimulus-difference which 
gives rise to a just noticeable difference of sensation. But it 
may be easily seen that this is impossible. The intensity of a 
stimulus-difference is always directly dependent on the total 
stimulus-intensity, and decreases with decrease of the latter. So 
that if the stimulus becomes infinitely small, we should be forced 
to assume that the stimulus-difference must also become infinite- 
ly small. That however is contradicted by experience, which 
shows us that every stimulus must have attained a definite 
measurable magnitude, if it is to produce a sensation. 

If, therefore, we follow our former method, and erect per- 
pendiculars to express the stimuli which correspond to the 
series of sensations, we must draw at the zero-point a line whose 
length represents the magnitude of the stimulus which occasions 
a just noticeable sensation. If we keep to our sensations of 
pressure, and find that J^- of a gramme is the magnitude of 
weight sufficient to excite a just noticeable pressure-sensation, 
we shall represent this weight by a perpendicular at the zero 
point. At I, which is removed from o by a just noticeable differ- 
ence, the vertical representing the stimulus will, in accordance 
with the dependency of sensation upon stimulus, be longer ; 
i.e., the stimulus whose original magnitude was -^ or y^- will 
here be -^4^-, etc. In short, we obtain the same relative increase 
of stimulus and sensation that we had upon our former scale 
(Fig. 3), the only difference being, that the new vertical at o 
now stands for -^ of a gramme, and not for I gramme. 

To answer all the questions that come up in any sense-depart- 
ment, then, two measurements are in general sufficient ; first, the 
measurement of the constant relation in which sensation-intensity 
varies with variations in the intensity of the stimulus ; and 

The Law of Sensation-intensities 39 

secondly, the measurement of the just noticeable sensation. 
The first measurement enables us to divide up the sensation- 
scale ; by calling in the aid of stimuli we can mark it off into 
equal parts. The second measurement gives us its zero-point, 
and thus renders the scale ready for practical use. If we have 
found in the sphere of pressure-sensations that the constant ratio 
is |-, and that the just noticeable sensation is produced by -^ 
gramme, we can dispense with all further measurement, and 
solve any problem presented to us. Suppose that we wish to 
know the intensity of a sensation excited by the pressure of I 
gramme. We take our scale, and begin with the zero-point. 
The pressure at O is J^- gramme ; the pressure at I is -^ greater ; 
the pressure at 2 is greater than it was at I, etc. We proceed 
in this way till we come to a pressure of I gramme, and then 
count up how many units of our sensation scale have been em- 
ployed up to that point. We shall find that we have used nearly 
14 units ; so that if we press upon the skin first with jfo, and 
then with I gramme, we have passed over 14 just noticeable 
differences. And the nearer we come to I gramme, the greater 
are the pressure-differences to which the just noticeable differ- 
ences correspond. The first unit corresponds to -^ of the original 
stimulus, or -j-^j- gramme. If the sensation increased directly as 
the stimulus, our 14 units would correspond to an increase of 4-| 
or not quite - gramme ; while, as a matter of fact, they require 
an increase of pressure of 4-^, or almost a whole gramme. 


This method of determining the intensity of sensation by pro- 
ceeding gradually from weak to strong stimuli through just 
noticeable differences would, however, be exceedingly tedious 
in practice. Direct observation would possess over it the advan- 
tage of greater brevity. The question, therefore, suggests itself, 
whether we cannot discover some shorter method, which would 
permit us to pass at one step from -fa to i gramme, instead of 
using, as we did above, no less than 14 intermediate stsges. 
This question may be answered in the affirmative, as a some- 
what closer consideration of the dependency existing between 
sensation and stimulus will convince us. 

Sensations and stimuli are interdependent magnitudes. Both 

4O Lectures on Human and Animal Psychology 

are capable of numerical expression. The numerical values 
which stand for sensations increase with the increase of the 
numerical values of stimulus. The simplest relation in such a 
case would plainly be this : that corresponding to the stimuli 
expressible by the numbers I, 2, 3, etc., there existed sensations 
which were also expressible by those numbers. We should then 
say that sensation-intensity is directly proportional to intensity 
of stimulus. This simple relation, however, does not hold ; 
stimuli increase far more rapidly than sensations. Now there 
are, of course, countless forms of the relations of dependency 
existing between numerical values, where one numerical series 
increases faster than the other. If, for instance, we multiply 
every number by itself, we obtain from the series, I, 2, 3, 4 
. . . another series, I, 4, 9, 16. . . . The first numbers 
are known as the square roots of the second ; the latter are 
called the squares, or second powers, of the first. So that if these 
two series expressed the relation of stimulus and sensation, we 
should say the sensation is equal to the square root of the 
stimulus. A similar numerical series, differing from this only 
by its more rapid increase, can be obtained by multiplying each 
number by itself twice or three times, and so obtaining its third 
or fourth power. If either of these series expressed the rate of 
stimulus increase, we should say that the sensation is equal to 
the third or fourth root of the stimulus. But sensation-intensity 
increases neither as the square root, nor the cube root, nor as 
any other root of the stimulus-intensity. This is plain from the 
fact that the stimulus-increments which condition definite in- 
creases of sensation-intensity stand in a constant ratio to the 
total stimulus-magnitude. Since, therefore, the relative stimulus- 
increments always remain equal, the relative numerical incre- 
ments in the series of numbers representing the stimuli must 
also be constant. This is not the case in the series cited. In 
the series, I, 4, 9, 16 . . . e. g., the numerical increments 
are successively 3, 5, 7, and the numbers to which these incre- 
ments are referable, i, 4, 9 ; but the ratios -f-, -|, |-, are not equal. 
If this case actually corresponded to the sensation-law, we must 
have obtained the fractions, -|, |-, *-, etc., or others which gave a 
constant result when the division was made. But neither the 
second nor the third nor any other powers give such a series. 

The Law of Sensation-intensities 41 

On the other hand, there is another numerical relation of very 
general application which exactly corresponds to the relation 
between stimulus and sensation. 

If we cast a glance at an ordinary table of logarithms, we 
notice that the numbers in it are entered in two columns ; one 
contains the ordinary numbers, the other the logarithmic 
numbers. We see at once that these latter increase more slowly 
than do the ordinary numbers ; just as magnitudes of sensation 
increase more slowly than magnitudes of stimulus. If the 
number i, e.g., stands on the one side, we find O on the other, as 
its logarithm. The logarithm of 10 is I, of 100 is 2, etc. Here 
also, then, in the case of numbers and their logarithms, we have 
two series which increase in very different ways. And if we 
look more closely, we find that this similarity is more than 
merely external. The logarithms of I, 10, 100, 1,000, are O, I, 

2, 3. What is the relation of the increase of those numbers to 
their magnitude ? When I is increased to 10, 9 is added ; when 
10 is increased to IOO, 90 ; when 100 to 1,000, 900. The ratios 
of this increase are, therefore, -f-, %$, |$$. But these ratios are 
all equal, i.e., all equal to 9. Now this is an expression of the 
law which regulates the increase of sensation. Sensations in- 
crease by equal magnitudes, when the increase of stimuli is such 
that each increment stands in a constant relation to the particular 
total stimulus-magnitude ; and the logarithms increase by equal 
magnitudes, when the increase of their numbers is such that 
each increment stands always in the same ratio to the corre- 
sponding numerical magnitude. So that we can say that sensa- 
tions increase as logarithms when stimuli increase as their 
numbers ; or, still more shortly since we may express any 
stimulus-magnitude by some definite number sensation increases 
<as the logarithm of stimulus. 

Logarithmic tables were naturally in use long before psycho- 
logy felt the necessity of them. Indeed, the expression of the 
dependency of sensation upon stimulus is merely that of a very 
simple relation, of frequent occurrence in the expression of the 
dependency of magnitudes in general. The logarithms o, 1,2, 

3, e.g., differ each from its neighbour by the same amount, I : 
while the corresponding numbers I, 10, 100, I.OOO, differ from 
one another by the same multiple : i.e., by ten times their value 

42 Lectures on Human and Animal Psychology 

in each instance. But if this were the only rule we possessed 
for finding logarithms, the process would be exceedingly tedious. 
The matter is happily very much simpler. If we raise a number 
to all its possible powers, we get from it, of course, other 
numbers. Thus lo 1 = 10 ; io 2 =ioo; iO 3 =i,OOO. It is clear 
that by thus raising the powers of a single number we can 
obtain any number whatsoever. For if we take the ij, i^-, ii 
powers of IO, they give us numbers lying between 10 and 100 ; 
the powers 2j, 2-i, 2\, give numbers between IOO and i,OOO. 
And if we take all the possible fractional powers, we shall obtain 
all the possible numbers between 10 and 100, between 100 and 
1,000, etc. In order to obtain also the numbers which are 
smaller than 10, we must not multiply the number 10, but 
divide it so many times by itself. We must raise it, as the 
mathematicians say, to negative powers. Thus IO- 1 -^ ; iO- a 
= -j-^j-, etc. But between IO 1 and IO- 1 stands 10 or lo 1 - 1 : i.e., I. 
If we take as well the intermediate fractions of these negative 
powers, there result all the possible fractional numbers ; while 
between the powers o and I come all the numbers between I 
and 10. We have, therefore, obtained every possible number 
simply by raising the single number 10 to all its powers. Now, 
if we compare the powers o, I, 2, 3, with the corresponding 
numbers I, 10, IOO, 1,000, we see that the latter stand to one 
another in the same ratio as the logarithms to their numbers. 
The former increase by equal increments, when the numbers 
resulting from the involution increase by equal multiples. The 
indices of the powers are therefore nothing but the logarithms 
of the numbers which we obtain by the process of involution. 
And we can now formulate the sensation-law as follows : sensa- 
tions stand to their stimuli as the indices to the numbers arising 
from involution. 


But now a certain doubt may arise with regard to this 
paralleling of indices and logarithms with sensations. There 
are negative indices, as we have seen ; and, consequently^ 
negative logarithms. If we divide the number 10 by itself once, 
twice, three times, and four times, we obtain the powers 
o, -i, -2, -3, or the logarithms o, -I, -2, -3. The number of these 

Significance of Negative Sensation-values 43 

negative logarithms is just as unlimited as the number of the 
positive. This will be perfectly intelligible when we remember 
that the negative powers and logarithms signify fractions. If we 
continue the series IO- 1 , io- 2 , io- 3 , or ^ y^j-, y^Vo-- we reach 
successively smaller and smaller fractions. Just as the series of 
whole numbers only terminates at infinity, so with the series of 
fractional numbers. If, then, we wish to reach zero by the 
method which we have described, it will be necessary to divide 
io by itself an infinite number of times. Thus the logarithm 
corresponding to zero is negative, and infinitely large. But is 
all this applicable to sensations ? Are sensations ever negative ? 
And can there be sensations which, besides being negative, are 
also infinite ? 

When we speak of negative sensations, we ordinarily under- 
stand by the term sensations which are opposite in direction to 
other sensations which we call positive. Cold, e.g., is a negative 
sensation as opposed to hot. But it would be equally correct 
to call cold positive, and thus to make hot a negative sensation. 
The terms ' positive ' and ' negative ' are, here as elsewhere, the 
expression of an opposition. The negative is by no means 
nothing : it is just as much a real magnitude as the positive ; 
and the terms we apply are in themselves arbitrary. A shop- 
keeper reckoning up his effects, counts everything which he has in 
the till, or that others owe him, as positive ; his own debts he 
regards as negative. If, on the other hand, he is estimating his 
debts, he considers them as positive, and the contents of the till 
and his loans as negative. The result is the same in both cases. 
Or if a geometrician wishes to distinguish directions in space, he 
names that direction negative which he does not name positive ; 
which becomes which is quite immaterial. Just in the same way 
we characterise the logarithms of fractions as negative because 
we have already used the positive denomination for the 
logarithms of whole numbers. We must guard ourselves against 
supposing that we have here anything more than a mere con- 
vention, even though this convention is the most natural and 

The question arises then whether we may not speak of 
negative sensations, using the word in the above sense of simple 
opposition. No one will hesitate to answer this question in the 

-44 Lectures on 'Human and Animal Psychology 

affirmative, if 'it can be once shown that such an opposition 
exists among sensations. It is of course unnecessary to say that 
oppositions like that of hot and cold do not concern us in the 
present instance. Hot and cold are differences of sensation- 
quality, about the nature of which we have here as little 
to inquire as about the differences between agreeable and 
disagreeable, pleasant and unpleasant. It is true that these 
attributes are predicated of sensations of opposite character. 
And if we were subjecting these to a special investigation, we 
might not only justifiably, but very naturally, express the 
antitheses of hot and cold, pleasurable and painful, by positive 
and negative magnitudes. But our business in this first instance 
is only with the intensity of sensation ; and all other sensation- 
properties are, therefore, excluded from our consideration. 

We found the natural zero-point of our scale to be the point 
where sensation begins, where we first sense at all. Can there 
;be sensations which are not sensed ; or does the putting of that 
question involve a contradiction of terms ? 

There certainly is a contradiction. But it is only an apparent 
one, due to an equivocal use of the word ' sense.' We have 
already seen that there exist sensation-differences which are 
not sensed (p. 22). It is obvious that two different meanings 
have been given to the word. In its first signification the sensa- 
tion is simply something which depends upon an alteration of 
stimulus, no matter whether we detect this alteration or not. 
But, secondly, it is our discovery of such alteration, which is 
denoted by sensation. And this is equally true for sensations 
taken absolutely. In speaking of sensations which are 
too weak to be sensed we are regarding them as something 
independent of our apprehension of them ; we are considering 
them merely as conditioned by external stimuli. We can put 
the matter in this way. A sensation-difference is not at all 
identical with a sensed difference ; the latter implies a definite 
intensity of the former, And a sensation may exist long before 
it can be sensed. We only sense it when it reaches a definite 
intensity. But though in this statement we recognise the 
equivocation, we have not done away with it. The equivocation 
is explained by the fact that when the word first appeared in 
language the nai've consciousness which produced it knew only 

Significance of Negative Sensation-values 45. 

those sensations and sensation-differences which it was itself able 
to recognise as such. Not till scientific reflection had arisen 
was the human mind forced to the conclusion that there must 
be sensations and sensation-differences, which it was inadequate 
to recognise for the reason that sensations neither arise nor 
alter abruptly, but only through continuous gradations. 

So that there is nothing left for us but to use the word ' sensa- 
tion ' here and in what follows to express all those sensations and 
sensation-differences which we do not perceive, but whose 
existence we must assume to explain those which we do perceive, 
as well as sensations in the narrower sense of processes which we 
are able clearly to apprehend. Where it becomes necessary to 
make a distinction we will call sensations and sensation-differ- 
ences of the latter class ' noticeable,' and of the former ' unnotice- 
able.' Now, since we observe that a sensation must have attained 
a certain magnitude if it is to become noticeable, and that, other 
things being equal, it gains in intensity the greater its magnitude 
becomes, we are surely justified in taking as the zero-point of 
our sensation-scale the point where sensation becomes just 
noticeable. That settled, we shall naturally call the noticeable 
sensations, to the right of that point, positive ; the unnoticeable 
sensations, to the left of it, negative. For noticeable and 
unnoticeable denote a direct antithesis, as valid as that of cold 
and hot, or of opposing directions in space. 

We conclude, therefore, that our comparison of the relation 
in which sensation stands to stimulus with the relation of 
logarithms to their numbers holds with regard to this further 
point of the opposition between positive and negative. And 
we can now produce our scale beyond the zero-point in a nega- 
tive direction until the stimulus vanishes, as has been done in 

Fig. 4. 

And now at length we have our sensation-law in its 

most general form. How many 
units must we enter on the 
negative side to the left of o 
before we reach the zero-point 
of the stimulus ? The stimu- 
lus zero-point in this connec- 
tion is not, of course, the ex- 
ternal process- of movement: 

46 Lectures on Human and Animal Psychology 

affecting our sense-organs, and which has just attained the 
lower limit of efficiency, but the internal stimulus in the brain 
resulting from the former, and paralleled as physical process with 
the mental process of sensation. For it may be assumed that 
there are external stimuli too weak to reach the brain, whether 
because of their inability to affect the organ of sense, or because 
they cannot be conducted from it to the brain. This assumed, 
where will the line which expresses the increase of stimulus 
with increase of sensation cut the sensation-scale? We can 
obviously extend our negative sensation-units to infinity with- 
out arriving at that point ; for if we suppose, e.g., that the 
stimulus decreases by i of its magnitude at each division of the 
scale, it yet decreases more and more slowly ; and though at 
last it becomes exceedingly small, it does not disappear so long 
as the negative sensation-units which we are positing are ex- 
pressible in numbers. Only when these numbers become 
infinite may we assume that the corresponding stimulus-magni- 
tudes are also infinitely small, i.e., so small that we may without 
hesitation regard them as zero. Once more, then, we have the 
same relation as that of logarithms to their numbers. If we 
extend further and further the fractional series -j^-, y^-, 1 ^ , we 
do not come upon any fraction, however small, which is not 
greater than o. We should only reach o at infinity ; and, there- 
fore, the negative logarithm corresponding to it is infinitely 
large. In the same way, we may conceive of a stimulus as 
divided and subdivided as long as we please, and nevertheless 
the smallest particle of it would still be a stimulus. The 
stimulus only becomes equal to zero at infinity, and the nega- 
tive sensation corresponding to a stimulus equal to zero must, 
therefore, be infinitely great ; and since a negative sensation 
means the same thing as an unnoticeable sensation, an in- 
finitely great negative sensation will simply be that sensation 
which is less noticeable than any other, just as it may be 
asserted of O and co that the first is smaller and the latter 
larger than any other number. 

Our analogy between the logarithmic law and the law of 
sensation is now incomplete in one point only. We saw that 
all possible numbers can be obtained by raising a single num- 
ber to all its possible powers The positive powers give us the 

Units of Stimulus and Sensation 47 

whole numbers ; the negative, the fractions ; and the zero 
power gives us unity. All these facts we have found to possess 
a definite significance in the case of sensation. But we have 
left one point still undetermined ; that is the number whose 
involution gives us all the other numbers that are possible. In 
the instance which we took, we raised the number 10 to the 
powers o, I, 2, 3, and obtained the series I, 10, 100, 1,000. Had 
we taken some other number than 10 and raised it to those 
powers, we should have obtained a different series. It is im- 
portant, therefore, to know what number it is which has been 
chosen as the base by whose involution the other numbers are 

It is obvious that this must also be an important question for 
the sensation-law, since sensations stand to stimuli as their in- 
dices to the numbers obtained by involution ; and it is evident 
that we can only say what stimulus-magnitudes correspond to 
the sensations I, 2, 3, if we know what definite number was 
taken as the base in this case of involution. Our choice of that 
number is entirely arbitrary. For our sensation-scale it is im- 
material ; it conditions only the divisions of the scale. We 
shall plainly have the most convenient division if it is so carried 
out that magnitudes of sensation may be calculated directly 
from magnitudes of stimulus, and vice versd. But this is possible 
only when sensation is the simple logarithm of stimulus, and 
not some multiple or fraction of this logarithm ; and this 
depends entirely on the absolute magnitude of our unit of 
stimulus and our unit of sensation. Both of these magnitudes 
may be arbitrarily chosen when we have once made it clear to our- 
selves what they mean. We have already seen that the stimulus 
must be taken as equal to i where the sensation is equal to o 
i.e., is just noticeable, for i, 10, 100, are all equal to I ; or, in 
other words, the logarithm of I is always o. That determines 
once for all the magnitude of the stimulus-unit. Now, if the 
sensation I is also to come at the point where its stimulus is the 
number corresponding to the logarithm I, we must mark it (10 
being, e.g., the base employed) at the point where the stimulus 
has attained the magnitude 10. Had 100 been the basal number, 
we must have placed I where the stimulus had the magnitude 
100, and so on. For io l =io, ioo 1 = ioo, and every number 

48 Lectures on Human and Animal Psychology 

raised to the first power is equal to itself. Further, if we mark 
in more of our sensation-units, the divisions 2, 3, 4, take their 
necessary places where the stimulus-magnitudes are 100, 1,000,. 
10,000, etc. For io 2 =ioo; io 3 =i,ooo; lo 1 ^ 10,000. This is 
all required by our law, as we have seen, if the stimulus 10 
corresponds to the sensation I ; so that now we have also de- 
termined our sensation-unit. It is equal to the number which 
we have chosen as base. Under these conditions, when the 
stimulus is represented by the number obtained by involution, 
the sensation corresponds to the index ; or the sensation is 
equal to the logarithm of the stimulus. 

In our ordinary logarithmic tables 10 is the base by whose 
involution all the numbers are obtained. So that, if we wish to 
calculate sensations from stimuli, we have only to call that 
sensation I which is occasioned by a stimulus-magnitude ten 
times as great as that which lies at the limit of noticeability. 
Having done this, it is only necessary, when a particular 
stimulus-intensity is given, to look up in the logarithmic tables 
the number which expresses that intensity ; the logarithm in 
the next column gives at once the magnitude of sensation. To 
return to our previous example, if a weight of -^ gramme pro- 
duces a just noticeable sensation, we call -$ gramme stimulus 
I. Pressure by ten times this stimulus, i.e., by -i- gramme, we 
call sensation I. Now it is easy to determine at what weight 
the sensation is any whole or fractional number of times 
greater, or by how much the weight must be increased in order 
to condition a particular increase of sensation. If we wish to 
get a sensation 2\ times as intense as sensation I, we refer to 
our table, and find for the logarithm 2*5 the number 316. That 
means 316 stimulus-units, or Vo~ 6= <5'3 grammes. Or if the 
problem is to determine how great the sensation is which is 
occasioned by a stimulus of 5,000 units (100 grammes), we look 
up the number 5,000, and find its logarithm, y6gS. That is, a 
pressure of 100 grammes produces a sensation which is 3'6Q8 
times as great in intensity as the sensation arising from the 
pressure of 4- of a gramme. 

We have now completely answered the question which was 
before us. Not only have we discovered the law of the de- 
pendency of sensation upon stimulus, but we have indicated 

Significance of Negative Sensation-values 49 

the method by which the intensity either of sensation or of 
stimulus can be calculated when the intensity of its correlate is 
given. This method is simplicity itself, for it presupposes no 
more knowledge than that of the multiplication-table and nc 
more apparatus than a book of logarithms. 






HOR the solution of all the problems which may arise in 
any definite sense-department, there are required, as we 
have seen, two kinds of measurement. First, we must 
know the constant relation in which alteration of sensation- 
intensity stands to alteration of the intensity of stimulus ; and, 
secondly, the magnitude of the just noticeable sensation must 
be determined. The first of these measurements we have 
carried out ; the second now remains to be performed. 

Pressure-sensations afford us the simplest conditions for our 
investigations. We lay upon that portion of the skin whose 
sensibility is to be tested small weights, preferably of cork or 
pith, and seek to ascertain what magnitude of weight is 
necessary for the production of a just noticeable sensation. 
Observations made in this way have shown that the sensibility 
of the skin at different parts of its surface is very far from being 
uniform. The most sensitive portions are the forehead, temples, 
eyelids, the outer surface of the fore-arm, and the back of the 
hand. We can usually sense on these parts weights of only 
.^-^ gramme. Less sensitive are the inside of the fore-arm, the 
cheeks, and the nose, and very much less sensitive than these 
the palm of the hand, the abdomen, and the thigh. Here the 
sensibility sinks to about -fa gramme. On some specially 
protected parts, e.g., the nails and the heel, the just noticeable 
weight rises as high as a whole gramme. 

Far more adequate for the apprehension of weak stimuli is 
our organ of hearing. A mere touch of the external auditory 

The Just Noticeable Sensation 5 1 

meatus or any contact with the tympanic membrane excites, as 
we all know, a fairly strong sound-sensation. And even a distant 
sound must be very weak indeed to be imperceptible. In 
making observations for the determination of the limit of 
auditory sensibility, we must, of course, never forget to take 
into account all the conditions upon which the intensity is 
dependent. If, e.g., we measure the sensibility of an ear by the 
sound produced by a falling weight, we must know, not only 
the magnitude and material of the weight, but also the material 
of the body upon which it falls. And we must, further, de- 
termine the rapidity of its fall and the distance of our ear from 
the place where the sound is produced. It has been discovered 
that a normally sensitive ear can just sense the sound made by 
a cork pellet, weighing I milligramme, in falling through a 
height of I millimetre, at a distance of 91 millimetres. That 
we may expect to find considerable difference in different 
individuals is a matter of course, justified by our everyday 
experience. Diseases of the sense-organ affect our hearing ; 
and, in addition to this, as old age draws on, the acuteness of 
this sense usually declines, passing through the most various 
stages from hardness of hearing to complete deafness one 
of the commonest defects of sense. 

If we are to use the sound-magnitude which we have just 
determined as a unit of stimulus, we must be able to compare 
with it the intensity of all other sounds which are employed as 
stimuli. The comparison is not difficult. Given the sound 
whose intensity is to be measured, we need only to remove it 
to the distance at which it just disappears. It is then precisely 
as great as the sound made by a cork weighing I milligramme, 
falling through a height of I millimetre upon a sheet of glass, 
at a distance of 9 i millimetres from the ear. That distance 
tells us at once how many times greater the given sound at the 
place of its production is than the just noticeable sound- 
intensity. An ordinary musket-shot is just audible at a 
distance of 7,000 metres. This distance is rather more than 
70,000 times as great as the distance 91 millimetres. Since the 
intensity of the sound decreases as the square of the distance, 
it follows that the sound-intensity of the musket-shot is more 
than 4,900,000,000 times that of the cork pellet which we 

52 Lectures on Human and Animal Psychology 

adopted as our unit. A similar comparison with the unit may 
be carried out for any other sound. We could easily determine, 
e.g., how many sensation-units are comprehended in a definite 
stroke of the sound-pendulum (p. 30). And since we can 
easily take the next step, and compare the various sound- 
intensities with one another, it is perfectly possible to express 
intensities of sound by means of a single scale like that which 
we used for weights. There is only one condition which we 
must be careful not to disregard : we must never make an 
observation while other noises are affecting the ear, or while the 
movement of the air renders the propagation of sound irregular. 
The quiet of night is therefore especially suitable for the 
experimental measurement of sound-sensations. 

The conditions are different when we are dealing with the 
sense of sight. It is obvious that we can only attempt to 
determine the just noticeable sensation, if there is possible for 
the sense-organ a state of absolute inactivity, during which 
there is no sensation whatever. This condition is realisable for 
the ear. We clearly distinguish noise from silence, as a state 
of things where auditory sensations are wanting. The corre- 
sponding distinction for the eye would be that between dark 
and bright. But visual darkness is something quite different 
from auditory silence. By greatly diminishing the intensity of 
light we may obtain darkness without there necessarily being 
any actual disappearance of the external light. Or if we close 
our eyes we are also in darkness ; but it does not follow that 
we are completely destitute of light-sensations. In nearly 
every case, a certain amount of external light penetrates to the 
closed eye. And not only that, but the closing of the eye is 
the cause of a light-sensation, the pressure on the eyeball 
serving as a retinal stimulus. You may easily convince your- 
selves of this by making the pressure somewhat stronger ; the 
weak shimmer which you still see, though you have closed your 
eyes, is thereby intensified, till finally the whole of the darkened 
field of vision is flooded by a sea of light. 

But even in the absence of this mechanical stimulus, and 
even in the darkest night, our eyes are never free from light- 
stimulation. With a little attention we can see that the dark- 
ness deepens and lifts, gives place here and there to a brighter 

The Just Noticeable Sensation 53 

twilight, which is in its turn followed by a still denser black- 
ness. We can even persuade ourselves at times that we re- 
cognise the blurred outlines of external objects ; now and again 
a brilliant flash of lightning seems to irradiate the shadows. 
So that the eye is always active, however complete the dark- 
ness, and we may easily find ourselves doubting whether it is 
due to a light from our eye or from the night itself that we 
are able to see. But we may readily convince ourselves that it 
is no external light which we have to thank for these phenomena 
of light and darkness. If we move, they accompany us ; they 
correspond to no external object ; they persist though we have 
assured ourselves with all possible care of our complete isolation 
from external light. But more, not only this changing shimmer 
which we observe in the dark, but even the deepest black 
that we can see, is always a light-sensation. When we close 
our eyes, our darkened field of vision possesses the same form 
as the bright field of the open eye. All that lies within the 
limit of this field we see black ; whatever lies outside of it 
we see not black, but not at all. When it is daylight, and our 
eyes are open, we do not say that the objects behind our back 
appear black to us. So that the blackest black which we can 
see is our weakest sensation of light. To sense this is not the 
same as to have no sensation whatever. And it follows that 
there are degrees of darkness, as of light ; that there are 
differences of blackness, that we may pass gradually from the 
deepest black to a brighter, from that to grey, and so finally 
to white. 

We see then that the view of the ancients, that the eye is 
itself the source of light, is not without a certain foundation. 
Only we can never see and recognise external objects by means 
of this light. The light-sensation which we have in the dark 
is caused by a stimulus within the eye. But if we are to see 
objects, the light-stimulus must proceed from them. That 
there should be a continual excitation of a sense-organ is 
certainly a peculiar state of things, probably not occurring else- 
where ; but it becomes intelligible when we remember that the 
eye is by far the most sensitive of the sense-organs. A stimulus 
which is not nearly strong enough to occasion a sensation of 
hearing or of pressure is considerably more than just noticeable 

54 Lectures on Human and Animal Psychology 

for the eye. In this latter case the normal physiological con- 
ditions of the organ may very probably furnish the occasion 
for a sensation ; the chemical processes which constitute nutri- 
tion may possibly serve to stimulate the neural epithelium of 
the eye. Less constant stimulation is caused by the pressure 
exerted upon the eyeball by the muscles which move it. This 
stimulus will always be operative even in rest, since the muscles 
are never entirely relaxed ; but it increases in intensity during 
movement. We can observe the same phenomenon in the 
light-sensations which we have in the dark. They, too, become 
more intensive when the eye is moved. 

It is now self-evident that the conditions of vision prevent us 
from measuring the magnitude of the stimulus which corre- 
sponds to a just noticeable light-sensation. The eye always has 
a sensation which is more than just noticeable, and all stimuli 
which affect us can, therefore, simply increase the intra-ocular 
light-sensation which is inevitably present. It only remains for 
us in this case to determine the least light-intensity, which is in 
absolute darkness just noticeably brighter than the black of the 
field of vision. We can most easily obtain very weak light- 
intensities of this kind by passing a constant current through 
a metal wire. As we increase the intensity of the current, the 
wire becomes hotter and hotter, till at a definite temperature it 
begins to be luminous. And since we can graduate the strength 
of a galvanic current at our pleasure, the intensity at which the 
luminosity of the wire becomes just noticeable can be readily 
determined. We have then only to compare its objective value 
with that of other known light-intensities. It has been found 
in this way that the just noticeable intensity of light is approxi- 
mately 3-^5- of the light of the full moon reflected from white 

The investigations to which we have referred furnish us ap- 
proximately with our units of sensation and stimulus for 
pressure, sound, and light, though in the latter case with the 
limitation rendered necessary by the existence of the intra- 
ocular light. No successful attempt has yet been made to deter- 
mine these units for the other sense-impressions, for taste, smell, 
and temperature. This is partly due to the fact that we are not 
able to control the operation of stimuli in these departments 

The Just Noticeable Sensation 55 

with sufficient accuracy ; and it is in part caused by the general 
impossibility of putting the organ into a condition of total free- 
dom from stimulation, a condition, that is, which would corre- 
spond to the zero-point of our stimulus-scale. 

Now that we have determined in this way the just noticeable 
stimulus- difference and the just noticeable stimulus-magnitude, 
the two magnitudes upon which our measurement of sensation 
depends, there arises a further question : do these two magni- 
tudes stand in any definite relation to each other ? If our 
sensibility to stimulus shows a certain variability, will not also 
our sensibility to stimulus-difference be variable ? We saw that 
this latter is expressible by certain constant fractions ; that, e.g., 
our sensibility to differences of pressure is , to differences of 
light is j^j- : in other words, that a pressure must be increased 
by % of its magnitude, a light by y^-g- of its intensity, if the 
difference is to become noticeable. Are these relations really 
constant, as we have asserted ; or is it not rather highly probable 
that they vary with variations in sensibility? 

Obvious as it may appear to answer the latter question in the 
affirmative, more careful reflection will at once convince us that 
the opposite is to be expected if the general law of the depend- 
ence of sensation upon stimulus holds. This law informs us, 
you remember, that a stimulus, whether great or small, must 
always increase in the same ratio, in order to condition a defi- 
nite sensation-difference. Suppose, therefore, that the sensibility 
of some sense were, in an exceptional case, reduced by one-half. 
It would then, of course, be necessary to take twice as great a 
stimulus as before in order to occasion a noticeable sensation ; 
and if we wished to increase this sensation again by a noticeable 
magnitude, the larger stimulus would naturally need, as the law 
says, a relatively greater increase. But there is not the least 
reason for supposing that this increase must be greater than the 
proportion originally required. 

This hypothesis is completely confirmed at every point by 
observation. If sensibility has changed, every stimulus is sensed 
more or less intensely than before ; but if two stimuli are com- 
pared, their difference is just as great in sensation as it was pre- 
vious to the change. If the sensation I is doubled, the sensation 
2 is also doubled. If a stimulus I had to be increased by to 

56 Lectures on Human and Animal Psychology 

alter sensation, then when, on account of the decrease of sen- 
sibility, the stimulus 2 must be substituted for it in order to pro- 
duce the same sensation, this latter stimulus must be increased 
by -|, if sensation is to be altered, etc. In short, sensibility to 
stimulation does not affect in any way the law of dependency of 
sensation upon stimulus. 


We may now return to the point from which we set out. Our 
object was to investigate the dependency of sensation upon 
stimulus. Stimulus, as the physical process directly parallel to 
the sensation, means here of course the internal stimulus, opera- 
tive in some sensory centre of the brain. But, to make our 
problem easier, we began by investigating the dependency of 
sensation upon external stimulus. The time has now come for 
raising the question whether it is at all probable that the trans- 
lation of external into internal stimulus has in any way influenced 
the connections which we have found. We have, in fact, already 
seen that stimulation- processes are only set up in the sense- 
organs and nerves when the external stimulus has attained to a 
certain intensity ; and since it is not until this process reaches 
the brain that it is immediately accompanied by sensation, a 
stimulus which is weaker than this is naturally the same to us as 
no stimulus at all. On the other hand, it is equally conceivable 
that the internal stimulation-process must have reached a definite 
intensity before it gave rise to a noticeable sensation. 

As a matter of fact, there can be no doubt that both these 
conditions are realised. It is a necessary consequence of the 
more or less protected position of the sensory nerves and their 
peripheral end-organs that exceedingly weak stimuli cannot 
affect them. And it is just as certain that the stimulation- 
process in the brain is only perceived by us at a certain intensity. 
This is sufficiently obvious if we consider the causes which 
condition change of sensibility. If we direct our attention to the 
impressions of any sense-organ, we can apprehend much weaker 
stimuli than is the case when our attention is first aroused 
through the force of the impressions themselves. But it is not 
probable that the conditions of conduction to the brain have 

Upper and Lower Limit of Weber s Law 57 

altered in the two cases. We are always subject to a large 
number of external impressions, but only a few of them are 
perceived by us. Nevertheless it is true that these impressions, 
acting jointly, are capable, unless they are very weak, not only 
of exciting the sensory nerves, but of passing along them to the 
central organ. 

Now, just as there is a lower limit, below which the external 
stimulus is too weak to occasion an internal stimulation, may 
there not also be an upper limit, above which it is impossible to 
arouse any stronger neural excitation ? If this is so, we shall 
expect to find that the law which is valid for moderate stimuli 
does not hold in the case of the strongest. 

As a matter of fact, it may easily be proved that neural 
excitations can never be increased beyond a certain point. The 
preservation of the nerves and their end-organs renders this 
necessary. If we stimulate the eye with stronger and stronger 
light, we shall at last injure the power of vision, or, indeed, 
entirely destroy it. The processes in the sensory nerves depend 
upon the constant renewal of the substances provided by the 
blood. The more intensive the sensory processes, the more 
energetically must the renewal be attended to. And since this 
cannot go on indefinitely, it is evident that the intensity of the 
neural processes has also its limit of increase. We do not as a 
rule reach this limiting point suddenly in the process of stimula- 
tion, but rather approach it gradually. At first the neural 
process increases in intensity in direct proportion to the external 
stimulus ; later this increase becomes somewhat slower ; finally it 
ceases altogether, however much we may continue to increase 
the intensity of the stimulus. We must, therefore, necessarily 
expect that the relation of the just noticeable sensation-difference 
to the total magnitude of stimulus is in reality not altogether 
constant, but slowly varies with the gradual increase of stimulus. 
If, e.g., a moderate pressure upon the skin must always be 
increased by ^-, a very intense pressure will require a somewhat 
greater increase ; and finally there will be a certain sensation of 
pressure an increase of which is absolutely impossible, however 
heavy the weights we place upon the stimulated part 

Many phenomena of our everyday life are to be explained on 
this principle. It is well known that extreme pain admits of no 

58 Lectures on Human and Animal Psychology 

degrees or distinctions ; that a very intense light blinds us ; an 
excessively loud sound deafens us. But the possibility of sensa- 
tion-increase does not cease abruptly, but by degrees. If we 
compare the shadow thrown by an object in moonlight with the 
shadow cast by the same object in sunlight, it will be at once 
seen that the former appears much darker than the latter. In a 
landscape seen by moonlight, this stronger contrast of light and 
shade makes the illumination far brighter, although it is absolutely 
much less intense. And from this fact we can distinguish at the 
first glance whether a picture represents a moonlight or a day- 
light scene. It is not in the power of the artist to mark this 
difference by an absolute difference of light-intensity. Both his 
paintings are equally bright; but he makes the difference between 
light and shadow greater in the first picture than in the second, 
and by this single device enables us to distinguish in a moment 
the night scene from the day scene. This device would be 
impossible if it were exactly and invariably true, as our law puts 
it, that an equal sensation-difference always corresponds to an 
equal difference-relation of light-intensity. For our two land- 
scapes are a case in point The moonlight shadows differ from 
moonlight by a quantity of light which is relatively to the in- 
tensity of the moonlight just as great as that separating the sun- 
light shadows from the sunlight ; so that the light-intensity of 
the moonlight shadows stands to that of the sunlight shadows 
as moonlight to sunlight. Nevertheless the light of the moon 
appears much brighter in relation to its shadow ; i.e., the sensa- 
tion-difference is greater here, where the light-stimulus is less, 
than it is in the case of sunlight, where the light-stimulus is more 

All the various influences which condition divergences from 
the simple law of the dependency of sensation upon stimulus 
have proved to be due to the intermediary processes of neural 
excitation. And it is a justifiable assumption that the law is 
literally valid as between internal stimulus and sensation. So 
that if we were able to measure directly the intensity of the 
stimulation-process in the brain, instead of the external stimulus, 
we should find the law holding without exception. In investi- 
gating the relations of sensation to stimulus, we have, as a 
matter of fact, been observing the effect of two laws : the law of 

Psychological Interpretation of the Law 59 

the dependency of internal upon external stimulus, and the law 
of the dependency of sensation upon the former class of stimuli- 
If we suppose that the intensity of the internal stimulation 
remains within certain limits proportional to that of the external 
stimulus which occasions it, but that, as the external stimulus 
continues to increase, it increases more and more slowly, we 
have a simple explanation of the deviations from the law of the 
logarithmic relation between stimulus and sensation. We have 
been unable to investigate this relation without the constant 
intrusion of the nervous excitation, an intermediary which has 
unfortunately remained hitherto inaccessible to the method of 
exact physiological examination. 

And now we approach the final question which is suggested 
by a consideration of the law of the dependency of sensation- 
intensities : the question of the psychological interpretation of 
the facts which we have hitherto ascertained. 


The discovery of a law only becomes of cardinal importance 
when we have learned to know its connections. The relation 
between sensation and stimulus is of importance because the 
knowledge of it allows us for the first time in the history of 
psychology to apply principles of exact measurement to mental 
magnitudes. But this measurement will not have its proper 
value until we have learned in what peculiarities of sensation, or 
of the organ which transforms the stimulus into a sensation, the 
law has its basis. Is the relation physically conditioned by 
processes in the nervous system ? Or is it psychically condi- 
tioned by the nature of the mind ? Or, finally, does it express 
the interconnection of the world without and the world within, 
which is conditioned by both these factors ? Is it, in a word, to 
be explained in terms of psychophysics ? 

It has been often assumed that our law possesses only a 
physiological significance. As the stimulus, even in acting upon 
the external sense-organs, must have reached a certain intensity, 
if it is to cause an excitation in them, this process of excitation 
will perhaps meet with increasing obstacles in the senary 
nerves, but especially in the central organ. May not now, it is 

6o Lectures on Human and Animal Psychology 

said, these central obstacles increase with increasing stimulation 
just so that finally, in the sensory centre, where alone the pro- 
cesses of stimulation are sensed, the magnitude of the stimulation 
is only proportional to the logarithm of the external stimulus ? 

As yet this conjecture can neither be proven nor contradicted. 
For we know almost nothing regarding the law of the trans- 
mission of stimulation-processes in the brain. Considered more 
closely, however, it falls into two hypotheses. According to the 
first of these, it should result from a comparison of the internal 
stimulation-processes in the sensory centre with the external 
stimuli, if such a comparison were possible, that the former 
does not increase in direct proportion to the latter, but more 
slowly ; and in such a way that the logarithmic law of depend- 
ency arises. This view seeks support in the general postulate 
of the parallelism of mental and physical processes, according 
to which the bare fact of the logarithmic increase of sensation- 
intensities requires that the physical processes in the brain, 
corresponding to the sensations, shall also behave as these latter 
behave. According to the second hypothesis, this behaviour is 
the result of the gradual decrease of the intensity of the stimula- 
tion-process during its propagation through the central organ. 
We shall hardly find any facts that can be adduced in favour of 
this latter assumption. At any rate, what little we do know of 
the propagation of stimulation in the central organ (e.g., the 
laws of reflex movement, when the intensity of the stimulation 
is increased) speaks rather against than for it. Moreover, we 
cannot regard the hypothesis as probable, even if we grant 
the applicability of the law of psychophysical parallelism to the 
present case. To imagine that the processes of sensation, of the 
apprehension of sensation, and of its comparative measurement 
depend on their physical side on a simple transmission from a 
sense-organ to a definite part of the brain is to have a very 
crude and inadequate idea of that principle. It is surely plain 
that the different degrees in the clearness and relative notice- 
ability of sensations are secondary sensation-characteristics, 
which must occupy the most important place in any explana- 
tion of Weber's law. And there must be certain physical 
processes running parallel to these characteristics, if that law is 
to be physiologically interpreted. But mental processes of so 

Psychological Interpretation of the Law 6i< 

complex a character would necessarily be paralleled by complex 
physical concomitants, by the complicated interaction of various 
central areas. Only by their means could justice be done to the 
fact, expressed in the law, of the relative decrease of stimulation, 
with absolute increase of stimulus-intensity. However, be this as 
it may, no unverifiable hypotheses of this kind should prevent us 
from raising the question whether it is not also possible to dis- 
cover a psychological interpretation. This would have to take 
its place alongside of the physiological, just as the physical and 
mental processes themselves, as we have seen > are parallel and', 
not mutually dependent. 

This latter consideration tells equally against the third of the 
interpretations of Weber's law which we mentioned above, the 
psychophysical. According to this view, neither physiological 
nor psychological conditions suffice for the establishment of the 
law. It is rather to be regarded as a specific principle of the 
interaction of physical and psychical, a fundamental law which 
as such is not capable of any further explanation. 

We must urge, in the first place, that it is extremely difficult 
to conceive of the existence of a law of this kind. It belongs 
neither to the one territory nor to the other, but only to their 
borderland ; and it disappears when we leave that on the one 
side or the other. It seems as though such a hypothesis must 
inevitably take us back to the spiritualistic doctrines which 
proved so unfruitful for the explanation of the facts of mental 
life, those which regarded bodily and mental existence as two 
generically distinct modes of being, whose interconnection is 
merely external. We shall, therefore, decline to adopt this 
theory of mutual influence, which explains nothing, unless 
we find that a physiological or a psychological explanation 
is impossible. But we must remember that it was through 
the observation of psychological facts, of sensations, that the 
law was discovered. It is, therefore, only natural that we should 
ask for a psychological interpretation of it. The physiological 
interpretation must remain at present a general postulate, be- 
cause a relation of external to internal stimulus, like that ex- 
pressed by the law, is as yet only a matter of hypothesis based 
on the principle of psychophysical parallelism, and can by no- 
means be proven. As a matter of fact, we shall see as we pro- 

62 Lectures on Human and Animal Psychology 

ceed that the view which makes our law a psychical uniformity 
is supported by many other phenomena of our mental life. A 
psychological value is assured to it by its actual universality. 
But more than that, we are also able to show its dependence 
on definite psychological conditions which are everywhere 

What the law tells us, first of all, is simply this : that our 
sensation furnishes no measure of absolute, but only of relative 
magnitudes ; or, in other words, that we can only estimate 
magnitudes by comparison. If an excitation of a pressure- 
nerve is increased from the intensity I to i|-, this is the same as 
if an excitation of the intensity 2 were increased to 2-. The 
two differences are equal, if we compare them in ignorance of 
the absolute intensity of the two excitation-processes. We 
possess no mental measure whatsoever of absolute mental 
magnitudes. We are no more able to conceive of an absolute 
magnitude of sensation than we can have an idea of absolute 
time-magnitude, or of any other magnitude of a mental nature. 
It is well known that we continually make mistakes in estimat- 
ing absolute distances with the unaided eye, without the assist- 
ance of instruments of measurement, whereas the eye is an 
exceedingly accurate instrument for estimating differences of 
distance. And the same holds in every case where we are re- 
stricted to the means with which nature has provided us : we 
can only measure relatively ; we can only compare the magni- 
tudes which are directly given us. 

Measurement of sensation in general is rendered possible 
simply and solely by our reference of all qualitatively similar 
sensations as regards their intensity to an arbitrary sensation- 
unit. We cannot compare all possible intensities at once with 
one another, and in this way refer them to the unit which we 
have chosen. That method of procedure is excluded by the 
very notion of comparison. In the first place, we can only com- 
pare individual things. We can, therefore, never unite more 
than two units of sensation in a single comparison. We re- 
present to ourselves first one and then the other of these in- 
tensities, and so determine which is the stronger sensation. We 
can only pass to a third sensation, and estimate its intensity by 
comparing it with one of the two sensations which have been 

Psychological Interpretation of the Law 63 

already compared. In this way it becomes possible for us to 
bring a large number of sensations into one continuous series. 
We can only do this by proceeding successively from sensation 
to sensation, from comparison to comparison. But if \ve can 
never compare simultaneously more than two, never three or 
more, magnitudes, the consequence is obvious that our measure 
of sensation is relative, i.e., is always limited to a determination 
of the ratio in which two sensations stand to each other. It is 
no argument against this relativity to urge that we can always 
proceed to new comparisons, and so measure all possible in- 
tensities ; for the series thus obtained is still only composed of 
individual comparisons. In fine, then, this law of the logarithmic 
relation of sensation to stimulus is a mathematical expression 
for a psychological process of universal validity. 

But in saying this we have already answered the question 
which we left undecided when we entered upon our considera- 
tion of the measurement of the intensity of sensation. The fact 
that the more intensive stimulus requires a greater addition for 
the production of an equally noticeable sensation-increase ad- 
mits, as we saw, of a twofold interpretation. Either the more 
intensive sensation demands the operation of a more intensive 
stimulus for its increase by an equal sensation-magnitude, or the 
more intensive sensation demands a more intensive sensation- 
increase, if this latter is to appear equally noticeable. Our 
reference of Weber's law to the principle of the relativity of 
sensations favours the second of these interpretations. Our 
comparison is always relative. In order that a more intensive 
sensation-magnitude may increase by as much as a lesser sensa- 
tion, the sensation-increase must be correspondingly greater ; 
and two sensation-increases which lie at different parts of the 
sensation-scale will be equally noticeable when they stand in 
equal relations to the stimulation-intensities to which they are 





rHE intensity of a sensation is only one side of it. Not 
only has every sensation its own intensity, but also a 
definite quality which renders it distinguishable from other 

The most extreme instances of qualitative difference are 
furnished by the sensations of the different sense-organs, sensa- 
tions of eye, ear, and skin. A colour and a tone, a sensation 
of pressure and one of warmth, are simply incomparable with 
each other. Hence they are denominated disparate sensations. 
But qualitative differences are also to be observed within the 
sphere of one and the same sense. Thus red, green, blue, and 
yellow are entirely different sensations, although they are all 
sensations of sight. The one thing that proves a closer inter- 
connection among these sensations of a single modality is the 
possibility of a continuous passage between any two of them, 
the one passing by slow degrees into the other. Thus red may 
pass into green, or a low into a high tone. So that the 
relation between two different sensations within one sense is 
analogous to that between two points which lie within one and 
the same spatial continuum ; while two disparate sensations 
may be compared to points which belong in entirely different 
spaces, and whose position with regard to each other is un- 

There is no sense-organ within whose sphere qualitative 
differences do not occur in greater or less number. Sometimes 

Quality of Sensation 65 

those differences are very few, as in the case of temperature, 
where cold and hot appear as the only two sensible qualities. 
Sometimes the differences are of such a nature that they will 
not submit to any definite method of classification. The sensa- 
tions of pressure, e.g., obviously present marked qualitative 
differences ; an ordinary pressure upon the skin is a very dif- 
ferent sensation from the prick of a needle-point or the scratch 
from a rough surface. But plain as these differences are, it is 
impossible for us to state them in terms of any definite recip- 
rocal relation. 

Nor are we in a much better position with regard to the 
sensations of smell and taste. It is true that certain groups of 
odoriferous substances, which are for the most part chemically 
related, give rise to similar scents ; e.g., many of the ethereal oils, 
the volatile fatty acids, the metals, etc. But we are entirely in 
ignorance of the relation in which these various scents stand to 
one another. 

With taste-sensations we can go one step farther. Here there 
can be no doubt that the number of existent sensations is more 
limited, and therefore their investigation easier. If we exclude 
everything which does not belong to the sense of taste itself, 
there remain, it seems, only six sharply differentiated sensa- 
tions : sweet, sour, alkaline, metallic, bitter, and salt. In saying 
this, we do not mean to assert that these six are the only taste- 
sensations possible at all. It is clear that, e.g., by combining 
sweet and bitter, we can produce a taste which is neither 
sweet nor bitter, although it has something of both qualities. 
The result is a mixed sensation, not a qualitatively simple one. 
In ordinary life we are apt to think that we possess a far 
greater number of taste-sensations. But this is only because 
we do not commonly distinguish taste from smell. When we 
are tasting, we are smelling at the same time, and so there 
arises a combination of smell and taste resulting in a mixed 
sensation, which is referred solely to taste simply because our 
attention is principally directed to that sense. How much 
really depends upon the sense of smell can be readily seen by 
recalling a bad cold. In that state we discover with amazement 
that many things have absolutely no taste at all. Or, again, 
the influence of smell can be still more certainly eliminated by 


66 Lectures on Human and Animal Psychology 

filling both nostrils with water. In this experiment we are 
entirely confined to the sensations of taste proper. And we 
find that our tongue cognises no more than those six definitely 
characterised sensations. 

This examination of the sensations of taste serves to show us 
the method to be pursued in a more exact investigation of the 
quality of sensation in general. In every case our first question 
must be, whether there are not discoverable certain sensation- 
qualities which are incomparable with one another, and which 
are therefore to be regarded as pure and simple. When we 
have found these, and definitely established their number for 
.some particular modality of sense, we have to ask further : what 
are the compound or mixed sensations which arise from the 
simultaneous occurrence of two or more of the simple ones ? 
That is to say, in investigating any given sensation, we adopt a 
similar mode of procedure to that which the chemist employs 
in the investigation of a given body. We must first determine 
the elements of which the sensation is composed, and then go 
on to show what relations these elements bear to one another in 
the combination. Here, just as in the case of the measurement 
of intensity, we have to set out from definite units of measure- 
ment. But then, of course, these were units of quantity ; now 
we shall be dealing with units of quality. These units are com- 
parable to atoms, from which the sensation is built up. But, as 
you all know, the term ' atom ' means two different things. For 
the physicist it is a unit of quantity, for the chemist a unit of 
quality. So that in splitting up our sensations into quantitative 
and qualitative units we are analysing these mental states in a 
way which recalls the two chief directions of analysis of the 
material world without us. 

In the case of the sensations which we have had under con- 
sideration up to the present time, this analysis into qualitative 
units has been either not carried out at all, or only very im- 
perfectly. It is quite otherwise with the two senses whose high 
degree of functional development has gained for them the title 
of the ' higher senses ' : those of sight and hearing. 

Tone-sensations; Beats 67 

The quality of auditory sensations is given, first of all, with 
their pilch. With this is always connected the clang, a peculiar 
colouring of tone-sensation. Noise we distinguish from both, as 
a sound-impression in which pitch can be perceived either only 
uncertainly or not at all. 

The simplest of these three forms of auditory sensation is the 
pitch, although in reality this can never be separated from the 
clang, since it is only in a clang that a particular pitch is per- 
ceived. That need not, however, prevent us from disregarding 
for the time being everything that gives to a tone its peculiar 
clang-character, and attending simply to that property of it 
which we call its pitch. Indeed, a psychological analysis of 
sensations demands that such an abstraction be made, for its 
business is to continue analysing every sense-impression until it 
reaches the ultimate elements, which cannot be any further 
divided. Now the pitch is easily separable from the other 
elements of a musical clang. It may remain unchanged, while 
the clang-character of the impression varies. This happens, e.g., 
when we strike the same tone upon a number of different 
musical instruments. On the other hand, the pitch may vary 
to a certain extent, while the clang-character of the impression 
does not change. This happens when we sound neighbouring 
tones on the same instrument. When, however, the tonal pitch 
of the two impressions is very different, the clang-quality gene- 
rally changes with it, as is easily seen by comparing, e.g., two 
widely distant tones upon a piano. 

It was known to the ancients that tones consist objectively of 
vibrations of the sounding bodies and of the air carrying the 
sounds. Indeed, in the case of the very deepest tones, these vibra- 
tions are actually perceptible by the eye. In the same way the 
vibrations of sounding strings can be easily perceived by the 
eye. The best means of showing the origin of tones from 
vibrations is afforded by the siren, a physical instrument especi- 
ally constructed for this purpose. It consists of a disc provided 
with a series of circular holes, and moving across a current of 
air in such a way that within any given time the current is 

68 Lectures on Human and Animal Psychology 

interrupted as ofte'n as unperforated por- 
tions of the disc alternate with perforated. 
By regulating the velocity of the rotation 
of the disc we can produce at will high 
or low tones. The slowest rate of air- 
vibrations which can give rise to the 
perception of a tone is about 16 in i", 
though under favourable conditions it 
may sink to 8. The best means of pro- 

FIG. 5. ..i 

ducing these very deepest tones is given 

by large tuning-forks or vibrating steel rods. As we approach 
the limen of perception, however, the tone becomes so faint that, 
however extensive the vibrations, it can only be heard at a quite 
short distance. The deepest tones of the musical scale lie be- 
tween 32 and IOO vibrations in i". As the number of vibrations 
increases the pitch steadily rises. When the vibration- rate has 
increased to about 40,000, tone ceases altogether, and we hear 
only a hissing noise. 

It is only in the case of the very deepest tones, which cannot 
be employed for musical purposes, that we are able to distin- 
guish the air-beats corresponding to their vibrations. Our 
knowledge of the increase of vibration-rate in the case of the 
higher tones does not, therefore, depend upon the immediate 
perception of the vibrations, but upon another observation, 
which is closely related to it. As early as the days of the 
Pythagoreans it was a familiar fact that a string shortened to 
half its length vibrates twice as rapidly as the whole string ; 
that one reduced to a third of its original length vibrates three 
times as rapidly, one reduced to a quarter four times, and so 
on. Now, the tone of the half-string is the octave of the tone 
of the whole string ; the tone of the third part, the fifth of this 
octave ; that of the fourth part, the double octave. So that this 
law of the uniform relation of the length of a string to its vibra- 
tion-rate contains in it another important law, that those rela- 
tions of tones which are apprehended as harmonious correspond 
to simple ratios of their vibration- rates. 

Harmonious relations of tones were originally distinguished 
from inharmonious only by the more pleasing character of the 
impression which they made in a tone-series. Singing and 

Tone-sensations; Beats 69 

playing in unison came long before part-singing and harmony. 
But as soon as the custom arose of employing several voices of 
different register in the rendering of a melody, there came to 
light other phenomena, connected with the simultaneous sounding 
of tones, and with the consequent simultaneity of air-vibrations 
of different velocities. Not only, that is, can we distinguish a 
single tone from a clang compound of several tones, but we can 
easily hear out of such a clang, supposing that it is harmonious, 
the separate single tones which compose it. It is a matter of 
direct perception, e.g.> that the common chord of c major is 
composed of the three tones c eg. Whenever a compound clang 
is harmonious, the separate simultaneous vibrations unite to 
produce a common movement of the air, itself consisting of very 
brief and uniformly recurring /"A A A A A /\ 

periods. Fig. 6 shows this for 2:2 \A\A.\J\ \J[ ,\A wj 
the three compound clangs of a 
tone plus its octave, its fifth, and 
its major third. The points at A A A A A A 

which a new period begins are 2:3 i ; I j 

indicated in each case by dotted ' / \/\f\j\/\f\J^\J\/\J 

vertical lines. In the octave the 

two vibration-rates which unite (\S\J\S\jf\J\S 

to form the compound clang 4:5 U A A A A U A A 
have in each period the ratio. / \J \J \J 

1:2; in the fifth, the ratio 2:3; FlG - 6 - 

in the major third 4:5. Similar simple periods are found to recur 
in the other harmonious two-clangs ; the ratios of the constituent 
vibration- rates are in the case of the fourth 3 : 4, in the minoi 
third 5 : 6, in the sixth 3 : 5. Since all these periods of compound 
vibration-rates are repeated just as regularly as the periods 
of simple vibration-rates, we can understand how it is that a 
harmonious compound clang produces upon us as uniform an 
impression as a single tone. It is true that we distinguish in it 
the presence of more tones than one ; but these unite to form a 
total sensation, which runs its course with perfect evenness. 

But it is quite different when two tones are sounded together 
whose vibration-rates do not stand in any simple and harmonious 
ratio, but bear a more complex relation to each other. In such 
a case there can be no production of the uniform periods, 

70 Lectures on Human and Animal Psychology 

recurring at very brief intervals of time, which we have found in 
the harmonious compound clangs. As a result of this, the inter- 
action of the vibrations causes a disturbance in the uniform 
course of sensation. Wherever two movements in the same 
direction coincide, as at a and b (Fig. 7), they strengthen each 

other ; and where two 
coincident movements 
have an opposite direc- 
tion, as at ;;/, they 
weaken each other. It 
IG ' 7 ' depends, of course, on 

the difference of vibration-rate, how often these pendular to-and- 
fro movements of the air particles agree or disagree with each 
other. If one tone makes exactly one vibration more in i" 
than another, there will occur in each second one such increase 
and decrease. For if at the beginning of the second, at a, both 
vibrations start at the same stage, there will meet in the middle 
of the second, at m, a forward movement of one wave and a 
backward movement of the other, so that the two cancel one 
another ; while again at the end of the second, at b, they will be 
travelling in the same direction, and will, therefore, assist each 
other. It is clear that just the same will happen if the difference 
of the two tones is one of a greater number of vibrations ; there 
will be as many increases and decreases, as many beats, as there 
are more vibrations in the one case than in the other. If the 
difference is very small, amounting, e.g., to one vibration in the 
course of several seconds, it will be scarcely noticed, the 
diminuendo and crescendo of the tone occurring continuously and 
gradually. If the change is spread over a sufficiently long space 
of time, it will not be perceived at all. But if one or more beats 
occur in i", they are clearly noticeable ; and if their number 
increases to 10 or more, their quick succession will be sensed 
as a very unpleasant whirring. 

The limit of rapidity at which the beats of dissonant tones 
may be perceived cannot be determined with any degree of 
certainty. For, in the first place, the beats, as they follow one 
another faster and faster, give rise to a general impression of 
harshness, more or less comparable to that which a rough 
surface produces in the sense of touch ; and when the rapidity 

Tone-Sensations : Beats 7 1 

becomes still greater, while the tones are not heard as a 
harmony, the beats and even the roughness of the clang also 
disappear. The extreme limit at which this harshness can still 
be distinguished appears to be reached in the neighbourhood of 
60 beats in i". 

Now these observations upon inharmonious compound clangs 
seem to imply a contradiction between the perception of the 
beats and the laws which we formulated above in terms of 
the vibration-rates of tones. For it is found that tones can still 
give rise to clearly perceptible beats when the difference between 
their vibration-rates amounts to considerably more than 60 in 
l". If we take, e.g., the two neighbouring tones c and d from the 
lower or middle region of the scale of pure temperament, and 
strike the clangs together, we shall obtain loud beats. This is 
perfectly intelligible from what has been said above. For if the 
tone c makes 128 vibrations in i", d, which is higher by the 
interval of a second, will make 9/8 x 128, or 144 vibrations 
The two tones must, therefore, give 16 beats in i". But if we 
strike with c, not the d. but the octave d ' we are giving a tone 

* * o o 

of 2 x 144, or 288 vibrations. Its difference from c amounts to 
160 vibrations. Yet although it is quite impossible to hear beats 
which follow one another as quickly as that, the compound 
clang is not merely inharmonious, but is also clearly accompanied 
by beats similar to, if not quite so strong as, those arising from 
the striking of two notes which are a single whole tone apart. 
What reason is there for the fact that the higher d' makes beats 
with c, while the tones c' or g, octave or fifth of the purely 
tempered scale, the differences of whose vibration-rates from 
that of c are smaller, give no noticeable beats at all ? The 
reason may be discovered from the following simple experiment. 
When we strike a piano or guitar-string that is stretched over 
the sounding-board, the result is, of course, a tone. If a bridge 
is placed in the middle of the string, so that only half of it can 
vibrate, the resulting tone rises, as we have said, an octave 
higher. By striking first the fundamental, and then the octave, 
we come to see that the latter was really contained in the former: 
that it sounded, though weakly, along with the fundamental. The 
case is the same if first the whole string, and then one-fourth 
of it, is struck. Here the double octave is seen to be sounded, 

~2 Lectures en Human and Animal Psychology 

though very weakly, with the fundamental, and so on. If we 
have trained our ear in the comparison of clangs, we are able to 
hear out these higher tones, the over-tones, from the fundamental 
It is found that every tone of our musical instruments and of the 
human voice contains a large number of over-tones ; so that 
strictly speaking, we never have the sensation -of a simple tone, 
but always that of several simultaneously sounding tones, one 
of which, however, the fundamental, is so much stronger than the 
rest that we usually fail to hear them. The phenomenon of over- 
tones finds its physical explanation in the fact that, in most 
forms of tonal stimulation, the wave-movement set up in the air 
is a compound one. When the string is struck, e.g., not only 
does it vibrate in its entire length, and so transmit the ground- 
tone to the air, but either half of it vibrates also, though not so 
violently, on its own account, and so produces the octave. In 
the same way each third and each fourth of the string vibrate ; 
thus giving rise to the third of the higher octave, the double 
octave, and so forth in decreasing series. These separate tones 
persist just as independently of one another as if several instru- 
ments were sounding at the same time. The only difference 
consists in the greater weakness of the over-tones. 

And now we are able to explain the very curious fact that the 
tone c beats not merely with the neighbouring d, but also with 
the d of the higher octave. Simultaneously with the funda- 
mental c there is given the octave c , and this beats with the d, 
which stands next to it. The beats are certainly not so pro- 
nounced as if the c had been directly sounded, partly because 
the over-tone is weaker, partly because the beats follow one 
another more quickly; but they are clear enough to be heard. 

This simultaneity of fundamental and over-tones is of im- 
portance not merely as throwing light on the consonance and 
dissonance of the tones in a compound impression, but also as 
influencing our apprehension of separate tones. The tones of a 
musical instrument and of the human voice are characterised 
not only by pitch, but also by a definite clang-character. If all 
tones depended simply upon the vibration-rate which determines 
their pitch, then, apart from the noises that may chance to 
accompany them, every tone of the same pitch would possess 
the same character, however it might be produced. This, of 

Clang-colour ; Simultaneous Clangs 73 

course, is not the case. The same tone sounds quite differently 
when given by flute, violin, clarionette, organ, etc. The vibra- 
tion-rates must have still other properties, varying for one and 
the same tone with the source of sound. As a matter of fact, 
we have found the over-tones regularly accompanying tones, and 
presenting differences which depend upon the mode of origin of 
clangs. There are tones in which scarcely any over-tones are 
noticeable. Those of the flute-pipes of an organ come very 
near to absolute purity, and those of a tuning-fork standing 
upon its resonance-box nearer still. If the resonance-space is 
exactly fitted to the primary tone of the tuning-fork, all the 
secondary tones are so weak in comparison that they are not 
heard as the tone rings off. On the other hand, wind and string 
instruments and the human voice always allow a large number 
of over-tones to be heard beside the ground-tone. As a general 
rule the intensity of the over-tones decreases with their height. 
The octave can be heard more clearly than the double octave, 
this than the third, etc. But there are considerable differences 
in the different instruments. Sometimes the higher octaves 
sound most strongly, as on the piano ; sometimes the higher 
fifths and thirds, as in the clarionette ; sometimes the first over- 
tones are heard at a comparatively uniform intensity, as on the 
harmonium ; and sometimes single very high over- tones are 
preferred, as in the trumpet and trombone. 

We have now inquired into all the conditions of the peculiar 
colouring of the different kinds of clangs. This depends partly 
upon the intensity of the over-tones in general, partly upon the 
character of the strongest among them. 


If the laws of concurrent vibrations which we have been 
discussing are true, no tone is ever entirely free from over-tones. 
Even though it should be so objectively, there would probably 
still attach to it subjectively some kind of clang-colour, due to 
the presence of very weak over-tones caused by the concurrent 
vibrations of certain parts of the auditory organ, attuned to the 
particular tone. 

The tone-sensation which, in virtue of its attendant partial 

74 Lectures on Human and Animal Psychology 

tones, is possessed of a definite clang-quality, we call a clang; 
the particular clang-quality produced by the over-tones, clang- 
colour. Every clang consists, therefore, of tone- pitch and clang- 
colour, the latter component being in its turn made up of s 
number of weaker tone-sensations accompanying the primary 
tone. The clang, that is, is a compound sensation, and since all 
tones are in reality clangs, our tone-sensations are never given 
in any other than a compound form. We can separate out the 
individual simple pitches only by either subjectively abstracting 
from the attendant secondary tones in the clang, or strengthen- 
ing the primary tone to such an extent that they disappear, as 
happens when a tuning-fork vibrates on its resonance-box. But 
even if a tone is comparatively rich in secondary tones, we 
apprehend it in idea as perfectly unitary and relatively simple, 
as is shown by the fact that we ascribe to it only a single pitch. 
Over-tones, on the other hand, even though they are strong 
enough to be clearly perceived, are not apprehended as separate 
pitches, but appear merely as a peculiar modification of the 
principal tone. It is plain that this cannot be explained simply 
in terms of the lesser intensity of the secondary tones. But it 
becomes intelligible when we consider that wherever definite 
simple sensations are given in constant connections these 
connections blend to form unitary ideas ; and that when it is a 
tone that is connected with harmonious secondary tones this 
blending process must be very materially furthered by the possi- 
bility of the co-existence, without mutual disturbance, of har- 
monious vibration-rates. Regarded from this point of view, the 
clang-idea presents to us a simple and typical example of a 
psychological process which we shall frequently meet with, for 
the most part in a more complex form, the process of sensation- 
fusion. All the elementary constituents in this fusion-process 
have lost the character which they possessed in their isolated 
condition ; in the stable connections into which they have 
entered they are determined by the character of the other 
elements present. Thus the octave c of a tone c, when it appears 
in the first over-tone of the latter, is something entirely different 
from what it is when sensed alone. In the latter case it would 
be an independent tone ; in the former it is perceived directly in 
its relation to the simultaneously given principal tone, and, since 

Clang-colour ; Simultaneous Clangs 75 

this is much the stronger, appears as a mere modification of its 

The compound clang is distinguished from the simple clang 
only by the number and relative intensity of the tones which 
enter into it. If we strike the chord c e g, we are sounding three 
tones which form part of the series of over-tones belonging to a 
lower C. The major third c e corresponds to the proportion of 
vibration-rates 4:5; the fifth c g y to 2 : 3 or 4 : 6. That is, the 
three tones occupy the fourth, fifth, and sixth places in the 
complete tonal series of a simple clang: I, 2, 3, 4, 5, 6, . . 
But while in a single clang these tones appear only as secondary 
tones of a lower fundamental, whose clang-colour they determine, 
in the second they constitute the chief elements in the whole 
impression, and are of equal intensity. In the compound clang, 
therefore, we sense at once a plurality of tones. Since compound 
clangs are only harmonious if the ratio of their vibration-rates 
is one of simple whole numbers, they may be regarded in every 
case as intensifications of the separate members of the tonal 
series of a single clang. 

But there is still another element in compound clangs. It 
consists in the appearance of lower tones, which are in harmony 
with the principal tones, and play a part in the determination of 
the character of a compound clang similar to that of the over- 
tones in the clang-colour of a single clang. 

Whenever there are set up simultaneously two harmonious 
sound-waves, whose vibrations strengthen and weaken one 
another alternately, at short and uniformly recurring intervals, 
there arises from this interaction a new tone, the vibration-rate 
of which corresponds to the difference between the vibration- 
rates of the two original tones. Look for a moment at the 
second pair of curves in Fig. 6, representing the fifth, eg. During 
two vibrations of the first and three of the second tone, there is 
one coincidence of hill and valley and one each of hill with hill 
and valley with valley. So there is set up a third wave- move- 
ment, which makes one vibration for every two of the first or 
three of the second tone. Such tones, which may be called 
either under-tones from their relation to the over-tones, or 
difference-tones, from the relation of their vibration-rates to those 
of the original tones, may be intensified in complex chords by 

76 Lectures on Human and Animal Psychology 

the fact of several of them falling upon the same note. In the 
chord c eg, for example, in which the vibration-rates of the tones 
stand in the ratio 4:5:6, both c and e, and e and g, produce 
the same under-tone i, a c lying two octaves below the lowest 
tone of the chord, while c and g give a tone 2, i.e., a c only one 
octave below it. 1 

To these harmonious under-tones, which we have found 
accompanying the compound clang, must always be added the 
over-tones of the single clangs. They may also strengthen one 
another in certain cases, where different clangs have similar 
terms in their series of partial tones. So that every chord, even 
the comparatively simple, is made up of a very large number of 
sensation-elements, some of which, the more intensive primary 
tones, stand out as clearly distinguishable qualities, while the 
others merely serve to determine the clang-character of the 
chord. The triple clang c e g, for instance, gives us the 
following tones : 


Principal tones. 


C, C 

I 2 

c e g 
4 5 6 

c l e 1 6 l _ d* 
8 10 12 15 16 18 

L gL 

20 24 

The first over-tones are usually the strongest ; only these are 
entered in the schema. Under-tones which appear as difference- 
tones of more than one two-clang, and over-tones which belong 
to more than one single clang, are underlined. The difference- 

1 When harmonious tones are simultaneously sounded, we have formed 
not only difference-tones, but a second kind of resultant tonal wave, depend- 
ing upon the fact that the hills and valleys of the primary waves are not 
perfectly coincident. The vibration-rate of these new tone-waves is the sum 
of the vibration-rates of the original tones. The tones themselves are, 
therefore, of higher pitch than the principal tones in the chord, and are 
termed, from their mode of origin, summation-font r. Thus the fifth, 2 : 3, 
has a summation-tone of the vibration-rate 2 + 3 = 5. Difference- and 
summation-tones together are sometimes called combination- tones. However, 
the interpretation of the summation-tones is not beyond doubt, many 
psychologists regarding them as high over-tones of the principal clangs. In 
any event, they are so weak as to exert no influence upon the clang-character 
of the chord unless they coincide with over-tones. We may, therefore, 
'leave them here out of account. 

Noises 77- 

tones which the over-tones form with one another, or with the 
principal tones, are not set down. In most cases they are so> 
weak as to be sensed with difficulty, or not at all. You see 
that, even in a completely harmonious chord, the over-tones of 
the second octave stand so near together as to produce very 
considerable dissonance. Indeed, the most perfectly attuned 
chords of an instrument whose clangs are rich in over-tones 
(organ, harmonium) allow the beats of these over-tones to be 
clearly perceived. They combine with the quality of the under 
and over-tones to determine the general character of the 
different chords. 


The compound clang arises irom the single clang by the 
strengthening of secondary tones to principal tones. The 
compound clang in its turn may pass over into the third 
general sound-quality, into noise, as soon as the dissonant 
elements, which we have found to be not altogether absent everu 
in harmonious chords, multiply to such an extent that har- 
monious tonal ratios cannot be any longer perceived. You 
may easily convince yourselves of the origin of noise from the 
compound clang by striking simultaneously upon some instru- 
ment of wide range, such as piano or harmonium, a large 
number of inharmonious tones. The separate tones make such 
strong beats with one another that the resultant sensation tends 
to lose its clang-character altogether. 

But when we seek to determine the point at which the clang 
ends and the noise begins, we find that there is no sharp line of 
division discoverable. In most noises we can distinguish one 
or more deep tones, but these are accompanied by a crowd of 
indistinguishable secondary tones, strong or weak, and of the 
most different pitch. That is, the difference between clang and 
noise is only one of degree. Noise and clang alike depend 
upon the simultaneity of several tone-sensations. Even in the 
clang, it is impossible to distinguish and identify the greater 
part of these tone-sensations ; they merely serve to colour the 
principal tone in a particular way, and it requires a sensitive ear 
and close attention, or special experimental aids, to refer the 
effect to its true cause. And the tact that the clang-colour 

78 Lectures on Human and Animal Psychology 

depends upon the occurrence of over-tones is further obscured 
by the presence of the principal tone itself. With noise the 
conditions are directly contrary : it is the mixture of tones 
which plays the principal part, and the separate tone, in conse- 
quence, tends entirely to disappear. 

It is, however, probable that this, though the customary, is 
not the only, mode of origin of noise. There is another, which 
sometimes co-operates with the first, sometimes appears alone. 
A vibration-rate the rapidity of which is high enough to 
transcend the upper limit of tonal sensation is perceived as a 
hissing noise, while very slow vibration-rates, which do not 
reach the lower limit of tone, give rise to a roaring noise. It is 
supposed that these sensations are caused not by the excitation 
of the cochlear apparatus which is attuned to tone, but by the 
vibrations of more simple organs, connected with the fibres of 
the auditory nerve, and situated in the vestibule of the labyrinth 
of the ear. Since the vestibule belongs to a much earlier stage 
of development than the cochlea, we might interpret these 
simple, absolutely toneless sensations as more primitive than 
clang-sensations, and as constituting the whole series of sourid- 
sensation for most of the lower animals. When once the tone- 
sensation has been developed, however, these toneless sensations 
are completely overshadowed, even in perceptions of noise, by 
the dissonant clang-constituents which enter, as we have seen, 
into the maiority of noises. 


If we abstract from these elementary noise-sensations, which 
are probably of high importance for the development of auditory 
sensation in the animal kingdom, but which play so small a 
part when once audition has been perfected, we may say that 
all kinds of auditory sensation, clangs, compound clangs, and 
noises, are combinations of simple tone-sensations. The 
simple tone-sensation itself, however, cannot be analysed into 
still more simple constituents : like every simple sensation, it 
possesses only the two attributes of intensity and quality, i.e., 
intensity and pitch. Pitch, like intensity, can vary only in two 
opposite directions, up and down ; we can pass from a given 

Measurement of Tonal Differences 79 

tone either to a higher or to a lower one, just as from any 
given point on a straight line we can proceed only in two 
directions and keep within the line. That is to say, our whole 
system of tone-sensations may be considered as a plurality in 
one dimension, or as a linear plurality. 

The analogy between the quality of tone-sensations and the 
intensity of any particular pitch holds in yet another connection. 
Intensity varies step for step with variations in the strength of 
the external stimulus. In the same way, tone-quality follows 
step for step variations in vibration-rate ; and we are as little 
able to appreciate any the least alteration in vibration-rate as a 
change in sensation-quality as we are to notice minute altera- 
tions in strength of stimulus as changes of sensation-intensity. 
In both cases there is a lower limit of discrimination. It is, of 
course, not possible to determine this limit on ordinary musical 
instruments with fixed note-values, like the piano, because the 
tones are separated by far more than just noticeable intervals. 
But if we take two similarly tuned strings or tuning-forks and 
gradually alter the pitch of one of them, we have no difficulty 
in discovering the point at which one tone sounds just notice- 
ably deeper than the other. It is necessary, in making this 
experiment, to strike the strings or forks successively, and to 
check the vibrations of one before the other is sounded, since 
otherwise beats would be produced, and the qualitative differ- 
ence of the tones known from them, and not from the differences 
in sensation. On the other hand, when the point at which a 
sensation-difference becomes noticeable is once found, we may 
employ the beats produced by the simultaneous sounding of 
the tones to tell us the objective difference of vibration-rate, 
corresponding to the bare possibility of their qualitative dis- 
crimination in sensation. The number of beats, you will 
remember, corresponds exactly to the difference between the 
vibration-rates of the two sounding bodies For instance, if 
two similar tuning-forks have been so far removed from unison, 
in the manner described above, that their successive tones are 
just distinguishable as different pitches, and if we find that they 
produce two beats in 10" when struck simultaneously, we may 
conclude that at their particular pitch the just noticeable 
sensation-difference is represented by a difference of 0*2 vibra- 

8o Lectures on Human and Animal Psychology 

tion in i". Careful experimentation on these lines has shown 
that the differences of pitch, which are just noticeable for succes- 
sive stimuli, remain absolutely constant over a large part of the 
musical scale. Between the limits of 200 and 1,000 vibrations 
in the i", we can sense a tone-difference represented by O'2 
vibration. For lower tones the fraction is somewhat smaller, 
for higher tones correspondingly larger ; but within the sphere 
of musical applicability there is no considerable deviation from 
this average value. Where the tones are high or low enough 
to approach the limit of sensibility, discrimination becomes, as 
we should expect, far less certain. You may convince your- 
selves of this by striking successive notes at either end of the 
key-board of a piano: differences of a whole half-tone are 
hardly perceptible. 

Let us apply the general conclusions which we arrived at in 
considering the question of the measurement of sensation-inten- 
sity to this particular case. The result of our observations may 
be summed up in a single sentence. Within wide limits, we have 
found, equal differences of tonal quality correspond to equal differ- 
ences of vibration-rate ; in other words, the sensation of pitch 
varies in direct proportion to the objective variation of tonal 
vibration. And there is another path which will lead us to the 
same point. We possess an especial capacity for the quantita- 
tive comparison of more than just noticeable differences of tone- 
sensation. Suppose that we strike first the two tones c and d in 
succession, and then the d and a of the same octave. Even the 
most unmusical person, who has no notion of the intervals be- 
tween the tones in the technical sense, is perfectly certain that 
a and d are further apart than c and d. Here, then, is another 
experimental method. Two tones at any distance from one 
another on the musical scale may be given in succession, and 
the observer required to estimate their exact sensation-mean, 
the pitch which is just as far removed from the first as it is from 
the second. It has been found that the tone selected as the 
mean is always approximately the tone whose vibration-rate 
lies midway between the vibration-rates of the two extremes. 

But there are facts, recurring in the musical experience of all 
times, which stand in apparent contradiction to these experi- 
mental observations. These facts are expressed in the tonai 

Tonal Scale 81 

relations of the musical scale. We have seen that the vibration- 
rate of the octave is invariably twice that of the ground-tone, 
that of the fifth 3/2 of it, that of the major third 5/4, and so on. 
The octave of the tone of 32 vibrations in i" makes 64 ; the 
octave of this, 128, etc. That is to say, the higher the pitch, the 
greater becomes the difference between the vibration-rates con- 
stituting any particular interval. Nevertheless, the sensation- 
difference between a tone and its octave appears to remain the 
same from whatever region of the scale the interval is taken : the 
difference in pitch seems unchanged whether we compare the tone 
of 32 vibrations with that of 64, or the tone of 64 with that of 128. 

The law regulating musical intervals, therefore, lays it down 
not that pitch varies in direct proportion to variation of stimulus, 
but that it varies more slowly than stimulus. And this law of 
slower variation is, again, a very simple one. To increase tonal 
qualities by equal increments, we must increase the rapidity of 
their vibration-rates by a magnitude which always bears the 
same relation to the rate to which it is added. To obtain the 
octave of a given tone, its vibration-rate must always be multi- 
plied by two ; to obtain the fifth, third, and fourth, its original 
rate multiplied respectively by 3/2, 5/4, 4/3. And this result is 
precisely parallel to that which we obtained when we were dis- 
cussing the pressure of weights, the strength of sound and light, 
in short, the intensity of sensation in general. In all depart- 
ments of quantitative comparison we found that for the sensa- 
tion to increase by absolutely equal magnitudes the external 
stimulus must be increased by relatively equal magnitudes. We 
have, then, only to write 'tonal quality' for 'sensation-inten- 
sity,' and we have our law, the same law which held for the 
general relation of stimulus and sensation in the sphere of in- 
tensity. If pitch is to increase by absolutely equal magnitudes, 
vibration-rate must be increased by relatively equal magnitudes ; 
or, more briefly,///^ increases in direct proportion to the logarithm 
of vibration-rate. 

Here, then, we are met by a very curious contradiction. 
According to the law which regulates the musical scale, the 
sensation of pitch is dependent upon stimulus in just the same 
way as sensation-intensity is dependent upon it. But so soon as 
we apply the methods which we used to measure intensity to 


82 Lectures on Human and Animal Psychology 

the measurement of quality in this particular case, we find, 
within certain limits, a direct proportionality between variation 
of sensation and alteration of vibration-rate. 

The contradiction is only apparent. The most obvious way 
to remove it, perhaps, would be to point out that sensation- 
intensity and tone-pitch are different things. If Weber's law is 
proved to hold of one of them, there is not the least reason to 
assume that it will hold of the other. How were the musical 
intervals established in the first place ? Not, at any rate, with 
conscious reference to the fact that the same interval presents 
the same difference for sensation from whatever region of the 
tone-scale it is taken. We must look for other conditions, con- 
ditions of our tonal sensibility in general, which give a definite 
character to each tone-interval, quite apart from its position upon 
the whole musical scale. They are not far to seek ; think of the 
over-tones which accompany every simple clang. When a pitch 
changes by the amount of some particular musical interval, the 
character given to the clang by its over-tones must change in 
the same way. Suppose that the change is that of a fifth. The 
vibration-rates of the principal tones bear the ratio 2 : 3. The 
clang-character of the lower one is determined by the over-tone 
series : 4, 6, 8, IO, 12 . . . ; that of the higher by the series : 6, 9, 
12, 15. . . . The relations of these two series remain the same 
whatever the absolute pitch of the principal tones may be. 

At the same time, the explanation is not satisfactory. Granted 
that this constancy of the relations of the secondary tones in 
every interval gives a reason for the dependency of the musical 
intervals upon constant relations of their constituent vibration- 
rates, still the problem is only pushed one step farther back, not 
solved. If we are to know that an interval is the same when it 
is given, say, first in the upper and then in the lower region of 
the scale, we must apprehend in sensation the likeness of the 
interrelations of all the partial tones in the two cases. But what 
is true of secondary tones will be true of their primaries. As a 
matter of fact, we are able to cognise the harmonic intervals of 
pure tones, which are practically free from over-tones, with 
almost as much accuracy and certainty as the intervals of clangs 
whose over-tones are numerous and intensive. That is to say, 
though our apprehension of harmonic intervals may be furthered 

Weber s Law and Tones 83 

by the compound character of the single clang, there must be 
operative some more ultimate influence than this, which we have 
not yet found. To which must be added that the contradiction 
resulting from the application of the customary methods of sen- 
sation-measurement to the intensity of sensation on the one 
hand, and to the pitch of tones on the other, is not in the least 
degree removed by the adoption of the proposed explanation. 
The answer that intensity and quality are two different things 
simply gives up any attempt at an adequate explanation of the 
incongruity of the two sets of results. As a matter of fact, there 
is a complete parallelism between the continuous gradations of 
stimulus-intensity and vibration-rate, of sensation-intensity and 
tonal pitch. 

But there is one way of escape still open to us. Recall to 
mind the psychological interpretation which we offered of 
Weber's law. We explained the law by assuming that in esti- 
mating differences of sensation it is the relative, and not the 
absolute magnitude of the compared sensations which we re- 
gard. But there is always the possibility of an absolute estima- 
tion by the side of this relative one. And we shall expect to 
find the possibility realised in all cases where a sensation, for 
some reason or another, is apprehended by itself, in isolation from 
the remaining terms of the sensation-series to which it belongs. 
That, again, will occur, and only occur, when the effect of the 
sensations upon consciousness is not such as to necessitate a 
reference to other sensations of the same kind. Now this refer- 
ence is inevitable in every apprehension of an intensity. A loud 
sound makes greater demands upon consciousness, so to speak, 
than a weak one. For a sensation-increment to possess the same 
magnitude in the two cases, therefore, the increase of the 
stronger stimulus must be greater in proportion as the stimulus 
itself and, consequently, its effect upon consciousness is greater. 
But with tone-pitch the case is quite different. A high and a 
low tone may well be upon a perfect equality as regards the in- 
tensity of their effect upon consciousness. That is, the criterion 
of our discrimination of two tones whose qualitative difference is 
just noticeable can only be their absolute difference in sensation, 
which is paralleled by the absolute difference between their 
vibration-rates. And a comparison of this kind will be possible 

84 Lectures on Hitman and Animal Psychology 

for tones whose difference is more than just noticeable ; so that 
in subdividing such a total difference into two equal smaller 
differences we shall always have in mind an absolute, and not a 
relative, standard of measurement. It is, of course, still another 
case, if we propose to find two tones at one part of the scale, the 
relation of whose qualities is similar to that of the qualities of 
two given tones from another part. In this case the difference 
is stated in the formulation of the problem to be a relative one ; 
and the interval is chosen with a view to this relativity. Coinci- 
dence of over-tones will certainly help us in this case to cognise 
the likeness of the two intervals compared ; all that we insist on 
is, that it is not the sole determinant of our estimation. It is 
true that the repetition of the fifth, eg, in a higher octave 
allows of a readier cognition of the particular interval than the 
giving of the tones da or fc. But, nevertheless, the likeness 
of these two intervals to the first does not remain doubtful for 
a single moment. 

Our views as to the significance of Weber's law are, then, 
partly confirmed and partly supplemented by the facts which 
we have learned as to our apprehension of differences of 
tonal quality and of tone-intervals. Partly confirmed : for we 
have found still further reason for our supposition that Weber's 
law is to be interpreted as a law of the relative estimation of sen- 
sation-magnitude. Tone-intervals furnish us with particularly 
convincing evidence of the truth of this law of relativity. Partly 
supplemented ; for we have found that where the conditions of 
our apprehension of different sensations suggest an absolute 
rather than a relative comparison, a simple proportionality takes 
the place of the logarithmic relation between stimulus and 
sensation. This fact serves at the same time to disprove once 
and for all the psychophysical theory of Weber's law, which 
saw in it an expression of the universally valid relations obtain- 
ing between the psychical and the physical. Such a hypothesis 
could only be maintained, if the sensation itself, apart from the 
psychological processes involved in its comparative apprehen- 
sion, were subject to the logarithmic law. Nor is the physio- 
logical theory, that is, in its customary form, less clearly dis- 
proved. It supposes that the conduction of the sense-excitation 
in the brain meets with obstacles, which increase as the magni- 

Weber 's Law and Tones 85 

tude of the stimulus increases, so that the excitation in the 
central organ itself increases more slowly than the external 
sense-stimulus, the exact amount of its inhibition being ex- 
pressed in the logarithmic formula. The fact that, under con- 
ditions which exclude the influence of comparison in estimation, 
the course of sensation and stimulus is within limits a perfectly 
parallel one, makes against this supposition. It must rather be 
the case that within these limits there is a direct proportion- 
ality between central and peripheral excitation. So that if we 
are led by the principle of psychophysical parallelism to look 
for a physiological basis, as well as for a psychological explana- 
tion of Weber's law, our field of search must be the relations of 
the stimulation-processes in some sensory centre of a higher 
order, where are aroused the physical excitations which underlie 
a relative apprehension of sensations. 

Our general conclusion, therefore, will be this : wherever we 
are able to bring about a continuous alteration in the intensity 
or quality of stimulus and sensation, we shall find certain limits 
within which the alteration of sensation runs directly parallel to 
the alteration of stimulus. On the other hand, when we are 
comparing different sensations with one another, we shall expect 
to find our estimation of their magnitude determined as absolute 
or relative by the special conditions of the investigation. A 
just noticeable difference in the intensity of sensation is always 
apprehended in relative terms ; for the amount of increase which 
is equally noticeable in different cases depends upon the de- 
mands made upon consciousness by the particular sensation. 
The greater the intensity of the sensation, of course, the greater 
its effect in and upon consciousness. Our estimation of tone- 
intervals is also relative : it is the relation of the terms, not 
their absolute value, upon which we have to direct our attention. 
Nevertheless, it is not difficult to perceive that the same interval 
represents a greater absolute sensation-difference in the upper 
region of the tonal scale than in the lower, unless the tones 
which compose it are so very low or high that our discrimination 
fails us. If we strike first the tone c followed by g, and then 
g followed by d', if we give, i.e., two opposed fifths in succes- 
sion, and concentrate the attention exclusively upon the absolute 
sensation-difference, we have no hesitation in declaring the 

86 Lectures on Human and Animal Psychology 

distance^/' to be greater than eg. And this helps to explain 
the converse fact : that when we are halving more than notice- 
able tonal differences simply in terms of their absolute value in 
sensation, and without regard to them as musical or unmusical 
intervals, or when we are determining just noticeable differ- 
ences of tonal quality, we estimate absolutely, and not relatively. 
So that if we are to sum up the result of this whole discussion 
in one general proposition, that will run somewhat as follows : 
Unless a sensation approaches the upper or lower limit of sensi- 
bility, alteration in it is directly proportional to the absolute magni- 
tude of alteration of its stimulus. But our apprehension of 
alteration in sensation remains only relative so long as its absolute 
perception is not made possible by the express introduction of 
especial conditions. 






VISUAL sensations have two qualities, colour and bright- 
ness. The latter term includes black, white, and all the 
various shades of grey. The number of distinguishable colours 
is enormous : it has never been determined. But it is a fact 
of immediate perception that the variety of colour in nature is 
not altogether heterogeneous : there are very many intermediate 
tints between this colour and that. And if we try to divide up 
the whole multifarious colour field, and to separate out the colours 
which are clearly and definitely unlike the rest, we can reduce the 
list of ' pure ' colours to very small dimensions. Red, yellow, 
green, blue, together with black and white, prove to be the 
simple and ultimate qualities which we are able, so to say, to 
abstract from the innumerable specifically different presenta- 
tions of nature. All other distinguishable colour-tones are 
intermediates, a fact which is very frequently expressed in the 
names given to them (purplish red, orange-yellow, yellowish 
green, violet-blue, etc.). But these six simple qualities, again, 
are not co-ordinates ; they evince different degrees of variety or 
resemblance. We are inclined to regard green as being nearer 
blue than yellow is, and to look on red and yellow as pretty 
closely related colours, even when the intermediates blue-green 
and orange are not present to suggest the comparison. It might 
be thought that this notion of colour relationship was due to our 
knowledge of colour sequence in the rainbow. But children 
who have never observed a rainbow with any degree of attention 
will usually connect blue with green, and red with yellow, when 
required to arrange the four qualities in the order of likeness. 


88 Lectures on Human and Animal Psychology 

The hypothesis that there is a limited number of simple 
colours, from which we may conceive all the other qualities of 
light to be compounded, is thus suggested by the subjective 
relationship obtaining between certain colour-tones. It is fur- 
ther confirmed by the familiar results of the mixture of pig- 
ments. The painter has long known that green can be obtained 
by mixing yellow and blue, violet and purple by mixing blue 
and red. It was an obvious corollary that every colour which 
could be produced in this way by the intermixture of other 
colours was in itself a complex, not a simple, sensation-quality. 
No distinction was drawn, you see, between objective light and 
subjective light-sensation ; if the external light is a complex, it 
was thought, the sensation corresponding to it must be a complex 
also. And even to-day the painter is wont to look upon red, 
blue, and yellow, together with black and white, as the simple 
qualities by whose intermixture all other colours are produced. 

The science of colour went still farther. Colours usually differ 
not only in tone, but in brightness ; red appears darker than 
yellow, etc. So it was thought possible to arrange all the colours 
in a series, the terminal members of which should be constituted 
by the two extremes of brightness, black and white. Aristotle, 
for instance, taught that black and white are the two fundamental 
qualities of light, and that every colour can be obtained from 
their intermixture in varying amounts. 

From the point of view of direct perception, the simplicity 
and universality of this hypothesis are very tempting. When 
we have once convinced ourselves that the great majority of 
colours in nature result from the intermixture of a small number 
of simple qualities, and that these themselves are sensibly related 
to one another, our mind will not be at rest till it has reduced all 
the phenomena to two polar opposites. And these can be nothing 
else than black and white. For all the true colours stand some- 
where between these two in brightness, approaching white if 
their brightness is increased and tending towards black if it 
is diminished. If all colour is to be derived from two opposites, 
those opposites must be black and white. 

The Aristotelian view of the origin of colour prevailed, then, 
until modern times. Goethe defended it, and many of his 
admirers are its enthusiastic champions. But it has been ban- 

Li ght -sensations 89 

ished from science these two hundred years, thanks to Newton's 
discoveries. Newton said to himself : If there really are simple 
kinds of light or simple colours, which intermix in various ways, 
we must be able both to isolate and to recombine the simple 
constituents of any given compound colour. That meant that 
the whole question was referred to the tribunal of experiment, 
where alone it could be definitely answered. For direct percep- 
tion is deceptive. Can the chemist ' see ' of what elements a body 
is composed ? Of course not. We know that bodies of very 
different chemical composition appear just alike. May not the 
same hold of light ? May not similar kinds of light give rise to 
different mixtures, and different kinds to similar mixtures ? So 
Newton looked round him for a means of analysing compound 
light, and by a happy accident found what he wanted in the 
refraction of light by the prism. 

If we allow a ray 
of light coming from 
a to pass through a 
prism, /, of glass or 
some other trans- 
parent substance, its 
course is not the 
straight line that it 


would be if no prism 


were in its way, but 
it is turned aside, refracted, as we say : so that an eye, o, behind 
the prism receives it as if coming from b or thereabouts ; the 
source of light is transposed from a to b. Moreover, the point 
from which the ray appears to come in the direction bd does not 
always remain the same. It varies with the quality of the light. 
If, e.g., a is light of a blue colour, and the ray is seen as though 
it came from b, a red a will emit a ray which appears to travel in 
the direction re, r lying higher than b and nearer to a. It follows 
that different kinds of light are not refracted in the same degree 
by the same prism and under similar experimental conditions. 
Red light is less strongly refracted than blue ; r is nearer a than 
b is. On comparing the different colours with one another, we 
find that they fall into a definite series in regard to refrangibility. 
.Red is least, violet most, refracted ; and the series runs red, 

90 Lectures on Human and Animal Psychology 

yellow, green, blue, violet. Tones interpolated between two- 
neighbouring colours possess an intermediate degree of refrangi- 
bility. Orange lies between red and yellow, greenish yellow 
between yellow and green, indigo-blue between blue and violet. 

How is it with regard to white ? White is, of course, the 
most widely diffused quality of light ; it is that of sunlight. It is 
the light which we ordinarily see, if its character is not modified by 
the peculiar colour of an object. A ray of white light sent through 
a prism is affected in this way : the eye that receives it after its 
passage finds it not white at all, but distributed into a whole 
number of colours ; so that if a is a point of white light, the ray 
proceeding from it is not refracted simply like a ray of mono- 
chromatic light, and its source transposed from a to r or b, but 
it seems to issue from a series of sources arranged in a vertical 
line, each showing a different colour. Violet stands at the 
bottom ; and then follow blue, green, yellow, and red. White 
sunlight is, therefore, not simple, but capable of analysis into a 
large number of simpler light-qualities. These, on the other 
hand, are not further decomposable. However often we pass 
pure red or pure yellow through a prism, it retains its character 
unaltered. You notice that the colour-series obtained by the 
refraction of white sunlight, whether experimentally or naturally 
the rainbow is caused by refraction in the particles of water 
suspended in the atmosphere, contains all the colours which 
occur in nature. By mixing its tones in the right proportions, 
we can produce any colour that we wish. This is really self- 
evident, since all the light that the earth receives is derived 
from the sun. So that, whether a natural body reflects or 
absorbs light, no effect can arise whose cause is not contained 
in the constitution of sunlight. As the intensity of white light 
decreases, we gradually arrive at darkness, or black. Black, 
that is, is not a colour, but the minimal degree of brightness of 
white light. 

The facts obtained by this exact analysis of light were, how- 
ever, difficult to reconcile with the results of the mixture of 
colours, which had also been reached by way of observation. 
The spectrum produced by the analysis of white sunlight has r 
you see, at least five colours, and still more if we count in the 
intermediate tints. But painters had long since noticed that 


all possible varieties of colour could be produced from three 
simple tones. It is true that the resulting mixtures are not so 
saturated as the spectral colours ; but still they are as saturated 
as most of the colours occurring in nature. The three colours, 
primary colours, they were called, which could be so inter- 
mixed as to give rise to any other colour-variety, were generally 
given, as we have said above, as red, yellow, and blue. But it 
is better to take red, green, and violet ; and it is better, instead 
of mixing pigments, to mix directly colours that have been 
separated out from sunlight by the prism, or to allow colour- 
impressions to follow one another so quickly that they fuse for 
sensation. The colours to be mixed can be painted upon 
the sectors of a round disc, and the disc rapidly rotated on a 
top or by clock-work. This gives us a perfectly uniform im- 
pression. Red, green, and violet employed in 
their right amounts produce white ; and every 
distinguishable colour-tone corresponds to 
some particular mixture of the three funda- 
mental colours with one another and with 
white. White can also be obtained by the 
mixture of two colours situated at the right 
distance from each other in the spectral series. FlG - 9- 

The constituents of such pairs as, taken together, give rise to- 
white, are called complementary colours. Green-blue, e.g., is 
complementary to red, blue to orange, indigo-blue to yellow. 
Green is the only pure spectral colour that has no comple- 
mentary. To produce white it must be mixed with purplish 
red, a combination of red and violet. That, of course, is 
equivalent to a mixture of the three primary colours. 

How is this contradiction between the analysis and syn- 
thesis of light to be got over? It is generally left unchallenged, 
as it was by Newton himself. He said : There are combined in 
white light particles of red, yellow, green, blue, and violet light, 
and the prism isolates each separate ray ; but when we are 
putting together particles of different kinds of light, three of 
them, red, yellow, and blue, are enough to produce all the 
phenomena. Analysis, you see, had come into conflict with 
synthesis ; and physical science was not sufficiently advanced to- 
set them at one again. 

92 Lectures on Human and Animal Psychology 

The first step towards reconciliation was taken when it was 
discovered that Newton's theory of light was incorrect. He 
held that the particles of light were themselves coloured, and 
that light was a substance continually emitted by the sun, and 
containing in it a multitude of particles of the most diverse 
colours. This view had often had objections urged against it ; 
but it was reserved for the French physicist Fresnel to adduce 
a direct disproof of it by experiment. Fresnel showed that 
when light meets light it is by no manner of means necessary 
that an increase of intensity should result. Were light a sub- 
stance, that must be the case. But, as a matter of fact, decrease 
of intensity is as common a phenomenon as increase. These 
observations of what is known as the ' interference ' of rays of 
light prove indisputably that light is not a substance, but a 
movement. Two intercrossing movements may result either in 
an increase or decrease of intensity : nothing else can. If two 
'balls travelling in opposite directions with equal force meet 
each other, their movement is annihilated ; if they are travelling 
in the same direction, it is accelerated. If two waves of water 
meet, there is increase of the wave where crest and crest come 
together, decrease or annihilation where crest meets valley. 
The phenomena of interference show that there are wave-crests 
and wave-valleys concerned when rays of light meet one 
another ; that light-intensities will increase at one point of 
junction, while they decrease at another ; in other words, that 
we have to regard light as a movement analogous to the move- 
ment of a wave in water. If you throw a stone into water, you 
start a wave, which extends in every direction. The shock of 
the stone gives rise to a vibration which is transmitted from one 
particle of the liquid to another. Light consists of vibrations 
of this kind, except that the substance in which they are set 
up is infinitely less dense than water. It is a form of matter 
which interpenetrates all physical bodies, solids and liquids 
as well as gases, besides filling the space between them. The 
particles of this ' luminous ether ' are set in vibration in the 
fiery atmosphere of the sun, and the movement is transmitted 
from particle to particle at the enormous velocity of 186,000 
miles in the second. What our eye senses as an impression of 
3ight, therefore, is not a substance penetrating to it from the 

Light- sensations 93- 

remote depths of space, but a movement which, to excite our 
sense-organ, must be continuous through all the vast distance 
which separates us from its place of origin. It is one and the 
same form of matter which occasions all the multifarious sensa- 
tions of light and colour ; so that difference in sensation can. 
only mean difference in the movement of the luminous ether. 
Accurate measurements of the effects of interference have en- 
abled the physicist to determine this difference in different cases, 
and it has been found that colour-differences depend upon 
differences in the velocity of the oscillations of the particles of 
the luminous ether. In red light, the number of oscillations is- 
something between 400 and 500 billions in i" ; in violet, it ap- 
proaches 800 billions. All the other colours lie between these 
extremes. Orange has 500, green 600, blue 650, and indigo- 
blue 700 billions in i" ; so that the spectral colours constitute 
a progressive series within whose limits the velocity of vibration* 
increases by nearly 400 billions. It is worth noticing that sun- 
light contains, besides these coloured rays, other vibrations 
which are invisible, not. sensed as light. There are rays less 
strongly refracted than the red, and rays more strongly refracted 
than the violet ; vibration-rates, i.e., both greater and less than 
those which the eye senses as light or colour. The invisible rays- 
beyond the red end of the spectrum manifest themselves as heat ; 
those beyond the violet, in certain forms of chemical action. 

It is, you see, only a comparatively narrow section of the 
ether-vibrations which has the power of exciting a retinal sensa- 
tion. The whole variety of colour-tone is included within 
these narrow limits, and a very slight change in vibration-rate 
suffices to produce a noticeable difference of colour-sensation. 

These brief remarks upon the physical nature of light will 
serve to show you that light and colour have no objective reality,. 
i.e., do not exist as light and colour outside of and around 
us, but that all the properties by which we discriminate light as 
such, and the various separate colours from one another, are 
within us, originating in our colour- and light-sensations. What 
we call light and colour is just our own sensation of light and 
colour. Outside of us there is no system of sensations, but only 
vibrations in the ether. And the proof that light and colour 
are subjective phenomena marks an important step in psycho- 

94 Lectures on Human and Animal Psychology 

logical as well as in physical knowledge. We now know that 
a complete explanation of the phenomena of light and colour 
cannot be based solely on a physical examination of light, but 
must also take account of the conditions under which we sense 
it. What we sense, once more, is not the ether-vibration, but 
the particular reaction of our eye and mind upon that vibration. 
Movements which are too fast or too slow to be perceived by 
the eye are evidently removed from the list of visual stimuli 
simply by reason of their velocity. But objectively they may 
be light, just as much as any other movements. 

So that, if we are attempting an explanation of the phenomena 
of light and colour, the result even of a purely physical investi- 
gation will be to refer us to the seeing subject. Now you 
remember that we had come across a contradiction. The 
analysis of light by the prism told us one thing ; the recompo- 
sition of the various light-qualities told us another. We can 
separate out from sunlight at least five simple colours, without 
counting intermediate tints, while we can produce every colour 
that occurs in nature by the intermixture in appropriate amounts 
of three colours only, best chosen as red, green, and violet. 
How is the contradiction to be met ? 

It is evident from what has been said above that the fact that 
white and all possible colours can be obtained from three 
primary colours does not mean that objective light is com- 
pounded of those three fundamental colours. Nor does it 
mean, as many physiologists would still have us believe it does, 
that all our subjective light-sensations are derived from three 
ultimate sensations corresponding to the three primary colours. 
All that the results of experiments upon colour-mixture 
prove is this : that three objectively simple modes of vibration 
are sufficient, when mixed in different amounts, to set up all 
those stimulation-processes in the organ of vision which can 
be occasioned by the colours of the solar spectrum and their 

The three primary colours could only possess any sig- 
nificance in physics if objective light were analysable into 
three modes of vibration, corresponding to them, and no more. 
We have seen that that is not the case. In one way the 
primary colours do occupy exceptional positions in the infinite 

Light-sensations 95 

gradation of light-vibrations. Red and violet stand at either 
end of the series of visible vibration-rates, and green is situated 
at its centre. But this, though a fact of some importance for an 
examination of the conditions under which the eye is stimulable 
by light, has no reference whatever to objective light as such, 
the light whose vibration-rates extend far beyond the limits 
within which the ether rays are visible. 

The primary colours would only possess significance in 
psychology if we could subjectively, in immediate sensation, 
analyse all our light-sensations into these three particular con- 
stituents. We can, certainly, say that orange is a sort of mix- 
ture of red and yellow, violet a mixture of blue and red, and so 
on. But even in these cases the phrase ' intermediate between ' 
would be more correct than ' mixture of.' It seems to me, at 
any rate, that my sensation of orange and violet makes of them 
impressions as simple as it does of red, blue, or yellow. And so 
much is indisputable : that no one can say he senses red and 
green in yellow, or red, green, and violet in white. Subjectively 
white is just as simple as any simple colour. And black we 
shall all incline to regard not only as the minimal degree of 
intensity of white, but at the same time as its qualitative 

There is only one possibility left If the existence of three 
primary colours is incapable either of physical or of psycho- 
logical explanation, it must depend simply and solely upon 
physiological conditions. If we accept the principle that to every 
difference in our subjective sensations there corresponds a differ- 
ence in the physiological stimulation-processes within the sense- 
organ, we must suppose that the three objective light-qualities 
in the red, green, and violet portions of the spectrum, when 
mixed in the right proportions, can set up as many physiological 
excitatory processes as there are subjectively discriminable 
sensations. How many of these stimulation-processes are 
possible we cannot directly determine : but they must be esti- 
mated from the number of distinguishable sensations, and not 
from the number of objective light-stimuli by which the sensa- 
tions are occasioned. 

96 Lectures on Human and Animal Psychology 


Simple as these considerations are, they have not as yet been 
able to command anything like universal assent. In the current 
theories of the nature of light and colour-sensation we find all 
too often a confusion between the physical and the physiological 
stimulus, and again between this latter and the sensation 
correlated with it ; or, if not that, the opposite error of a deriva- 
tion of the objective conditions of light-excitation from the 
subjective differences in sensation, a hypothesis arbitrary in, 
form and contradictory of experience in content. 

For instance, Thomas Young, an English physician and 
physicist of the beginning of the present century, maintained 
that all our sensations of light and colour are compounded of 
the primary sensations red, green, and violet. There exist in- 
the eye, he said, three kinds of nerve-fibres, sensitive respectively 

to the red > green, and 
violet rays. We can 
represent the laws of 

Gelb f \J3Zau colour - mixture by 

drawing a triangle, as 
in Fig. 10, the three 

Orange/ Weiss. \ Indigo angles of which are 

constituted by the 
three primary colours, 
Violett while the intermediate 

furpur spectral tints are placed 

FlG - I0 - along its sides, purple, 

e.g., between the red and violet of which it is compounded, and 
white occupies the centre of its area. Such a triangle, accord- 
ing to Young, would express equally well the conditions of 
visual sensation and visual stimulation. Orange and yellow, for 
example, would excite the fibres sensitive to red and green, the 
red predominating in orange, and the green in yellow, while the 
sensation of white would result from the excitation of all three 
fibres in approximately equal intensity. The sensation of 
white, that is, is simply a mixture of the three sensations red, 
green, and violet. The impression of a whitish colour occurs, 
on the other hand, if one or two of the primary colours 

The Three Primary Colours 97 

predominate in the mixture. These whitish colour-tones may 
therefore be written in upon that portion of the area of the 
triangle which lies between the centre (white) and the sides. 

Thomas Young regarded his own assumption of three funda- 
mental sensations as no more than a provisional hypothesis, 
especially useful for the explanation of the phenomena of colour- 
mixture. But many physiologists and physicists of recent times 
have imagined this supposed structure of the organ of vision to 
be a necessary corollary from the facts of colour-blindness, and 
so have made the hypothesis a certainty. Colour-blindness does 
not, as a rule, signify blindness to colour in general, but only 
insensibility to certain particular colours. If it is connate, the 
colour-blind person need not be conscious of his defect, which 
simply consists in the confusion of special colours, red and 
green, e.g., clearly distinguishable by the normal eye. Experi- 
ence shows that in the great majority of cases colour-blindness is 
red-blindness, though green-blindness also occurs. But red and 
green are fundamental colours, so that we seem to have in the 
phenomena a confirmation of Young's theory. The condition 
of the eye, you see, is quite easily explicable in terms of it. We 
have simply to regard one of the three sets of nerve-fibres or 
terminal organs of the normal eye as absent or functionless : in 
red-blindness those sensitive to the red rays ; in green-blindness 
those sensitive to green. 

Nevertheless, the proof is not so unexceptionable as has often 
been thought. If we grant that the only forms of colour-blind- 
ness are the ' red ' and ' green ' types, we may, perhaps, find it 
to be a necessary inference that there are normally particular 
portions of the retina especially sensitive to red and green light 
and that these, for some unknown reason, are in the defective 
organ either absent or insensible ; but we shall not find the least 
ground for supposing that the sensation of yellow, e.g., is a mix- 
ture of the sensations of red and green, or that stimulation by 
yellow light simply means a stimulation of the elements sensi- 
tive to red and green. The first proposition is negatived by the 
character of the sensation ; yellow is qualitatively different both 
from red and green, and shows no trace of a mixture of the two. 
The second we can hardly regard as probable, unless we are will- 
ing entirely to give up the principle which has stood us in such 


98 Lectures on Human and Animal Psychology 

good stead heretofore, the principle of parallelism of differences 
in physiological stimulation with differences in sensation. But, 
besides this, the progress of our knowledge of the phenomena of 
colour-blindness has brought facts to light which are irreconcil- 
able with Young's hypothesis. First of all, it has been found 
that ' red ' and ' green '-blindness, though the commonest, are 
not the sole types of abnormal colour-sensibility. Cases are 
known of insensibility or diminished sensibility to other rays in 
the spectrum, especially the yellow and blue. Secondly, there 
are extant observations of unilateral or monocular colour-blind- 
ness. Now, on Young's hypothesis, white must here be com- 
posed of different primary sensations in the two eyes ; in 
unilateral red-blindness, e.g., it would be a mixture of red, green, 
and violet in the normal eye, of green and violet only in the 
other. The same white light would therefore be differently 
sensed by the two eyes ; to the normal it would appear white, 
to the other, which has no sensation of red, greenish. As a 
matter of fact, this is not the case ; the same white appears 
precisely the same to both eyes. Thirdly and lastly, we have 
instances of total colour-blindness. This generally results from 
disease of the eye, and is frequently confined to one retina, or 
even to portions of it. Black and white, with all the inter- 
mediate greys, can still be sensed, but there is no sign of a 
colour-sensation of any kind. A picture looks like a drawing ; 
light and shade are clearly distinguished, but there is absolutely 
no perception of colour. It is obvious that such a state of 
things could never come about if every light-sensation were the 
resultant of a mixture of the three primary colours. Total 
colour-blindness would be impossible unless the sensation of 
brightness and that of colour were correlated with different, and 
under certain circumstances separable, excitatory processes in 
the visual organ. And here we have a fresh proof of the validity 
of our principle that difference of physiological stimulation- 
process runs parallel to difference in sensation. For this 
independence of white from the various colours might have been 
inferred before from the independence of its quality in sensation. 

Leonardo s Foiir Principal Colours 99 


These facts have, of course, told against Young's theory ; and 
an attempt has been made in quite recent times to replace it by 
another. To this end the views of Leonardo da Vinci a name 
as conspicuous in the history of science as in that of art have 
been revived. Leonardo regarded four colours, which he called 
principal colours, as of prime importance, and looked upon all the 
others as intermediate or mixed colours. These principal colours 
were red, yellow, green, and blue. To them must be added black 
and white. From the six fundamental qualities so obtained and 
from their intermixture in various amounts all our sensations of 
light and colour could, he urged, be derived. Orange, e.g., is 
given in immediate sensation as a mixture of red and yellow, 
violet as a mixture of red and blue. 

Leonardo's view is entirely based upon the subjective cha- 
racter of our sensations. And it might perhaps have held its own, 
being not without interest as an expression of the psychological 
side of the matter, if the attempt had not been made to graft 
further hypotheses upon it for the explanation of the objective 
laws of light-stimulation and colour-mixture. But it was 
assumed, e.g., that between the members of each pair of principal 
colours there existed an 'antagonism ' analogous to that between 
white and black. Antagonistic colours were defined as those 
which cancelled each other when mixed, leaving only the sensa- 
tion of brightness, which accompanies every colour-sensation of 
whatever quality. Red and green, blue and yellow, were 
regarded as antagonistic in this sense. To make the dominant 
idea of the theory still more definite, it was supposed that there 
are intermingled in the retina three different kinds of sensitive 
substance, in each of which two antagonistic processes may be 
set up, corresponding in a manner to the processes of anabolism, 
or assimilation, and catabolism, or dissimilation, which are found 
together throughout organic nature. 

For the sake of brevity, we will term these processes a and d, 
and call the three substances, after the sensations which they 
mediate, the black-white, the red-green, and the blue-yellow. 
The assumption then is, that the sensation of black is due to an 
^-process, white to a (/-process, in the black-white substance ; in 

TOO Lectures on Human and Animal Psychology 

the red-green, red is the a and green the </-process, or vice versa ; 
and similarly with the blue-yellow. But every stimulation of a 
colour-substance involves the excitation of the black-white ; and 
so it comes about that if the a and ^-processes of one or both 
colour-substances cancel each other, we still sense brightness. 

It cannot be denied that this theory has done good service. 
For the first time since the banishment from science of the older 
colour-theory of Aristotle and Goethe, it called attention to the 
fact that white light is just as simple in sensation as any mono- 
chromatic light, and that black and white are not only to be re- 
garded as different intensities of a single quality, but also, and 
indeed predominantly, as qualitative opposites. In all other 
respects, however, it is simply an arbitrary combination of arbi- 
trary assumptions. Even the analogy between the ' antagonis- 
tic ' colours and the black-white pair cannot be carried through. 
When we mix black and white we get grey. And grey is 
directly sensed as a quality intermediate between the two ex- 
tremes. But when we mix red and green or blue and yellow, 
there is no mixture, but only mutual disturbance in sensation ; 
the only thing that is left is white, and white was present from 
the beginning, except that the colours were too strong for it. 
Again, the primary colours, indicated by the laws of mixture, 
have to be accommodated to Leonardo's principal colours. That 
is not altogether easy. We are obliged to change the names of 
the colours to suit our theory. For the antagonistic colours are 
not what we generally call pure red and pure green, pure yellow 
and pure blue ; if we are to obtain complementariness, the red 
must be tinged with purple, and the blue with indigo-blue, in 
other words, both these principal colours be mixed with a con- 
siderable amount of violet. And, lastly, it is a necessary conse- 
quence of the theory that there should be no partial colour- 
blindness without annihilation or diminution of the sensibility 
to the two members of some pair of antagonistic colours. For 
it can only explain the affection as the result of some defect in 
one of the two colour-substances. We ought, that is, to find 
red-green and blue-yellow blindness, but no other form. As a 
matter of fact, there cannot be any doubt at all that red-blindness 
may occur without green, and green without red-blindness. 

But not only does the assumption of antagonistic colour pairs 

Leonardo s Four Principal Colours 101 

come at every point into conflict with facts : its psychological 
foundation is exceedingly questionable. The four principal 
colours, red, yellow, green, and blue, are, you see, the only simple 
sensations ; all other colours are immediately given as compound 
in sensation. What support is there for that assertion ? Plainly 
this, first of all : that when we have once recognised these four 
as fundamental sensations the others fall into place readily 
enough as subjective intermediates ; and secondly this : that 
the names of these four are by far the oldest colour-names, the 
others bearing the evidence of modernity upon their face. Now 
the first of these facts is only of importance in the light of the 
second. If language had originally contained special names, 
say, for orange and violet instead of for red and yellow, it is 
quite probable that we should be inclined to look upon these 
latter as intermediate tones. So that everything depends upon 
the answer to be returned to the quescion : why is it that these 
particular four ' principal ' colours were the first to receive definite 
colour-names ? 

From the point of view of the theory which is under considera- 
tion, it is a matter of course that the reason can only be looked 
for among the immediate facts of sensation ; that red and yellow 
are just given as simple, while orange is sensed as a compound 
impression. Language, that is, preferred at the outset only to 
name what was simple in sensation. Now this view obviously 
proceeds from a wrong postulate as regards the origin of word- 
symbols in language. In the first place, it is not true that a 
separate word must exist for every qualitatively simple sensation. 
More than one philologist has called attention to the fact that 
a sharply distinguished term for ' red ' appears earlier than one 
for 'blue.' In the ancient literatures, e.g., in Homer, the 
expressions for the blue of the sky are such as could be used for 
any dark or grey object. Now and again the conclusion has 
actually been drawn from this that the Greeks of Homer's time 
had not yet sensed blue ; that is, that the colour-sense in man 
has within this comparatively short period undergone a very 
considerable development. We shall hardly feel tempted to 
assent to that conclusion. Language does not distinguish every- 
thing that sensation distinguishes ; it contents itself with creating 
special terms for those impressions whose discrimination is 

IO2 Lectiires on Human and Animal Psychology 

necessary for the expression of thought and its communication 
to others. Are we to suppose that it is only since Newton's day 
that mankind can distinguish orange from yellow, or indigo-blue 
from sky-blue ? Surely not. These new names of colour-tones 
simply came into use when they were needed for optical or 
technical purposes. To make assurance doubly sure, it has been 
quite recently proved that the scale of colour-sensations in 
various savage races presents no differences from our own. 

At the same time the four principal colours do in one sense 
constitute a special case. Whenever different colour-names 
occur, they are ultimately reducible to these four. So that there 
is some show of reason for assuming their original sensational 
preference. Nevertheless the law holds even here that language 
does not name sensations because of any subjective peculiarities 
that they may possess, but merely with reference to their 
objective significance. So that whenever we can follow a colour- 
name back to its original meaning we find it indicating some 
external object by which the colour-sensation is occasioned. 
Orange, indigo-blue, and violet are named from the colour of the 
fruit, the dye-stuff, and the flower. Now what are the colours 
which mankind would have named first, on this principle ? 
Surely either those which excited in them the strongest feelings, 
or those which were commonest among the natural objects of 
their environment : the red of blood, perhaps, the green of 
vegetation, and the blue of the sky, against which the bare earth 
and the light of sun, moon, and stars looked yellow instead of 
white. So that there is no need for us, in our search for the 
origin of the four principal colours, to descend into the bottom- 
less abyss of theory, and postulate a sensation that is independent 
of any impression. And, of course, these colour-impressions, 
which frequent repetition or some other reason has brought 
more prominently than the others before consciousness, must 
obtain an advantage not only as regards linguistic expression, 
but also in sensation itself, in that all other sensations are 
arranged with reference to them. If once red and yellow are 
given, orange can only be looked upon as an intermediate tint. 
In the same way purple and violet fall into place between blue 
and red. And since there is a continuous transition from colour 
to colour, while the range of sensation is strictly limited, the four 

Theory of Light-sensations 103 

principal colours were amply sufficient to allow of a permanent 
arrangement of all the possible colour grades. Had the dominant 
impressions, which determined at once colour-arrangement and 
colour nomenclature, been orange, yellowish green, greenish blue, 
and violet, instead of red, yellow, green, and blue, we should 
undoubtedly have sensed red as an intermediate between violet 
and orange, and green as a tint lying between yellowish green 
and greenish blue. Isolate any one of these colours for your- 
selves from the spectrum, and ask yourselves whether the 
impression it makes is not that of absolute simplicity, if once 
you abstract from the relations into which it has been brought 
with other colours by the customary arrangement of the colour- 


The principal colours owe their prominence, then, not to any- 
thing in the quality of sensation itself, but to external circum- 
stances which have nothing at all to do with sensation-quality. 
Subjectively every colour-impression is a simple quality, 
resembling its nearest neighbours, but only resembling them 
because of the continuity of tints in the colour series. And the 
mention of this fact of continuity leads us to another point, 
which the two colour theories discussed above leave entirely out 
of account, the relation of the two terminal colours of the 
spectrum to each other. Red and violet are not the most 
different of all colour-sensations, as their spatial positions on the 
colour scale might lead us to expect ; on the contrary, they are 
as much alike as any two colours can be. Here again colour- 
sensation and brightness-sensation are diametrical opposites. 
The maximal differences of vibration-rate produce in the one 
case similar subjective effects ; while in the other the extremes 
of luminous intensity correspond to opposite qualities of sensa- 
tion, black and white, between which the whole series of bright- 
ness sensations is arranged in continuous progression. Now, 
whatever hypothesis we adopt, we must seek to do justice to all 
these peculiarities of sensation, besides taking account of the 
laws of colour-mixture. Bearing that in mind, we arrive at 
something like the following theory. 

We may suppose that every retinal light-stimulation is com- 

104 Lectures on Human and Animal Psychology 

pounded of two separable constituents, a colour-excitation and 
a brightness-excitation. The brightness or ' achromatic ' excita- 
tion may occur by itself. When that is the case, we sense black, 
white, or grey. The colour or ' chromatic ' excitation always 
implies the presence of the achromatic. When it occurs, we 
sense saturated or whitish colour, according to the intensity of 
the concomitant achromatic process. Differences in this latter 
we may regard as ultimately conditioned by objective differences 
in the intensity of light. It consists always of two qualitatively 
opposite part-processes, one attending stimulation by light 
and corresponding to the sensation of white, and the other 
accompanying the recuperation of the retina after stimulation 
and corresponding to the sensation of black. This part-process 
of recuperation, you observe, is not only present when the visual 



FIG. ii. 

organ is entirely free from stimulation, but accompanies the 
more moderate degrees of stimulation, as a reaction of the 
stimulated substance in the direction of restitution of the de- 
composed chemical compounds. In the event of very weak 
excitation, its intensity may even exceed that of the other part- 
process, that of stimulation. We can therefore represent the 
system of brightness-sensations by a straight line, terminating 
in black and white and having the various shades of dark grey, 
grey, and bright grey arranged along its length in all their 
manifold but perfectly continuous gradations. Such a line is 
drawn in Fig. II. The opposed processes are represented by 
vertical lines falling upon it, the intensity of the excitation- 
process being shown by the length of the ascending verticals, 
and the intensity of the recuperation-process by that of the 

Theory of Light-sensations 105 

descending ones. At absolute black the excitation-value must 
be regarded as zero, while recuperation is at its maximum ; at 
the brightest white recuperation is at its vanishing point, and 
excitation at a maximum. Every intermediate brightness- 
sensation presupposes a mixture of both processes. The total 
physiological process corresponding to a mean grey, e.g., is com- 
posed of a stimulation of the magnitude a w and of a recupera- 
tion of the extent a s. These two part-processes do not cancel 
one another, but intermix, so that in sensation grey is a quality 
intermediate between those of black and white, and equally 
related to both. 

If we attempt to construct a similar geometrical diagram for 
colour-sensations on the basis of their subjective peculiarities, 
we must substitute for the straight line a curved one ; its two 
ends must approximate, to indicate the subjective similarity of 
red and violet. We may choose a circle, as the simplest line of 
the description required. Then, as in Fig. 12, all the saturated 
colours may be arranged round its periphery. But as the colours 
of the solar spectrum leave a gap 
between red and violet, we must 
fill the series out by introducing 
purple. Purple is obtained by 
mixing red and violet, and all its 
various shades lie between those 
two qualities. The process of 
chromatic stimulation, on the 
principle of parallelism of sensa- 
tion and physiological stimulation, 
must be regarded as recurrent or 
periodic ; the processes set up in 

the retina by the quickest vibrations must resemble those 
resulting from the slowest. You may find an analogy in the 
octave. Although the ground-tone is further removed from its 
octave than is any other tone of the chromatic scale, yet these 
two are more nearly related in clang-character than any two 
others. Indeed, this may be something more than an analogy. 
The vibration-rate of the octave is twice that of its ground-tone, 
and the vibration-rate of the light rays at the extreme violet 
end of the spectrum is approximately twice that of the red rays 

io6 Lectures on Human and Animal Psychology 

At the same time, the conditions of light and sound-stimulation 
are in other respects so different that the attempt occasionally 
made to discover in the series of intermediate colours the exact 
physical and sensational correlates of the principal intervals of 
the tonal scale, fifth, fourth, the two-thirds, etc., has in every 
case resulted in failure. 

The great difference between colours and tones, which pre- 
vents any such attempt from succeeding, is seen most clearly in 
the different effects of the mixture of light and sound-waves. 
When we mix sound-waves, we get a compound sensation, con- 
sisting in the last resort of just as many simultaneously dis- 
tinguishable elements as it contains successively distinguishable 
tones. But when we mix light-waves, we always obtain a simple 
sensation. White, which is composed of all the rays of the solar 
spectrum, is in the last resort just as simple as is any colour 
which contains but one single wave-modality. 

These peculiarities of light-sensations lead us to two facts of 
general import, which must be taken account of by any theory 
of chromatic stimulation equally with the subjective similarity 
of the terminal spectral colours. First, there may be substituted 
for any simple colour a mixture of the two bordering colours 
between which it lies. We can obtain orange from red and 
yellow, yellow from orange and yellowish green, a pure green 
from yellowish green and greenish blue. Secondly, any colour 
mixed with its subjective opposite in the right intensive propor- 
tions gives us the sensation of white; the two are complementary 
colours. The first of these facts indicates that the process of 
retinal stimulation is not continuous, but graduated. For the 
operation of any particular grade of vibration rapidity there can 
be substituted the mutually supplementary actions of two other 
grades not too remote from it, one of a less and one of a greater 
rapidity. These combine to give an intermediate result. The 
second fact, like the recurrent course of the colour-line, proves 
that colour-stimulation is a circular process, in the sense that 
every form of it is correlated with an opposite form ; so that 
when the two excitations concur they cancel one another 
leaving only the concomitant achromatic stimulation in sensation. 
We can express this in our diagrammatic construction (Fig. 12) 
by supposing the colours on the periphery of the circle to be so. 

Theory of Light-sensations 


arranged that the members of every complementary pair are 
directly opposite to each other, and can be connected by a 
straight line passing through the centre of the figure. At the 
centre itself we may put white, and on the area between centre 
and periphery write in the whitish colour-tones, as they gradually 
change from complete saturation to absence of colour or 

All the simple qualities of visual sensation are contained in 
this figure, with the exception, first, of black and the greys 
intermediate between it and white, and, secondly, of the blackish 
or greyish colours, brown, olive-green, etc. If these dark shades 
of colour and brightness are to be introduced into the diagram, 
we must first draw a perpendicular to the centre of the circle, 
white (Fig. 12), and arrange along its length the series of 

achromatic sensations, and 
then fill in the dark colour- 
shades and the intermediates 
between them and grey or 
black on concentric circles 
placed at different heights 
round the perpendicular line 
of brightness-sensation (Fig. 
13). In this way we get a 
cone with a circular base, 
the apex corresponding to 
FIG- 13- black, the centre of the base 

to white, and the periphery of the base to the saturated colours, 
while all other qualities are distributed over and through the 
body of the cone, their exact position being determined by 
reference to the fixed positions of these fundamental qualities. 





Blass, pale. 
Blau, blue. 
Dunkel, dark. 

Griin, green. 
Hellgrau, bright grey. 

Schiuarz, black. 
Strohgelb, straw-yellow. 

Himmelblau, sky-blue. Violet '/, violet. 

Fleischroth, flesh-coloured. Purpur, purple. 
Gelb, yellow. Rosa, rose. 

IVasserblau, water- blue. 
Weiss, white. 




IN the light of what we already know regarding the pro- 
perties of sight and hearing, we can see that the best way 
to understand the relation between these two senses is to turn 
to their two essential differences. The first is this : that, 
whereas a subjectively simple tone-sensation can only be caused 
by a simple objective movement of the air, in the case of sight 
any vibration of ether, simple and complex alike, produces a 
-simple light-sensation. And the second consists in the fact 
that the simple tonal qualities can only be varied in two 
directions to produce higher and lower tones, whereas light- 
sensations form two series, the chromatic and the achromatic, 
-each consisting of a number of fundamental qualities and 
intermediate tints, while the chromatic series, owing to the 
subjective relationship of the terminal colours of the spectrum, 
further constitutes a manifold of qualities which returns upon 
itself. The system of simple tones, therefore, can be represented 
by a straight line ; that of light-sensations requires a figure of 
three dimensions for its expression. Besides these general 
differences, there are further distinctions dependent on various 
properties which attach to the sense of sight, either exclusively. 
or to a greater extent than to any of the other senses. 


If we produce a tone (e.g., by striking a string or tuning-fork), 
.and then suddenly arrest the vibrations of the sounding body, 


After- images. \ 09. 

the tone-sensation immediately ceases. Even if it actually lasts 
somewhat longer than the stimulus, the duration of its after- 
effect is so short as normally to escape our notice. It is quite 
different with the sensation of light and colour. You know 
that if a red-hot ember is swung round at a moderate rapidity, 
you see a complete circle of fire. This phenomenon shows that 
the light-impression in the eye must persist at least during the 
time that the ember takes to pass from any point in its course 
back again in a circle to the same position. We can obtain 
more exact knowledge of the after-effect of light-stimulation by 
fixating a luminous object for some time and then suddenly 
closing the eyes. We then see on the dark field of vision an 
after-image which resembles the object, but which gradually 
undergoes very curious changes as regards its light and colour- 
properties. For the first moment it is exactly like the external 
object, then its intensity decreases somewhat, and again after a 
little time its quality alters to just the opposite of the original. 
If the fixated object were white, the after-image turns black, 
darker, i.e., than the dark field of vision on which it is seen, if 
the object were black, the after-image becomes white, brighter 
than the surrounding field of vision. Finally, if the impression 
were coloured, the after-image assumes the complementary 
colour, greenish if the object were red, reddish if it were green. 
In its first stage, therefore, the after-image is called positive or 
same-coloured; in its later phases, negative or complementary. 
Both phenomena, but especially the more persistent comple- 
mentary images, are observable with open eyes, if the object 
fixated is extremely bright. If you glance, e.g., at the setting 
sun, and then look at the roadway or a grey wall, you may see 
a clear green after-image of the solar disc at the point of fixa- 

The phenomena of after-images prove, in the first place, that 
the stimulation-process in the retina outlasts the external 
stimulus by a considerable period, often amounting to several 
seconds. But they show also that the stage of direct continu- 
ance of stimulation, which is manifested in the positive after- 
image, is followed by an opposite condition of the sense-organ 
during which bright objects are seen as dark, and dark ones as 
bright, and in general every colour as its complementary. 

i io Lectures on Human and Animal Psychology 

This complementary after-effect may be easily explained on 
the assumption of a partial exhaustion of the retina. If, e.g., 
our sensibility to red is exhausted at a particular place, we 
shall see white light at that place just as if no red rays were 
affecting it ; in other words, since the subtraction of red from 
white gives green, a red object will leave behind on the exhausted 
retina a greenish after-image. The quality of these comple- 
mentary after-images may be enhanced by contrast with its 
background. This holds especially of the cases mentioned 
above, where the after-images of white and black objects were 
observed upon a dark field of vision. 


These striking differences in the after-effects produced by 
stimulation of the two sense-organs render it probable that the 
processes which excite the sensations of tone and light are 
totally different in character. Indeed, it seems probable that 
sound vibrations set up corresponding vibrations in the 
membrane of the cochlear canal, in which the fibres of the 
auditory nerve terminate ; and that our separate perception of 
the constituent tones of a clang is due to the tuning of the 
individual fibres of this membrane to various tones. The 
stimulation-process would, therefore, here be a mechanical one, 
and the sensation is consequently interrupted as soon as the 
vibratory movement of the stimulated fibres ceases. That 
must, of course, occur very soon after the cessation of the 
external air vibrations which set up the excitation. It is 
probable, once more, that the rapid arrest of the vibrations of 
the membrane is promoted by the action of certain solid 
cuticular structures, semicircularly arranged, which rest upon 
it, and whose function must be analogous to that of the damper 
of a pianoforte. But when a light-stimulus affects the retina, 
we have obviously a very different state of affairs. There are 
many observations which go to prove that the sensitivity of the 
retina to light is like that of the prepared plate in the dark 
room of a photographer. One of the most striking is this : in 
the dark, the retina is of a deep red colour ; exposed to light, 
it gradually bleaches, and finally turns white. This renders 
it practically certain that light-stimulation is a photochemical 

Mechanical and Chemical Senses 1 1 1 

process. Such processes play an important part in organic 
nature ; e.g., in the breathing of the green portions of plants 
and in the production of the colours of flowers. Now a chemi- 
cal process, even if it is comparatively soon over, always 
requires a considerably longer time for its completion than docs 
a simple transmission of motion. Regarded from this point of 
view, the after-image appears simply as a subjective indication 
of the duration of a photochemical action ; and its two phases 
point to the fact that there are two processes which run their 
course during that action. The positive after-image gives us 
the duration of the chemical decomposition occasioned by the 
light-stimulus ; the negative or complementary after-image 
shows us an after-effect of this decomposition. This latter is a 
phenomenon analogous to those of exhaustion in other living 
organs, ^.,in nerve and muscle, all alike manifest themselves 
in decreased excitability for stimuli of the same kind as those 
previously operative. 

The senses of sight and hearing may, therefore, be regarded 
as the principal representatives of two fundamental forms of 
sensory excitation, the mechanical and the chemical. In setting 
these expressions over against one another, we must be careful 
not to refer them to processes in the sensory nerves. These 
consist, probably in every case, of very rapid chemical decompo- 
sitions. Our purpose is merely to discriminate between the 
different ways in which external stimuli affect the nervous 
terminations in the organs of sense. With the mechanical 
senses may probably be reckoned (besides hearing) that of 
cutaneous pressure, as is shown by the brief continuance of the 
after-effect of impression in sensation. To the chemical senses 
belong (besides that of sight) the cutaneous temperature-sense 
and the senses of smell and taste. The organ of touch is the 
earliest, and in the lowest forms of organic life the only, sense- 
organ. And the fact that it includes both a mechanical and a 


chemical sense is, perhaps, not without significance for the 
physiological history of the development of sense-functions. 


The phenomena of negative and complementary after-images 
which we have been discussing require little more than purely 

H2 Lectures on Human and Animal Psychology 

physiological explanations. There is, however, a very large 
number of cases in which our sensations of light and colour 
undergo change where this cannot be derived from the after- 
effects of stimulation and from the influence of exhaustion, or 
where at least these phenomena afford only a partial explana- 
tion. Of course, wherever after-images manifest themselves at 
the same time, they may be explained in physiological terms. 

If we cause light to pass through a sheet of red glass, and 
then cast a shadow somewhere upon the red illuminated surface, 
we ought really to see this shadow grey. For it contains nothing 
but diffused white light, the intensity of which has been lessened 
by the shade. As a matter of fact, however, the shadow looks 
not grey, but green. This same green tint may sometimes be 
observed in nature, in the shadows cast by trees, when the 
setting sun gives a reddish light in consequence of the stronger 
absorption of the refrangible rays by the atmosphere. 

An experiment which exactly reproduces the conditions of 
this subjective colouring of natural shadows can easily be made 
by means of the colour top and the rapidly rotating discs which 
we employed to illustrate the mixture of colour-sensations. We 
take a disc with small coloured sectors on a white background. 
Somewhere about half-way between the centre and periphery, 
the coloured sectors are replaced by narrower portions of a 
black band (Fig. 14). If we let the top rotate at a high speed, 
the colour of the sectors fuses with the white of the background 
to a whitish tint ; and where we inserted our black bands, we 

get a mixture of black and white, i.e.,. 
grey. This grey, you see, corresponds 
completely to a circular shadow cast 
upon the coloured background. But 
this objectively grey ring does not 
appear to us as really grey, but as 
coloured, and as coloured comple- 
mentarily to the background. If the 
sectors are green, the ring looks red ; 
if they are red, we see it as green, etc. 

r IG. 14- 

Here is a still simpler experiment. 
Take a sheet of thin white notepaper and a sheet of coloured 
paper of the same size ; lay the white paper on the coloured 

Phenomena of Contrast 1 1 3 

so that it exactly covers it ; and then push a little square of 
grey or black paper between the two. If the coloured paper 
which is underneath is green, e.g., its colour shines through the thin 
notepaper, except at the place where the grey square lies. This 
should appear grey, but, as a matter of fact, looks red. If our 
coloured paper had been red, the grey square would have looked 
green. In short, it always takes on that colour which, mixed with 
the colour of its surroundings, would produce white. The pheno- 
menon is not so striking, but yet clearly apparent, when you 
simply lay a piece of grey paper on a coloured background, with- 
out covering it with the thin notepaper. Suppose, e.g., that you 
cut little squares from the same grey paper, and lay them upon 
red, green, yellow, and blue papers, placed side by side. They 
all look different ; that on the red is greenish, that on the green 
reddish, that on the yellow bluish, that on the blue yellowish. 
And the effect of the background is just as obvious, if you em- 
ploy colourless lights of various brightnesses, instead of colours. 
Fasten two similar grey squares upon black and white sheets of 
paper respectively ; the former appears so bright as to be almost 
white : the latter looks dark, and under favourable conditions 
may approach to black. 

In describing all these phenomena and enumerating their 
conditions, we have left out of account one important factor, 
the extent of the surfaces employed. If the shadow cast in 
coloured light is very large, the centre will look grey ; only at 
the edges will its colour become apparent We can best show 
the dependency of subjective colouring upon the space-relations 
of the surfaces compared by having recourse again to revolving 
discs. We furnish a disc with sectors in the way shown in Fig. 
15 ; the sectors are cut into steps, the absolute magnitude of 
which is in every case the same. 
If now the sectors (B} are blue, 
and the background () yellow, 
we shall expect, when the disc is 
rotated, to see a bluish-yellow mix- 
ture, which is perfectly uniform 
within each step of the sectors B, 
but which is changed from step to 

step in such a way that the yellow 

r 1G. 


ii4 Lectures on Human and Animal Psychology 

predominates more and more in the mixture as we approach the 
periphery ; for objectively the colour of each step remains the 
same throughout its whole extent. In reality, however, the 
inner and outer borders of each blue-yellow ring are differently 
coloured, and the intermediate points show a gradual transition 
from one shade to the other. Each of these constituent colours 
shows most strongly when contiguous to a ring which contains 
less of it in its mixture. On our own disc, therefore, the outer 
edge of every ring is blue, and the inner yellow. We have, that 
is, a succession of yellow and blue rings outlined upon a back- 
ground of mixed colour. 

The same experiment can be made in a different way. We 
will take a white disc and paste upon it not coloured, but black, 
sectors, as has been done in Fig. 15. We should expect to 
obtain by rotation concentric grey rings, increasing in brightness 
toward the periphery of the disc, but showing no variation within 
each ring. We find instead that all the rings look brighter on 
the inside, where the contiguous ring is darker, and darker on 
the outside. The difference is so great that the brightness of 
the different rings oftentimes appears to be the same through- 
out ; so that when rotated rapidly the disc seems to be com- 
posed simply of alternating black and white circles. 

These experiments illustrate a group of phenomena to which 
the general name of contrast phenomena is given. The name is 
an obvious one, for their principal condition is clearly the oppo- 
sition of two colours or of two degrees of brightness. And for 
the same reason the subjective colour due to this opposition (the 
green tint of the shadow in red light) is also termed contrast 
colour. We may, therefore, use the expressions 'contrast colour' 
and 'complementary colour' as synonymous. 

The identity of these two concepts has been the chief agent 
in determining a large number of physiologists to refer the 
phenomena of contrast to like causes with complementary 
after-images. Just as the after-image arises from persistence of 
the excitation in the stimulated region of the retina, so contrast, 
they say, is a result of a diffusion of excitation over the sur- 
rounding portions of the retina. This analogy is, of course, 
condemned at once by the fact that we distinguish a positive 
after-image, resembling the original stimulation, from a negative 


phase of it, which is opposed to the former ; while the effect of 
contrast is always negative and antagonistic. Moreover, a con- 
siderable time is necessary for the development of the after- 
image, whereas the contrast change of sensations is instanta- 
neous. But we have still other and weightier reasons to urge 
against the analogy. A spatial diffusion of stimulation should 
appear in more intensive form the greater the strength of the 
original stimulus. This is, however, not by any means the case. 

A weak stimulus may under favourable circumstances bring 
about a stronger contrast than a strong stimulus. This we 
learned from our experiment with colourless objects laid upon a 
coloured background. You remember that if we cover them with 
transparent notepaper, the contrast is much more obvious than if 
they are left uncovered, though their colour is weakened by the 
covering sheet. This suggests that contrast-effects in general may 
consist not in a direct alteration of sensation, but in changes in our 
subjective mode of apprehending sensations ; and the suggestion 
is fully confirmed by other observations. Let us vary our last 
experiment a little. By the side of the grey square, which has 
taken on the colour complementary to its background, we place 
a second square of exactly the same brightness. Now that com- 
parison of the two is possible, the contrast colour disappears, 
but it reappears so soon as the second grey is removed. 

Since the comparison of impressions is so plainly important in 
this connection, it has been maintained, as against the physio- 
logical hypothesis mentioned above, that contrast depends upon 
a deception of judgment, A shadow seen in red light looks green 
to us, it is said, because we are accustomed to regard ordinary 
diffuse daylight as white, and to judge of colours in comparison 
with it. So that if in a particular case it is not white, but red, a 
shadow cast in it must seem to us to be green ; for if reddish 
light looks white, light which is actually white will be no longer 
regarded as such, but will necessarily appear to have had a cer- 
tain quantum of red taken from it, *>., will seem greenish. If 
we mix all the colours of the spectrum together with the excep- 
tion of red, we obtain, of course, a green shade. The same prin- 
ciple of wrong judgment is employed to explain the facts 
observed in the experiment with transparent paper. If we lay 
a transparent white sheet upon a dark red one, it looks bright 

116 Lectures on Human and Animal Psychology 

red. Now we push our grey square in between the two sheets. 
This, of course, renders that portion of the surface colourless ; 
but we judge that the bright red paper extends over it also, and 
so see it in the colour which an object must possess if it is to 
appear colourless when seen through a red medium, i.e., as green ; 
or, to put it more generally, we see it in the colour complement- 
ary to that of the transparent covering. 

Now there are weighty objections to be urged against this 
derivation of contrast phenomena from deceptions of judgment. 
First of all, it is not right to say that because we are accustomed 
to see diffuse daylight white, therefore we must do so in the ex- 
ceptional cases when it is not really white, but coloured. We see 
clearly enough that the illumination of the setting sun is reddish, 
and we are not in the least inclined to regard the light that comes 
to us through a sheet of coloured glass as white ; we sense it 
precisely in its own proper colour. These instances serve to show 
that the assumption from which the explanation proceeds is not 
justifiable. And consider, furthermore, how extraordinarily com- 
plicated this series of inferences is, in terms of which we are said 
to sense. In the experiment with transparent paper, e.g., we are 
not only supposed to take account of the actual colour of the 
objects seen, but also of the influence of the medium through 
which the light rays have to pass before they reach our eye. 
And yet the experiments may be easily performed under con- 
ditions which definitely exclude any such influence. We have 
already seen that the contrast is also apparent when we lay our 
grey square directly upon the coloured background. Now, if 
pains are taken to choose saturations and brightnesses exactly 
corresponding to the modified colour and brightness appearing 
in the former case through the transparent paper, the contrast is 
just as clearly observable as before. So that it is not by betray- 
ing us into a false judgment that the transparent medium 
heightens contrast, but rather by producing degrees of satura- 
tion and brightness, which are especially favourable for it. 
Colour contrast always appears most strongly when the coloured 
object and the colourless one which it tinges by contrast are as 
nearly as possible of the same brightness. This condition is 
more nearly realised in all cases where the brightnesses of the 
objects are originally different, and their relations consequently 

Phenomena of Contrast \ \ j 

not so favourable for the production of contrast, by the use of 
the transparent paper as a covering. 

But although the judgment theory is untenable, the expression 
'comparison ' has a certain justifiable application to the process 
underlying contrast. We observe that not only is a really grey 
object modified by its surroundings so as to appear complement- 
arily coloured, but also that this contrast colour disappears as 
soon as we destroy the influence of the surroundings by intro- 
ducing another object of the same grey colour. It is, therefore, 
not amiss to say at least that this removal of contrast is the 
result of a comparison ; and if the expression is permissible in 
this case, it must be so also for our original phenomenon. For 
it is, indeed, conceivable that the result of one comparison could 
be cancelled by that of another, but hardly that an excitation- 
process due to definite physiological causes could be annulled 
by such an act. If we objectively tinge a grey paper with green, 
no amount of comparison of it with a pure grey of the same 
brightness can deceive us with regard to the existence of the 
colour. And we learn from the phenomena of after-images that 
a colour due to subjective retinal excitation behaves in this con- 
nection just like one objectively produced. But apart from these 
considerations, there is nothing at all in the phenomena of con- 
trast which is not done full justice to by this reference to a rela- 
tive comparison. A grey paper on a black background looks 
brighter than it does upon white. This is perfectly intelligible 
if we assume that our apprehension of a particular light-sensa- 
tion is not something absolutely unalterable, but is dependent 
upon other light-sensations which are at the same time present 
in consciousness, and in relation to which it is, so to speak, 
measured. The same point of view may be adopted for the ex- 
planation of colour contrast. If we gradually decrease the satu- 
ration of any colour, it passes over finally into white or grey. 
We may therefore regard the absence of colour as the lower 
limit of the various stages of saturation of any given colour. 
The principle of relativity which we have just been discussing 
renders it inevitable that the saturation of a colour should be in- 
creased, if its surroundings, presented at the same time with it, 
are complementarily tinged. This being so, the minimum of 
saturation will, like every other degree, pass over into a more 

iiS Lectures on Human and Animal Psychology 

complete saturation ; i.e., a colourless surface will in contrast to a 
colour appear complementarily coloured. 

The notion of comparison is, therefore, generally applicable 
to the mental process underlying contrast. But the judgment 
theory is wholly wrong in regarding the process as a compara- 
tive judgment, for the production of which all manner of com- 
plicated reflections have to be called into account. When it is 
said that these reflections take place unconsciously, a suicidal 
admission is made. At the same time that the mental process 
is resolved into logical reflections, it is really granted that these 
reflections do not actually exist, but only represent the transla- 
tion of a process of an entirely different kind into the form in 
which we usually cast our reflections. Now such a translation 
is quite allowable when it is simply used as a means of making 
intelligible the mode of operation of the elements which have 
been empirically shown to be present in some particular process. 
But though allowable to a certain extent, for the sake of clear- 
ness, this method of interpretation oversteps its rightful limits 
when it leads us to ascribe to judgment purely imaginary pre- 
liminary stages, such as, e.g., the reflection how a colour ought 
to appear when it is seen through another colour. That is what 
it does, you see, in the logical theory of contrast which we have 
been discussing, logical : for the theory is really logical, and not 
psychological. If we wish, therefore, to retain the concept of 
comparison to designate the processes under discussion, we shall 
only be able to mean by it an associative comparison ; i.e., a con- 
nection of two sensations, where the quality of each is deter- 
mined by its relation to that of the other. 


We have thus referred the phenomena of contrast to the 
same principle of sensational relativity with which we are 
already familiar as the general psychological expression for the 
facts of Weber's law. Our apprehension of stimulus-intensity 
or of tonal pitch is not something absolute, but depends upon 
the character of other stimuli and tones with which we bring the 
given sensation into relation. In the same way, the subjective 
effect of a particular light and colour is determined by the rela- 

General Law of Relativity. 119 

tions in which they stand to other impressions of light and 
colour which are affecting us at the same time. 

We may now subsume all these phenomena tonal intervals, 
light contrast, the geometrical increase of stimulus-intensity for 
equally noticeable sensation-differences under one general law: 
the law of relativity. The psychological significance of them all 
is the same. We may formulate this law, which may be re- 
garded as the most general expression of the results of our psy- 
chological analysis of sensations, as follows: wherever there 
occurs a quantitative apprehension of sensations, whether as re- 
gards intensity or degree of quality, the individual sensation is 
estimated by tJie relation in which it stands to other sensations of 
the same sense-modality. 

Sometimes this relative estimation of particular sensations is 
determined by the impressions which immediately precede or 
follow it ; this is usually the case in the measurement of equally 
noticeable stimulus-differences. Sometimes it is determined by 
sensations simultaneously present, as well as by those coming 
before and after; this is the case with tone-pitch. Sometimes 
finally, only the simultaneous impressions come into account, as 
in light and colour contrast. Which of these three conditions is 
fulfilled depends upon the special circumstances of the experi- 
ment, and upon the peculiarities of the special sense with whicli 
we are dealing. The matter does not at all affect the signifi- 
cance of the law. And this significance is, you see, predomi- 
nantly psychological. For the most obvious interpretation of the 
law is this : that we never apprehend the intensity of a mental 
state as if it stood alone ; we never estimate an isolated magni- 
tude ; but measurement implies a direct comparison of one con- 
scious state with another. And so we shall expect to find that 
the law of relativity is not restricted to the sphere of sensation, 
but is applicable in every case where the intensity of a mental 
process is quantitatively apprehended and compared with that 
of others. 




WE have seen that sensation is the ultimate source of a 
very large number of the concrete processes that go to 
make up our mental life. Everything that we meet 
with in our world of ideas is derived in the last resort from sen- 
sations, and ideas are the raw material of all the higher mental 
activities. It may be questioned whether the stream of thought 
could not continue without any reinforcement from the outside 
world ; but this at least is certain, that it has its source in sensi- 
bility, and that sensibility does perpetually interfere to deter- 
mine its direction. In this way our investigation of sensation 
leads on directly to the consideration of a second and very 
important psychological problem, the problem of the composi- 
tion of ideas from the sensational elements that enter into them 
in so great number and variety. 

In our previous discussions we examined the physical con- 
ditions of sensation, external sense-excitations, before we pro- 
ceeded to sensation itself. Now the further question arises : 
what is the immediate consequence of a sensation ? Observation 
furnishes us with the unhesitating answer : every sensation 
which is of sufficient intensity and not inhibited by opposing 
influences is followed by a muscular movement. This muscular 
movement we term a reflex movement, and the name is appro- 
priate. There have plainly taken place a transference of nervous 
excitation within the central organs of the nervous system from 
sensory to motor fibres, and a similar transmission of stimulation 
from these to their particular muscles : the stimulus is, as it 
were, thrown back, reflected. Reflexes are observed so long as 

Reflex Movements 121 

the nerves of sense retain their connection with the central 
organs of the nervous system, and these remain united by motor 
nerves to the muscles. But it is not necessary that the whole 
-central nervous system should be capable of exercising its 
normal functions. The cord may be severed from the brain, and 
reflexes still be mediated by the nerves which enter it ; indeed, 
quite a small section of the cord will suffice for the necessary 
transference of excitation. 

This transmission of stimulus by the sensory to the motor 
nerves is provided for in the elementary organisation of the 
nervous system. If we examine brain and cord microscopically, 
we find, first, a number of nerve-fibres of varying thickness, 
continuations of sensory and motor nerve-trunks, and secondly, 
besides these, cells of varying size, viscous structures, like most 
organic cells, containing a more solid nucleus and a number of 
small granules. These nerve-cells are characteristic of and 
peculiar to the central organs of the nervous system. How 
important they are functionally you may judge from the fact 
that they always stand in connection with both incoming 
(afferent) and outgoing (efferent) nerve-fibres. As a rule each 
one sends out several nerve-fibres. We may look upon these 
cells, then, partly as ultimate terminal organs, partly as organs 
for the mediation of connection between conducting fibres. To 
make the mechanism of the reflex clear, we have 
only to suppose a nerve-cell interpolated be- 
tween two fibres, one of which (e) leads from a 
sense-organ, the other (b] to a muscle. We 
have there the scheme of the reflex. As a 
matter of fact, however, the nervous connections 
are very much more complicated and involved, 
as we should expect, knowing what we do of the 
complexity of structure of the central organs. 

The intensity and extent of the movement constituting the 
reflex response to a sense-stimulus are enormously different in 
different cases. In general they increase with increase in the 
intensity of stimulus. The very weakest stimuli do not as a 
rule excite reflex movements at all ; moderate intensities arouse 
a moderately intensive movement, confined to some particular 
group of muscles ; as the intensity of stimulus is further in- 

122 Lectures on Human and Animal Psychology 

creased, the reflex answer becomes more general, till it finally 
involves practically the whole organism. This law of the in- 
crease of reflex movement with increase of stimulus remains 
constant despite the fact that individual and temporal differences 
may be by no means inconsiderable. These are mainly referable 
to varying excitability. The more excitable the sensory nerves 
and nervous centres, the earlier does the reflex make its appear- 
ance, and the more quickly does it run through the gamut of its 
intensive and extensive changes. Again, it may be intensified 
or reduced by the action of various influences upon the organism. 
Beheading, removal of the brain, increases the reflexes, until 
death occurs. Many amphibia may be kept alive for months 
after beheading, and their reflex excitability is throughout this 
whole period abnormally great. There are also certain chemical 
substances the effect of which upon the nervous tissue of the 
central organs is to occasion a decided intensification of the 
reflexes. Besides some alkaloids, whose influence is not very 
great, we may especially mention strychnine in this connection. 
It produces such an excess of sensibility that the lightest 
pressure upon the cutaneous nerves, such as normally would fail 
to arouse any movement at all, brings on reflex twitchings over 
the whole body. In deep sleep or swoon, on the other hand, the 
reflex excitability is diminished, as it is by the influence of opium 
and other cognate poisons. 

What the chemical changes within the nerve-cells are upon 
which the influence of strychnine and similar poisons depends is 
still unknown. But their general effect may be easily deduced 
from the universal laws of the action of forces. It cannot be 
supposed that a material heterogeneous to the constituents of 
nervous tissue creates fresh nervous force. The only tenable 
hypothesis is, that it facilitates the actualising of forces already 
latently present ; i.e., that it overcomes certain of the inhibitions- 
making against the transformation of stored into kinetic energy, 
and so renders the transforming force more effective. Substances 
of an opposite nature will, on the contrary, increase the number 
of inhibitions, and thereby increase the amount of external force 
necessary to release the cell from its state of tension. You will 
understand this easily enough if you take a simple mechanical 

Reflex Movements \ 2 3 

Suppose that you have a clock whose movement can at any 
moment be arrested or started by the throwing in or out of gear 
of some mechanical appliance, say a spring, which stops the 
works or not according as it is tightened or relaxed. So long 
as it is tightened, the weight which moves the clock will be 
exerting a pressure against it. This represents the stresses to 
which the mechanism is subject. As soon as it is relaxed, these 
are transformed into energy of motion. To effect the relaxation 
some small sum of work must be expended. Its magnitude will 
be proportional to the resilience of the elastic spring, and it can 
very easily be increased or decreased by varying the tension of 
the spring. 

The movement of the clock in this illustration is the reflex 
movement, the relaxation of the spring is the operation of a 
sense-stimulus, and the greater or less tension represents the 
influence of the substances which are exerting a specific action 
upon the mechanism of transformation of energy. Just as a 
greater tension of the spring increases the difficulty of starting 
the clock, so the alteration produced by opium in the nervous 
system makes against the release of a reflex movement ; and 
just as a less degree of tension facilitates the starting of the 
clock, so strychnine facilitates reflex movement. Every clock 
goes for a certain period, at the expiration of which it runs down 
and requires to be rewound ; in other words, there is in it a 
certain amount of potential energy which it takes a definite time 
to use up and transform into energy of motion, and which then 
requires renewal. Without this renewal the clock cannot go 
any more. Here again there is a complete analogy with the 
mechanics of the nervous system. There is a definite amount 
of potential energy contained in the central organ. This is. 
partly, only renewed when it has been almost exhausted (as in 
the cloclf), the restoration taking place during sleep ; but, partly, 
there is a continual process of renewal, as, indeed, there must 
be, if the chemical composition of the nervous elements is not to 
be so radically disturbed that a return to the normal condition 
is altogether impossible. So that the inevitable result of too 
heavy a draft upon the potential energy is death. Strych- 
nine and similar poisons bring on death simply by exhausting 
the energy of the central organs, and particularly of the cord. 

J24 Lee hires on Human and Animal Psychology 

The other tissues of the body are left uninterfered with ; and 
even the nerve-fibres, as has been shown by their severance from 
the cord, remain capable of taking up and transmitting stimuli. 
Now the removal of the brain has the same effect upon the 
reflexes as any one of these poisons, which helps to set the 
reflex mechanism in action. But it is plain that this effect must 
be differently produced. The facts are these. The nerve-cells 
of the cord, which connect sensory with motor fibres, are them- 
selves very complexly interrelated, and, moreover, send out fine 
nerve-fibres which run to the brain, and there terminate in the 
central cell plexus. These relations are represented schematic- 
ally in Fig. 17, where r r are cord-cells, functioning as reflex 

centres, and cc central brain-cells. The 
stimulus acting on the termination of 
a sensory nerve (ee) is not simply 
transmitted by way of the reflex arc to 
a motor fibre (bb'\ but is conducted to 
the higher cells (cc), and there diffused 
it may be with practically no check 
or restriction. The figure shows, 
moreover, that a single sensory fibre is 
not always correlated with a single 
motor fibre. The conduction paths 
between cell and cell are so numerous 
that every sensory fibre is connected with a number of motor, 
and even with other sensory, fibres. So that if an excitation 
arises at e, we shall expect force to be released not only in b t 
but also in b' and /. To get any idea at all of the organization 
of the central organs you must multiply these schematic connec- 
tions indefinitely. A sense-impression will result not only in 
the movement of a definite group of muscles, but in movements 
and sensations at quite different parts of the body, extending 
perhaps to a whole number of muscle groups and to several 

There can, indeed, be no manner of doubt as to the existence 
of reflected sensations, sensations occasioned not by the stimu- 
lation of their own sensory nerve, but by that of some other. 
Normally, it is true, they are very weak ; and it is only in 
conditions of pathologically increased excitability that they 

FIG. 17. 

Reflex Movements \ 2 ; 

attain to a more considerable intensity. They appear, then, to 
be essentially different from reflex movements in that they 
are never so intensive as the directly stimulated sensations, but, 
as a rule, only perceptible at all with strained and careful atten- 

Far more important is the other side of the matter, the ex- 
tension of the reflex movement due to the transmission of the im- 
pulse given by sensation to parts of the body increasingly remote 
from its origin. Heightening of stimulus-intensity will produce 
a very considerable diffusion of movement, but the quite 
universal reaction is only gained in conditions of abnormal excita- 
bility, e.g., under the influence of strychnine. Here there is no 
definite limit set to the extension of the reflex response ;. 
practically all the muscles of the body are violently convulsed 
during the action of the sense-stimulus. 

But, these abnormal states apart, we can assign definite rules 
to the diffusion of the reflex. It is uniformly dependent upon 
intensity of external stimulus. When the stimulus is only just 
intensive enough to call forth a reflex answer, the movement 
is always restricted to the group of muscles most immedi- 
ately connected with the sensitive part. If it is the retina 
that is stimulated, the resultant movement is exclusively eye- 
movement ; if it is the skin of one of the four limbs, only 
that limb moves ; if it is some portion of the skin of the head 
or trunk, there is movement in the nearest muscle group, and 
generally also in the limb which bears the closest relation to- 
the stimulated spot. Thus a weak stimulus applied to the left 
cheek leads to a contortion of the left side of the face and a 
movement of the left arm. The four limbs, which with the eye 
constitute the most motile parts of the body, are most liable to- 
be thrown into movement by cutaneous stimuli. 

If the intensity of stimulus increases, the reflex movement 
becomes more widely diffused, but is still confined for some 
time to muscles in the neighbourhood of the organ stimulated. 
Thus it may extend from one leg or arm to the other. As the 
intensity increases further, the response grows more and more 
general ; at the very highest intensity all four limbs are generally 
in motion together. The movement is at first a flexion, but is 


changed to extension at the highest intensities. This fact. 


126 Lectures on Human and Animal Psychology 

that a stimulus which affects all the motor fibres equally always 
gives rise to a movement of extension, seems to show that 
nerves of the extensor muscles are not so intimately connected 
with the sensory fibres as those of the flexor muscles. It is not 
till the stimulus becomes very intensive that the extensors are 
affected, though when once the reflex response has become 
maximal, extension tends to get the upper hand of flexion. 


In both of its two principal forms, the reflex possesses the 
characteristic of purposiveness. If the movement is restricted to 
the muscle group underlying or immediately adjoining the point 
stimulated, its result is to free the part affected from the influence 
of the stimulus. If the response is more diffuse, the action of the 
muscles is primarily auxiliary to the movement of escape ; and 
it is only in extreme cases, where a whole number of muscle 
groups are thrown into movements of extension, that all evidence 
of purposiveness is absent. 

The principal reason for this purposiveness of character is 
that the reflex movement is usually directed towards some 
end, and ceases when it has been attained. The end is contact 
with the part stimulated. If you stimulate a headless frog by 
applying a sharp point to the posterior portion of the trunk, one 
of the legs is violently moved towards the injured spot. Con- 
tact with this seems to be the end of the whole movement ; and 
contact is effected in the simplest possible way, by that limb 
and by those muscles which can mediate it with the least ex- 
penditure of effort. 

Stimulation of the eye produces results analogous to those 
obtained by stimulation of the skin. If you look at the eyes of 
a new-born child, you cannot fail to notice the fixity of their 
gaze. The eye moves, it is true, and especially if light-stimuli 
fall upon it ; but the movement is altogether irregular, and there 
seems to be no definite connection between it and the locality 
of the visual impression. This relation is only gradually built up 
If you bring a light into the visual field of a child several days 
or weeks old, you will find that it turns its eyes towards you and 
looks fixedly at the light. If you introduce two or more lights, 

Purposive ness of the Reflex \ 2 7 

it generally alternates between one and the other. Hut its gaze 
is riveted upon the light ; the eye is held to that by a kind of 
mechanical necessity, and can only leave it when its impression 
is weakened by fatigue, or when some other stimulus has 
appeared to oust it. We are in presence, that is, of a similar 
phenomenon to that of the touching of the stimulated part of 
the skin by means of a reflex movement. When a light-stimulus 
makes its appearance in the field of vision, the eye moves 
towards it, just as the hand moves towards the irritated spot 
upon the skin. 

We must notice, however, that the reflex mechanism of the 
eye is twofold. On the one hand, there is the connection 
between light-sensation and the muscles that contract the pupil 
and close the lids ; on the other, that between light-sensation 
and the muscles which move the eyeball. The lid reflex may 
be occasioned by quite weak light-stimuli, if the eyes have 
previously been in the dark. Thus the first opening of the eye 
of the new-born child to light is at once followed by a violent 
and convulsive closing reflex. But the organ very quickly 
begins to grow accustomed to the light ; and then the connec- 
tion between light-sensation and muscles of the eyeball makes 
its appearance. At first, as we have seen, the entrance of a 
light-stimulus into the field of vision merely produces move- 
ment ; we cannot say that the movement is governed or directed 
in any way. It is still only an uncertain groping for the light. 
But between the second and fourth weeks after birth some 
amount of regularity is observable. The child begins to fixate ; 
and every light-stimulus that appears within the field of vision 
excites a tendency to fixate it. Fixation consists in the as- 
sumption by the eye of the position in which a definite light- 
impression forms an image upon the most sensitive portion of the 
retina. This spot lies approximately at the centre of the whole 
tinal surface, a little to the outside of the point at which the optic 
nerve enters the eye, and from which it radiates over the retina. 
It is characterised anatomically by an extremely close packing 
of the retinal elements which take up the light-stimulus, and by 
a yellow coloration. On this latter account it is commonly 
known as the yellow spot. 

The way in which the reflexes of the eye develop, then, is 

128 Lectures ou Human and Animal Psychology 

this. First of all, light-stimuli excite merely irregular move- 
ments. After a time, these take on a definite form, and serve a 
definite purpose : the eye moves in such a way as to bring the 
image of the stimulating light upon the yellow spot. If at this 
stage a light moves to and fro within the field of vision, the eye 
follows it with an equally continuous movement. 

By what steps does this regularity emerge from the initial 
irregularity ? It is clear that the determinate relation of the 
yellow spot to the reflex movements cannot be given in the form 
of an interconnection of the nervous elements conditioning 
them. If that were the case, the first ray that fell upon the 
retina would release a reflex movement of the same character 
as those which we observe later. Now, apart from the evidence 
against this view that is furnished by the facts of experience, 
there is nothing to suggest it in the manner in which the force 
\vhich releases the movement is transmitted within the central 
organs. The force transmitted from the sensory to the motor 
fibres is dependent as regards intensity and diffusion upon the 
intensity of stimulation, and the temporary condition of the 
organs which subserve its transmission. So that there is no 


reason why an intensive stimulus in the neighbourhood of the 
yellow spot should excite only a very slight eye-movement, 
while a weak stimulus at the periphery of the retina is followed 
by an extensive one. No ! there must obviously be influences 
at work during the development of the sense which gradually 
bring it about that, while the release of a reflex movement is 
effected by the physical mechanism, its extent and direction 
depend entirely upon the place of the part stimulated ; so that 
the greater sensitivity of the organ or the increased intensity of 
stimulation can only find expression in a greater energy and 
rapidity of movement. To obtain a complete explanation of 
how this happens, we must look a little more closely at the 
structure of the sense-organs. 


The skin over the entire surface of the body is sensitive to 
stimulation. And the entire retina is similarly sensitive, with 
the exception of the place of entry of the optic nerve, where 

Development of Rejlexes of Touch and Sight, i 29 

there are none of the peculiar end-organs which subserve vision. 
But the sensations derived from various parts of the skin or 
retina are by no means entirely similar. So far as the skin is 
concerned, you can convince yourselves of this very simply. 
Touch your cheek and the palm of your hand with your finger, 
being careful to exert the same amount of pressure in each case. 
The two sensations are quite clearly different. And it is just 
the same if you compare the palm with the back of the hand, 
or the neck with the nape, or the chest with the back, or any 
t\vo portions of the skin which are some little distance apart. 
More than that, if you observe carefully, you will find appreci- 
able differences in the quality of sensations coming from portions 
which are fairly close together. As you pass from one point 
upon the sensitive surface to another, you experience a gradual 
and continuous change in cutaneous sensation, although the 
nature of the external pressure has remained perfectly constant. 
Even the sensations from corresponding places on the two 
halves of the body, alike as they are, are not altogether the 
same. If you touch first the back of one hand, and then that of 
the other, you find that there is a slight qualitative difference 
between the two sensations. 

A similar variation can be shown to exist in the retina. Fix- 
ate a piece of red paper held in the hand, and then move it 
slowly away, without allowing your eye to follow it as it disap- 
pears. Its image falls at first, of course, upon the yellow spot ; 
and then travels across the retina towards the periphery. You 
will observe that during the lateral movement of the object the 
sensation of red undergoes a gradual change : the colour-tone 
becomes darker, appears to take on a tinge of blue, and finally 
passes over into pure black. Any colour that you choose, as 
well as white, will show similar alterations. The last stage in 
sensation is always black. 

The obvious explanation of this phenomenon is, that we sense 
differently with the different parts of the retina ; and that the 
sensation gradually changes as the impression moves from its 
centre towards the outlying regions. So far as we can tell, the 
alteration takes place in just the same manner whatever the 
direction of movement ; but, and this is noteworthy, it takes 
place with different rapidity in different directions. If the object 


130 Lectures on Human and Animal Psychology 

moves from the centre outwards or upwards, the series of colour- 
tones is passed through more quickly than if it moves inwards 
or downwards ; so that a body imaged on the outside or upper 
portion of the retina looks black, while if its image falls on a 
corresponding portion on the inside or below, it is still seen as 

When you have made these experiments upon the colour- 
change undergone by small objects seen with outlying portions 
of the retina, you will ask yourselves with astonishment how it 
is that you do not always notice the differences ; that the blue 
sky or the red front of a house is not surrounded by a black 
border. Indeed, we should naturally expect that if a blue or 
red piece of paper gradually blackens, as it is brought upon the 
lateral parts of the retina, the blue of the sky and the red of the 
bricks would also show some change towards the periphery of 
the field of vision. And we might find considerable difficulty in 
answering the question, if we had not already become acquainted 
in our discussion of sensation with a large number of facts which 
point out the way in which it is to be met. A sensation, we 
must remember, is not anything determinate and invariable, but 
the product of a comparison, or, more exactly, of its associative 
relation to other sensations, which is never conditioned exclu- 
sively by the character of the single excitation, but also by that 
of simultaneous and preceding impressions. It does not stand 
alone, but is brought into relations. If these co-operative de- 
terminants are so strong as to make us, under certain circum- 
stances, see blue red and red blue, you remember the pheno- 
mena of colour-contrast, why, they may perhaps be able to 
prevent our noticing the differences in colour-tone which are 
produced by moving an object over the different portions of the 

We have every reason to suppose that the 'Sensations which 
we get from ordinary contemplation of large and uniformly 
coloured surfaces are themselves entirely uniform. For if we 
move the eye, and fixate successively different portions of the 
colour surface, i.e., bring them in order upon the yellow spot, 
we receive precisely the same colour-impression in every case. 
There are certainly differences given originally in sensation ; but 
these we have eliminated in thousands of experiences by refer- 

Development of Reflexes of Touch and Sight. 1 3 r 

ring them directly to the spatial distribution of impressions, 
the association being carried out with the certainty and precision 
of a machine. So that when we are looking at large and con- 
tinuous coloured surfaces, we simply do not notice the differ- 
ences due to the place of the impression : sensation has emanci- 
pated itself from them. 

This fact serves to illustrate what is a universal rule in the 
sphere of sense-perception, and one which we shall often have 
occasion to refer to in the future. We entirely neglect a good 
many of the different characteristics of a particular sensation 
simply because they are not directly connected with the objec- 
tive contents of the corresponding perception. If \ve are con- 
sidering a colour as produced by some external object, we take 
pains to be accurate as regards its quality. But when there is 
some peculiarity in the colour-tone which has nothing to do with 
the nature of the external impression, we only perceive it by the 
aid of special instruments, or by an extraordinary exercise of the 
attention. We can hardly doubt, therefore, that the sensation- 
differences in the different regions of the cutaneous surface or of 
the retina are really far greater than they appear in our ex- 
periments. Prepare ourselves as we may, we still tend to notice 
only those sensation-characteristics which are conditioned by 
the nature of the external impression ; the mere intention tn 
free ourselves from a rule which has been adhered to in the 
whole course of sense-development, and to which we necessarily 
and unconsciously conform, does not suffice to abrogate it. We 
must accordingly not content ourselves with the fact that a 
peculiarity of sensation, dependent upon the position of the 
portion of the sense-organ stimulated, can only be demonstrated 
in the rough or in a general way. But we may certainly sup- 
pose that such differences exist and are effective in cases where 
the deficiency of our observational methods prevents our cogni- 
tion of them. For the facts mentioned abo/e show that the 
sensation-difference must have become quite extraordinarily 
large, if it is to be apprehende 1 as a subjective difference of 
sensation, and not simply referred to local differences in the 
objective stimulation. 

Now what is the cause of these peculiar differences in sensa- 
tion ? It is plainly entirely local ; and it must, therefore, be 

132 Lectures on Human and Animal Psychology 

looked for in the structure of the sense-organs. Differences of 
tone and colour we referred ultimately to differences in the ter- 
minal organs of ear and eye. So these further qualitative differ- 
ences, peculiar to the sense of sight, must be ascribed to slighter 
variations in the structure or chemical constitution of the retinal 
end-organs. The assumption is not by any means without 
factual support. Observation shows that it is especially the 
sensitivity to red light that decreases towards the periphery of 
the retina. This is interesting, because the most frequent type of 
partial colour-blindness, as we saw before, is red-blindness. So 
that ordinary red-blindness possibly means nothing more than 
an extension of the normal sensibility of the lateral parts of the 
retina to the centre. 

In the skin, too, there are many local differences which may 
serve to explain the qualitative variation of sensation with the 
place of the portion of the organ stimulated. The bulb-shaped 
end-organs which take up tactual stimuli are distributed in 
various quantities, like the retinal elements. There are far more 
of them, e.g., at the sensitive finger-tips than in the comparatively 
insensitive back or thigh. And there are further differences in 
the thickness of the epidermis, and in the nerve-supply of neigh- 
bouring tracts of skin, which may bring it about that one and 
the same impression is differently sensed at different parts of 
the surface of the body. 

And now we have established a fact which may help us to 
answer the question raised a little while ago. We had asked 
for the conditions under which a system of reflexes at first 
entirely irregular could give rise to one of regular and uniform 
movements. We have found that the skin and the eye, the two 
sense-organs whose stimulation is the principal incentive to 
reflex movement, present structural peculiarities which imply 
definite local differences in sensation. What must our inference 
be, then, as regards sensation ? Evidently, the result will be 
identical with that which we have when a colour is recognised 
as the same colour, or a tone as the same tone. Each particular 
sensation will be recognised in terms of this attribute of depend- 
ence upon the place of the impression ; and we shall be able 
to recognise from the attribute the locality of the sensation itself 
when we have once had experience of its position. 

Development of Reflexes of Touch and Sight. 133 

This whole group of facts, then, leads us to a single conclu- 
sion : that we have in it the principal condition of the purposive 
development of the reflex. That development consists, once 
more, in this, that a movement which is at first purposeless 
comes to have a definite object, the object being the sensitive 
spot which was stimulated by the external impression that 
released the reflex. That this spot may be discovered by the 
reflex movement, it is necessary for it to be recognised in each 
particular case. And just as colour and tone are only appre- 
hended as like or different because their sensations are indis- 
tinguishable or distinguishable, so the recognition of the locality 
of an impression can only be effected by means of definite 
sensation-characteristics, characteristics which depend solely 
upon that locality. We have shown that there are such charac- 
teristics. And with that proof we have given the first condition 
for the regularity and uniformity of the reflex : it must be looked 
for in the structure of the organs themselves. At the same time 
we have not yet given an altogether complete and satisfactory 
account of reflex development. We find that the movements 
always take the shortest and simplest road to their purpose ; 
and the structure of the sense-organs cannot, of course, explain 
that. It can only tell us how it is possible for the reflexes to 
have a purpose, not how they can attain it. There must be a 
further explanation of this, to be discovered in the movement 
itself. We must, therefore, go on to ask whether any such 
explanation is admissible, and what are the terms of it. 




T~) EFLEX movements become transformed from their 
[]^ original purposelessness and irregularity, so that they 
conform to definite ends and follow definite laws. And the 
essential factor in this transformation can only be looked for in 
the movements themselves. So the question arises as to our 
measurement of muscular movement. 

When we move the legs in walking, we measure off the length 
of each step, without having to follow the movement with the 
eye. The practised pianist has acquired such skill in estimating 
the distances of the various keys, that his fingers scarcely go 
wrong by a hair's-breadth. And we possess an accurate judg- 
ment of the force of muscular movement. We discriminate the 
magnitude of different weights by lifting them. (It has been 
already shown that this discrimination is not made in terms of 
the pressure of the weights upon the skin, but by reference to 
the act of lifting, You will remember that if lifting is allowed, 
a difference of -^ can be distinguished ; while in the case of 
simple pressure upon the skin a difference of - is only just 
noticeable. Cf. above, p. 27.) So that we possess a very 
accurate measure of the force and extent of movement in the 
movement itself. And such a measure can have been acquired 
only by aid of the sensation which accompanies muscular move- 
.ment. For sensations are, as we know, the only means by which 
we receive intimation of changes, whether outside of us or within 
our own body. 


Muscle-sensation. \ 3 5 

Now, if we attend closely to our movements, we become aware 
that they are, in fact, always attended by sensations from the 
muscles. As a rule, it is true, these sensations are so weak that 
they escape our notice. It is only when we are exerting a 
certain amount of effort, e.g., moving a whole limb, that we 
observe with any clearness the strain-sensation in our muscles, 
although much less extensive movements are capable of pro- 
ducing sensations of considerable intensity, if they are frequently 
repeated, and so occasion fatigue. Fatigue manifests itself by 
a muscle-sensation, sometimes present while we are at rest, 
sometimes only appearing (or at least only becoming actually 
painful) when we move. 

The fact that muscle-sensations must be unusually intensive 
before they can attract our attention depends upon the ultimate 
character of our sense-perception. We saw above that sensa- 
tions which cannot be referred to properties of external objects 
are very easily overlooked. The local colouring of the sensations 
of sight and touch escapes immediate observation, because we 
ordinarily direct our attention only upon the place from which 
the external impression comes. In like manner, we take no 
account of our muscle-sensations as sensations> but regard only 
the perception whose instruments they are, the force and extent 
of the movement made. The sensation calls up at once the 
complex idea of which it is a constituent ; and we require special 
experimental methods, or an unusual intensity of sensation, if 
we are to become conscious of it as such. 

The sensations which accompany muscular contraction are 
probably occasioned by the pressure which the contracted 
muscle exerts upon the sensory nerve- fibres contained in it. 
But besides these sensations accompanying actual contraction 
and the cutaneous sensations of pressure and strain which are 
always connected with them, there are still other sensations 
involved in movement, whether executed or merely intended. 
Our sensations of movement are by no means dependent solely 
upon the external or internal work performed by the muscles, 
but are influenced also by the intensity of the impulse to move- 
ment proceeding from the central organ in which the motor 
nerves have their roots. This fact is most clearly indicated by 
observations on pathological changes in muscular activity. A 

136 Lectures ou Human and Animal Psychology 

patient who is partly paralysed in leg or arm, so that he can only 
move the limb with very great effort, has a distinct sensation of 
this effort : the limb seems heavier than it used to be, as though 
weightco with lead ; that is to say, there is a sensation of greater 
expenditure of force than before, although the work actually 
done is the same or even less. For the performance of this 
amount of work there is required an innervation of abnormal 
intensity. In the same way, the patient will deceive himself, 
especially in the first stages of the disease, with regard to the 
extent of his movements. His steps are short and uncertain ; 
his hand misses the objects which he is reaching for. By degrees, 
if his condition remains unchanged for a long time, he regains 
more or less precision of movement ; practice gives him familiar- 
ity with his new system of muscle-sensations. 

Sometimes this state of partial paralysis is confined to a single 
muscle group, or even to an individual muscle. In the eye, e.g., 
the partial paralysis may affect merely the single muscle which 
turns the eyeball outward, and which anatomists call the ex- 
ternal rectus. There then arises a very curious alteration of 
vision. The patient has a wrong idea of the locality of the 
objects he sees on the side of the diseased eye : he places every- 
thing farther outwards than it really is. If he tries to take 
anything in his hand, he reaches out beyond it. A day-labourer 
whose work was stone-breaking, and who was attacked by the 
disease, began to hammer the hand that held the stone instead 
of the stone itself. But in these cases, too, it has always been 
found that, if the condition persists unchanged, the patients 
gradually become accustomed to their state, and regain the 
power of accurate movement, the only abnormality being the 
feeling of greater effort in the diseased part. 

These phenomena of partial muscular paralysis render it in- 
telligible that even in cases of complete paralysis there may 
still persist the idea of active movement of the paralysed muscle. 
If a patient whose leg is completely paralysed makes a firm 
resolve to move it, he may have a distinct sensation of muscular 
strain, and consequently an idea that the leg has really moved. 
By calling in the aid of sight he can, of course, convince himself 
that he has been deceived : in the dark the illusion is complete. 
The same thing happens when it is the eye that is the totally 


v) / 

paralysed organ. The idea of an actual movement is connected 
with the inefficient resolution to perform that movement. The 
result is an optical illusion ; external objects appear to have 
moved in the direction of the purposed movement of the diseased 
eye. This apparent objective movement is evidently a necessary 
consequence of the subjective illusion. If the .eye had really 
executed the intended movement, the images cast by external 
objects upon its retina could only have retained their positions 
unchanged, if the objects themselves had moved in the direction 
of the eye and in complete accord with it. 

It has sometimes been thought that the act of will suffices of 
itself to explain these subjective movement-illusions. If I will 
to move an organ which is dependent upon my volition, it is 
said, there is necessarily connected with my resolve the idea of 
its actual movement. But it is difficult to see how a resolution 
can contain in it that peculiar sensation of muscular effort by 
which the magnitude of movement is measured alike in cases of 
partial and complete paralysis. It is surely evident that this 
sensation is a process accompanying the act of will, and capable 
of varying in degree, for one and the same volition, with vari- 
ation of its particular conditions. Under ordinary circumstances 
the sensation is demonstrably caused by the stimulation of the 
sensory nerves following the contraction of the muscle. How 
can it arise in cases where the muscle is not able to contract ? 

We must remember that muscle-sensations always accompany 
the particular volition. Hence, whenever a volition is repeated, 
the appropriate muscle-sensations will be connected with it. 
And since they are familiar to us from numberless previous 
perceptions, and are inseparably and invariably connected with 
the will-process, they will be found along with this latter even in 
cases where the muscle is unable to contract and so to furni>h 
the usual sense-stimulus. Now we give a special name to all 
those sensations and ideas which, though not occasioned by 
external, but by internal, stimuli, are yet completely dependent 
for their determination upon previous external excitation : we 
call them reproduced sensations and ideas. And we may accord- 
ingly consider these sensations of muscular effort which are the 
invariable accompaniments of volition as reproduced muscle- 
sensations, while we distinguish them from others of like nature 

138 Lectures on Human and Animal Psychology 

by the intimacy of their connection with special processes of con- 
sciousness, volitions, and by the constancy with which they 
accompany these. They will, of course, accompany them in 
cases where the influence of the will upon the muscles is 
normal and effective ; but there they will at once fuse with the 
actual muscle-sensations occasioned by the stimulus of con- 
traction. In other words, their effects can only be separately 
followed out when, as in the illustrations given above, partial 
or total muscular paralysis has disturbed or entirely destroyed 
the other muscle-sensations which are peripherally excited. 

For any historical investigation into the development of our 
sense-perceptions, the question of the origin of sensations of 
muscular effort is of far less importance than the analysis of the 
phenomena which are brought to light by their disturbance. 
The gradual adaptation to the diseased condition in cases of 
partial muscular paralysis, like those which we have noticed, 
seems in particular to be at least as instructive as the condition 
itself. It shows what sort of influence muscle-sensations may 
have exercised upon the development of the senses. Our recog- 
nition of the position of an object is normally based upon the 
sensation of effort attending the movement of the sensing organ 
to the object. If this power of localisation may be gained afresh 
after a total transformation of the whole system of muscle- 
sensations has taken place, there is not the slightest difficulty in 
the hypothesis that, when sense-perception in general was in 
process of development, the establishment of a relation between 
muscle-sensations and the place from which an external stimulus 
operates was a matter of slow and gradual growth. And this 
takes us directly back to our original problem. We set out from 
the proposition that, if reflex movements admit of accurate 
measurement, the measure can only be looked for in the move- 
ments themselves. We found the measure required in sensations 
varying with the force and extent of movement. We have now 
proved, by an appeal to experience, that it is by means of 
these movements that our limbs and organs of touch acquire 
their accuracy of function. Any alteration in the muscle- 

Sensations of Movement and Other Sensations. \ 39 

sensation does away with this accuracy. And it can only be 
regained, if at all, by a fresh course of practice. 

Our view of the development of reflex movements will, then, 
be somewhat as follows. They owe their origin in the first 
instance solely to neural connections existing within the central 
organ ; that is their only primary condition. The sensation 
occasioned by a stimulus gives rise to a more or less extended 
movement, and this in its turn to a muscle-sensation. The 
movement is, therefore, only a middle term between two sensa- 
tions : between the original sensation caused by the external 
stimulus and the muscle-sensation which results from the move- 
ment. But there is more in the whole process than this. When 
we move our limbs, it is either that they themselves may come into 
contact with the sensitive surface of the sense-organ, or that they 
may transfer the stimulus from one portion of that organ to 
another. Suppose that a stimulus operates upon the skin. In the 
movements which are aroused this or that part of the skin is 
touched; in other words, there results a second sensation of contact, 
beside that already caused by the external stimulus. And this 
naturally arises in the near neighbourhood of the other, since 
the excitation-process which underlies sensations of moderate 
intensity extends only to adjoining nerve-connections, and so 
sets in movement only the contiguous muscle groups. The 
total process now consists not of two, but of three, sensations. 
The last two of these (sensation of movement and secondary 
sensation of contact) are at first of indefinite extension. But 
very soon there comes to the front some particular contact- 
sensation, one which is similar in character to the sensation 
which stood first in the entire series, and which was directly 
occasioned by the external stimulus. And this similarity is 
obviously conditioned by contact with the place upon which the 
stimulus originally operated. We have seen that there attach 
to every portion of the skin certain local characters, by means 
of which it can be distinguished and recognised. The end of the 
movement will be then the production of a sensation of contact 
at the place upon which the stimulus operated. This end is 
easy both of proposition and attainment. For we can recognise 
not only the peculiar character of the contact-sensation, but als< 
that of the muscle-sensation corresponding to it. If we suppose 

140 Lectures on Hitman and Animal Psychology 

all this to have happened in a large number of cases, we see that 
a firm connection will have been established between the two 
sensations. So soon as a stimulus operates, and a sensation is 
aroused, the corresponding sensation of movement is awakened, 
and with it the movement, which is responsible for the final 
term of the whole series, a sensation of contact identical in 
local character with the sensation constituting its initial term. 

In the eye we find these phenomena modified by the peculiar 
structure of the organ. The nervous connections of the retina 
place it in reflex relation to the muscles which move the eyeball. 
One portion of the retina is characterised by an especial clear- 
ness of sensation. While upon the peripheral parts different 
colour-impressions are practically all sensed as the same uniform 
grey, and even as they approach the centre remain for some 
time indistinct in tone, upon the 'yellow spot' they are clearly 
and accurately distinguished. Hence the law which governs the 
development of the reflex movements of the eye : every im- 
pression, upon whatever part of the retina it fall, is brought to 
the place of clearest vision, the yellow spot. From the whole 
series of purposeless reflex movements there comes into promin- 
ence this particular one, the effect of which is to place the eye 
directly in such a position, that the stimulus can act upon the 
yellow spot. Here too, then, a definite movement-sensation, 
whose purpose is to regulate this definite movement, becomes 
connected with the local character of the retinal sensation, wher- 
ever it may be aroused. And the final term in the total process 
is always a sensation which is recognisable because character- 
ised by its relation to the yellow spot. 

But while we have been describing the development of re- 
flexes, as we observe it in experience, we may, perhaps, have 
fallen into a grave error. We seem to have been ascribing to 
the organism, at this early stage of its mental development, 
definite tendencies and purposive actions. Should not the phe- 
nomena under observation be rather regarded as subject only to 
mechanical laws ? In other words, is it not we who are putting 
purposiveness into them, while the sensing and moving subject 
itself knows no more of that than the stone knows of the inten- 
tion of the boy who picks it up and throws it ? 

We cannot, it is true, predicate ' purpose ' and ' intention ' 

Sensations of Movement and Other Sennit is us. \ \\ 

of these elementary processes of sense-perception in the > 
fication which the words possess for ourselves. Nor, a> a 
matter of fact, is anything of the kind presupposed in the above 
account of them. The processes which we assumed as necessary 
for the regulation of reflex movements are of just the same 
nature as those present in the discrimination of sensations ac- 
cording to intensity and quality. They consist in the connection 
of sensations which are excited simultaneously or in immediate 
succession by the operation of a sense-stimulus. Such connec- 
tions are termed in general associations, and are distinguished as 
simultaneous and successive. Now the unfailing characteristic of 
an association between two sensations, a and b, is this : that 
when one of them, a, is given, b is added to it, even though the 
external stimulus for b is not present. In other words, the cri- 
terion of an association is the spontaneous reproduction of one 
of the members of an association complex. Taking this criterion 
as our guide, and carefully observing the facts, we find that the 
association between definite sensations becomes stable in pro- 
portion to the frequency of its repetition. But besides frequency 
of repetition, which is involved in all the phenomena of practice 
and habituation, there is a second influence of importance at 
work in this particular case. If a sensation, a, enters into two 
associations, one with a similar or related sensation, b, and one 
with a quite different sensation, c, the resulting complexes 
possess a different character. In the association a + b, the com- 
ponents are apprehended as similar and associated ; in the 
association a + c, as dissimilar and associated. So that all 
associations of sensations may be divided, again, into associa- 
tions of similar and associations of dissimilar sensations. Two 
musical clangs, e.g., which constitute a harmonic interval, form 
a similarity-association ; they are related, as we have seen, b\ 
certain common partial tones. But two completely different 
sound-impressions, which have no elements in common, form an 
association of dissimilar sensations. If now a sensation, a, has 
become associated with several other sensations, b, c, </..., ot 
which b is more like it than c, d... are, the similarity-association 
has the advantage, other things equal : it forms more readily 
than the alternative associations. This is not difficult to ex- 
plain. The transition from a to b will be facilitated by the 

142 Lectures on Human and Animal Psychology 

existence of properties common to both ; b is partly contained 
in <7, and so is already partly present when a is present. 

Apply these considerations to the phenomena which we have 
been examining. It is at once clear that they can all be ex- 
plained as association-processes. Light-sensations, e.g., form 
associations with the corresponding sensations of the ocular 
muscles. And these have become so stable, oxving to the func- 
tional connection of light-impressions with the resultant reflex 
movements, that even if actual movement is prevented, the 
reproduced sensation of movement is still present. Or, again, 
the relation of light-impressions in the visual field to the spot 
of clearest vision is a characteristic instance of the association of 
similar sensations. This association, being preferred to all the 
other possible ones, is, of course, rendered still more stable by 
continual repetition. So that if in describing the development 
of these reflex connections just now we chanced to speak of the 
' recognition ' of an impression, of its ' being brought upon ' 
the spot of clearest vision, that must not be interpreted to imply 
that deliberation and reflection are at work, in the general sig- 
nification of these terms. It was necessary, in order to be 
intelligible, to translate certain processes of what we may call 
mental mechanics, certain association-processes, into the lan- 
guage of logical thought. Logical thinking is the form of mental 
activity with which we first become directly acquainted in our 
internal experience. And so it offers a ready means of making 
clear the connection of separate elements in a mental process, 
although the process itself may not belong at all to the sphere 
of logical reflection. But we must be careful. The logical 
formulae which we often find so useful in explaining the con- 
nection of mental processes must not be confused with the pro- 
cesses themselves. These associations, into which the processes 
of sense-perception may be analysed, form the basis upon 
which all the higher mental activities, including logical thinking, 
rest. It is always possible to put this farther back, to find it 
in the elementary processes ; or, as it might, perhaps, be better 
stated, it is always possible to translate the results of the mental 
association-mechanics into the language of logical reflection, of 
which it itself is wholly ignorant. 

Influence of Practice upon Rejlex Movements. 143 


The mental associations which we have been discussing are 
of a very simple kind. And the parallel connections of phy- 
sical processes are also of a comparatively simple nature. It is 
not at all difficult to represent the entire complex, which we 
have been looking; at from the psychical point of view, in purely 
physiological terms. In attempting this, we may, of course, 
leave entirely out of account the sensations which accompany 
the stimulation-processes in the organs of sense and of move- 
ment. It is true that we are obliged at times to interpolate 
hypothetical links in the chain of known facts. But the>e cor- 
respond well enough to familiar physical laws for us to be fairly 
certain that our physiological picture of the functions of sense, 
determined by the scheme of the reflex process, approximates 
pretty closely to the truth. 

For our present purpose we may regard the general process of 
stimulation in sensory nerves as some kind of movement. Of 
the real nature of this movement we know nothing. We will, 
therefore, not seek to define it ; we will only assume that it 
obeys the universal laws of mechanics. This movement is pro- 
pagated, as we saw, through nerve-cells to the fibres of the 
motor nerves ; it excites a greater or less number of fibres 
according to the intensity of stimulus and the degree of sensi- 
bility. The weakest stimuli are confined within the particular 
nerve-channel which is most directly connected with the stimu- 
lated sensory nerve ; stronger stimuli have more diffused effects. 
It follows that the reflex process which is set up by the stimu- 
lation of a definite sensory nerve is, in the great majority of 
-cases, kept within one definite nerve-channel. That channel 
will always be employed when reflex activity is awakened ; 
whilst it will only happen occasionally that other channels are 
occupied. It is an obvious conjecture that this preierential 
nerve-channel is the one by which a movement is conducted 
to the stimulated part ; that is, that the uniform sequence of 
events in the reflex is given with the uniform arrangement of its 
nervous connections. Indeed, this hypothesis may be looked 
upon as exceedingly probable. Wherever the mind interferes 
in the series of bodily processes, we find the conditions of its 

144 Lectures on Human and Animal Psychology 

action given in the bodily organisation. Locomotion is deter- 
mined by the structure of the skeleton and the arrangement of 
the skeletal muscles, sensation by the character of the nerve- 
endings in the sense-organs. 

Now it is a fact of common observation in external nature that 
a movement which takes place again and again in the same 
direction comes by degrees to follow this direction more readily 
than any other, and will presently be unaffected by influences 
which at first would have had no difficulty in diverting it. 
When water is poured upon the ground, it forms a channel for 
itself. Its initial direction may have been determined by the 
merest accident ; but, once determined, is adhered to, and the 
more certainly the oftener we pour. When a machine is set in 
motion, there is always the same resistance of mass to be over- 
come in its various parts ; but friction is lessened by the wearing 
and smoothing of part against part : so that a machine which 
has been going for some time usually runs more easily than a 
new one, or one which has lain for a long time unused. If you 
let your watch run down, and do not wind it up for a fortnight, 
you know that it is always liable to stop until it has been going 
again for a week or so. Now there is good evidence for the 
view that the same thing holds of neural processes. If we are 
in the habit of executing some definite muscular movement, we 
know that it gradually becomes easier, i.e., can be made with less 
expenditure of force. What we call ' practice ' consists simply 
in changes of this sort. The execution of a practised movement 
becomes easier because the stimulation-process in nerve and 
muscle is the more easily set up the more frequently it is re- 
peated. This process is originated by an increased supply of 
the elements essential to the tissues ; so that exercised muscles 
show an increase in the mass of their contractile substance. 

Observation of purposive practice puts it beyond all doubt 
that this restriction of innervation to a definite channel is a 
matter of very common occurrence. Most persons are unable 
to move certain fingers, the third and fourth, separately. But 
a little practice enables one to move either finger independently 
of the other. At first it requires a very considerable effort to do 
this ; but, as practice is continued, the separate movement be- 
comes so easy that it takes place almost of itself. 

Arrangement in Space an Association 

The course of practice in cases like these is approximately as 
follows. The first time that we attempt to move the particular 
muscle by itself, we are not entirely successful. However great 
the effort, adjoining muscle-groups are also involved in the 
movement made. With continued practice, however, this attend- 
ant movement becomes weaker and weaker, and finally ceases 
altogether. The uniform tendency in practice, then, is this : a 
larger and larger amount of the total excitation follows the 
channel of the nerve connected with the particular muscle, until, 
when the process has been often enough repeated, the whole of 
the excitation is confined to this single nerve-path. And that is 
precisely what we observe in the development of the constant 
reflexes. The only difference is, that in the present case this 
transmission of the major portion of the excitation by the par- 
ticular nerve-channel is a matter of volition and intention, while 
in the reflexes it comes about of itself, through the connection 
of sensory and motor fibres. Moreover, it is obvious that in the 
present case as well it is really not will, but a frequent repetition 
of the same physical processes in the nerves, which directly pro- 
duces the effects of practice. If it were will, we should expect 
the desired isolation of the movement to be attained at once ; 
whereas, in fact, however great the effort of will, practice is in- 
dispensable. On the other hand, when once isolation has been 
accomplished, it is not always necessary that the will should 
intervene for the production of the isolated movement. 


We have arrived, then, by different roads, at a single result. 
First of all we considered the development of purposive reflexes 
as a mental process. Its end proved to be the uniform limita- 
tion of reflex movements. Secondly, we attempted to analyse 
the process in physical terms. And here we are confronted witli 
the same limitation as the result of physiological practice. So 
that the two investigations together, the psychological and the 
physiological, furnish a concrete illustration of the principle 
which we have already found to be universally valid in the 
sphere of simple sensation-processes, the principle of psyclu 
f/tysical parallelism. But we must now return to our psycholo- 
gical problem, and ask : what becomes of the reflex movements 


146 Lectures on Hiiman and Animal Psychology 

after they have been uniformly limited in the manner described ? 
What influence have the associations which have arisen by way 
of the reflex upon the further development of mental processes ? 

We will attempt, first of all, to answer this question with refer- 
ence to the eye, because the mechanism of movement is there 
obviously the more simple. Eye-movements are from the first 
confined to the few muscles which turn the eyeball. And the 
very special sensibility of the yellow spot brings them into a 
definite relation to this part of the retina. In the skin, on the 
contrary, there are numerous tracts of equal importance for sensa- 
tion ; or, in other words, what is given only once in the eye is 
here repeated many times over. 

The reflex movements of the eyeball are, as we saw, so dis- 
posed that any light-stimulus, wherever in the field of vision it 
may appear, is carried towards the yellow spot, the place of 
clearest vision. Every light-sensation occasions a movement, 
which transmits the stimulus by the shortest possible road to 
the yellow spot, and with which the corresponding sensation of 
movement is indissolubly connected. Another result of the 
movement is to change the local colouring of the primary sensa- 
tion ; this takes on the quality peculiar to the place of clearest 
vision. The more remote from the centre the original stimula- 
tion, the greater the modification of the sensation. Now the in- 
tensity of the movement-sensation is also proportional to this 
remoteness. If I lift a weight two feet, I have a more intense 
sensation than if I lifted it only half as high. All our sensations 
of movement are intensively graduated in proportion to the 
magnitude of the movement made. The qualitative alteration 
of the light-sensations, therefore, runs parallel throughout to 
an intensive modification of the sensations of ocular movement. 
We recognise the relation of a particular light-stimulus to the spot 
of clearest vision by the local colouring which it takes on ; and 
we measure this relation quantitatively in terms of the result- 
ing movement-sensation. When a stationary light-stimulus is 
brought upon the different portions of the retina by a movement 
of the eye, the character of the aroused sensation varies from 
point to point. And every such variation is paralleled by a 
movement-sensation. And so we associate this sensation of 
movement in the most intimate fashion with the variation, sepa- 

Arrangement in Space an Association 1^7 

rating the subjective sensation-differences from those which arc- 
due to the action of an objective stimulus. This does not mean 
that we apprehend them as subjective, that we distinguish them 
as something in us from the things outside of us : there can 
plainly be no question of any such distinction in these processes 
of perception, entirely dependent as they are upon the mental 
mechanics of association. Rather are these mental processes of 
wh'ch we are treating the material from which the subject- 
object distinction is gradually built up. They are but the first 
step on the road to the conscious discrimination of the self. On 
the other hand, there is no doubt that even at this stage a well- 
defined distinction exists: the subjective differences form one 
group of sensation-qualities, the other characters of sensation 
another and a quite different one. And this is the fact which con- 
cerns us here. A series of constantly recurring sensation-differ- 
ences is brought into a relation of dependence with an entirely 
similar series of sensations of movement. At the same time, in 
saying this we postulate one condition the actual presence of 
which might be doubted, the condition that when once the eye 
has brought an impression upon the yellow spot it leaves it again 
and turns to another ; thereby, perhaps, bringing the original 
one back to the particular portion of the retina which it had 
stimulated in the first instance. (That must needs be the case, 
if it is to be possible for us to recognise at all that a visual sen- 
sation has remained unchanged.) Now there can be no doubt 
that the adult human being can move his eye hither and thither, 
to this point and that. He takes in any number of impressions, 
one after the other, just as he pleases. But can the same thing 
be assumed for that earlier stage of development at which the 
simple reflex mechanism is still undethroned ? As a in itter of 
fact, there is one influence already at work there whicii render/ 
this variation of fixation possible, and without whose prepara- 
tory operation the possibility of voluntary variation would cer- 
tainly never have been realised. I mean the influence of 
fatigue ; the weakening of light-sensation after long-continued 
operation of external stimulus. 

Every stimulus which acts upon the peripheral portions of the 
retina arouses a reflex movement, which brings its image upon 
the place of clearest vision. There the impression is retained 

148 Lectures on Human and Animal Psychology 

for some little time, until fatigue sets in and relaxes the mecha- 
nism. After this some other peripheral impression of a different 
kind, for which the retina is still unexhausted, may become the 
centre of interest, and arouse a second reflex movement corre- 
sponding to it. In this way you can see how a large number of 
external impressions may be successively apprehended at the 
spot of clearest vision. First of all will always come the most 
intensive, or that whose place of stimulation stands in the 
closest reflex connection with the yellow spot ; the others will 
follow in definite order. Now suppose that there are presented 
to the eye two luminous points at some distance from one 
another. Even if the external impressions are perfectly 
similar, the sensations which they excite will be possessed of 
a different local colouring. If the eye moves from its original 
position to another, in which the second luminous point falls 
upon the place previously stimulated by the first, the second 
sensation is made qualitatively identical with the first, while the 
latter has changed. The sensation of movement is there to mea- 
sure the distance traversed ; that is, the distance of the two 
luminous points from one another. 

Every particular connection of a sensation of movement with 
the corresponding series of local sensation-colourings is a long- 
practised association. The number of such associations is very 
great ; and they enter again into associative connections with 
each other. The sensations of movement forming a quantita- 
tively graduated series, and the local sensation-differences being 
qualitatively graduated, there arises at the same time a com- 
plete parallelism of the two associatively related sensation-series. 
And the result of this compound association-process ? We must 
anticipate a little to state it. Since this process connects to- 
gether the whole number of sensations excited in and round the 
eye, it will also systematise those sensory processes which begin 
with the simple light-sensation ; it will determine the form in 
which the eye transforms its sensations into perception. 

This form is space-perception. So that our observations, even 
at this stage, lead us to the conclusion that the perception of 
space, psychologically regarded, is not an innate possession of 
the mind, but the product of an association of sensations. It 
will now be our task to test this conclusion in detail by investi- 
gating the properties of spatial perception. 



ALL our previous considerations have been based upon em- 
pirical facts. The laws of reflex movement, the muscular 
sensibility, the local differences in sensations of light and touch, 
the exhaustion consequent upon long exposure to sense-stimu- 
lus, all these are phenomena which may be verified in experi- 
ence. But at the conclusion of these considerations we seem to 
have left the firm ground of experience far behind us. We have 
ventured upon a psychological construction of space, from the 
associative co-operation of the specified factors. Is that not more 
than experience can ever achieve ? Is not space a connate pos- 
session of the mind ? Or, if not that, is it not at least an en- 
tirely new element in our knowledge, which is sui generis, and 
therefore not a derivative from anything else ? 

It is certainly true that the perception of space is a new ele- 
ment in our knowledge. But in this sense every psychological 
fact is new which arises from some particular combination of the 
elements of our mental life. The laws of this composition are 
such that the properties of the mental resultants to which they 
give rise can never be predicted from the properties of the ele- 
ments which enter into them, although we are afterwards able to 
see the connection of these elements and their combinations in 
the complex result. Thus, e.g., after we have completed a com- 
plicated process of inference, \ve recognise that the conclusion 
follows necessarily from the premises. But still it is, as con- 


150 Lectures on Human and Animal Psychology 

trasted with the premises, something new, something which had 
to be deduced by definite acts of thought. Nothing at all is 
gained by such general assertions as that the perception of 
space is a connate possession of the mind, or that spatial quality 
is an original property of our sensations of sight and touch. 
Not only are these statements not capable of proof, but those 
who formulated them have not even taken the trouble of exa- 
mining the psychological problem before them. For a problem it 
surely remains to ascertain whether the laws of the movement 
of the eye and the organ of touch, and the sensational associa- 
tions connected therewith, exert any influence upon the percep- 
tions of these senses. 

It is a well-known fact that we are able to compare distances 
fairly accurately by means of the eyes. But it frequently 
happens that two distances which are not exactly equal are 
nevertheless regarded as such, just as in the case of simple 
sensations the perception of a difference only becomes clear 
when it has attained a certain magnitude, determined in each 
particular case by the character of the sense involved. Now 
in the present instance, just as in the sphere of sensation- 
intensity, we can determine by measurement how great the 
difference between two magnitudes must be for it to become 
just perceptible. 

We draw two horizontal lines of equal or almost equal length, 
and ask an observer who knows nothing of their objective 
relations to say whether they appear to him equal or not. If 
we begin by taking the two lines equal, and gradually lengthen 
one of them, we shall reach a point where the longer line is 
perceived to be just noticeably greater than the other. Here 
the experiment is interrupted, and the difference between the 
two lines measured. If this procedure is repeated for various 
lengths, we obtain a series of different values which tell us how 
the apprehension of differences of distance varies with the 
gradual increase of the distances compared. 

The experiment is therefore essentially the same as that 
which we made earlier to determine the dependency of sensa- 
tion upon stimulus. We have only substituted space-magnitude 
for stimulus-magnitude. If the two lines with which we begin 
are one decimetre long, and if we gradually increase one of 

Influence of Ocular Movement on 'Spatial Vision 151 

them, the difference is noticed when the increment amounts to 
about -^ decimetre, or 2 millimetres. But if the distance 
with which we set out is only decimetre, the just noticeable 
difference will be correspondingly smaller ; it will now be found 
to be y^-j- decimetre, or one millimetre. And this ratio rcm.iin> 
constant whatever other standards of measurement we may 
apply. Within certain upper and lower limits the difference is 
always approximately T ' (7 of the total distance with which we 
are dealing. Of the two horizontal lines in Fig. 18 the left 
is 26 and the right 25 
millimetres long. We 

see at once that the FIG. 18. 

f >rmer is the longer ; 

but if it is made just a little shorter, the difference is no longer 
noticed. You may convince yourselves by experiment that if 
the lines are drawn twice or three times as long, their differ- 
ences must also be two or three times as large. 

It is at once obvious that we have here the same law which 
we found to hold for the dependency of the just noticeable 
sensation-difference upon stimulus-difference. The just notice- 
able increment of spatial distance always bears the same ratio to 
the total distance. And it is plain that this coincidence may be 
explained most simply by reference to the fact that we possess 
in sensation a measure for the perception of spatial relations ; 
and that the sensations which give us this measure most directly 
are those resulting from the movements of the eyeball, the 
intensity of which must increase with the length of the path 
along which the eye travels. 

We hold before the face a box, ss (Fig. 19), open on one side 
and having a horizontal slit upon the opposite side, through 
which both eyes can look towards a white screen, zv, and see it 
without perceiving any of the other objects in the room. Now 
we hang between the screen and the eye a vertical thread,/ 
kept taut by a weight. Each eye will of itself take up such a 
position that the thread / forms an image at the yellow spot, 
the place of clearest vision. The line drawn in space from this 
point through the centre of the eye is called the visual axis 
We may say, therefore, that the visual axes of the two eyes cut 
one another or intersect in/ If we now alter the position of. 

152 Lectures on Pluman and Animal Psychology 

the thread somewhat, by bringing it nearer or removing it 
farther from the eyes, the angle formed by the intersection of 
the visual axes is changed at the same time ; for the eyes 
always follow the thread and remain directed upon it. If the 
thread is removed to a greater distance, both eyes turn out- 
wards, and the angle of intersection becomes more acute ; if 
the thread is brought nearer, the eyes turn inwards, and the 
angle of intersection becomes more obtuse. When we know 

- 'he alteration in the distance of the 


thread, we can easily determine how 

far each eye has turned round its 
centre. If the thread is moved little 
by little, the alterations in its distance 
will not be perceived at all ; i.e., the 
turning of the eyes round their centre 
is so slight that the accompanying 
movement-sensation is not notice- 
able. This sensation of movement 
only makes its appearance when the 

alteration in the position of the thread has reached a certain 
magnitude ; and then we perceive that the thread has been 
brought nearer or removed farther off. This limiting point 
must be determined in a long series of experiments, and with 
different distances of the thread from the eye. We should find 
that the eye possesses the finest sensibility for its own move- 
ments when the two visual axes are practically parallel, i.e., 
when the eyes are approximately in their position of rest. 
Under th jse conditions we can perceive an alteration of distance 
if the revolution of each eye round its centre amounts only to 
about the sixtieth part of an angular degree, or to i'. 

But so soon as the eyes have turned inwards to any con- 
siderable distance, which happens, of course, when the thread 
is brought nearer, the just perceptible movement is very much 
larger. And we shall find that the magnitude of this just 
noticeable movement increases in direct proportion to the dis- 
tance of the eye from its position of rest. 

We are plainly here only dealing with a further confirmation 
of the universal law of the dependency of just noticeable 
sensation upon stimulus. The turning of the eye inwards 

Influence of Ocular Movement on Spatial Vision 153 
brings about a definite sensation of movement. The matini- 


tude of the movement corresponds to the intensity of the 
stimulus ; the greater the movement already present, i.e., the 
greater the stimulus already operative, the greater must be the 
increase of movement or the increment of stimulus. And if the 
apprehension of sensations of movement follows the same law 
as that of the sensations of the external senses, it is to be ex- 
pected that the increment of movement corresponding to the 
equally noticeable increment of sensation will always bear the 
same ratio to the total movement already present. As a 
matter of fact, experiment proves that this relation is approxi- 
mately constant. Even such deviations as occur correspond to 
the rule which we have found to hold in the case of sensations 
of the external senses : that is, when the extent of the move- 
ment is very great, the fineness of discrimination becomes 
somewhat less than we should expect it to be according to the 
law. But the increment of movement which just suffices to 
produce a noticeable sensation amounts approximately to 
-gJjy of the total movement-magnitude. This result is in 
complete accord with what we have already obtained from the 
comparison of spatial magnitudes : a longer line can be just 
distinguished from a shorter when the difference between them 
amounts to -^ of the length of the latter. Hut if the 
perception of a spatial distance is directly proportional to the 
effort of movement made by the eye in traversing this distance, 
we must conclude that the effort of movement is the criterion 
of perception. And since we can only have know- 
ledge of the effort through the movement-sensation^ 
the influence of the latter is also demonstrated. 

These experiments on the connection of the 
sensation of movement with the estimation of dis- 
tance may be supplemented by the following observa- 
tion. We suspend two black threads, side by side 
and parallel to one another, at a little distance 
from a bright background, and fixate them with one 
eye (Fig. 20). We then move gradually away from 
them, keeping them constantly fixated as we move. ^ ^ 
Since distant objects look smaller than near ones. 
ihe apparent distance between the threads continually decreases. 

154 Lectures on Human and Animal Psychology 

until a point is reached where the two appear as one. Now 
the decrease in the size of an object as we move away from 
it is due to decrease of the magnitude of its image upon 
our retina. So that the experiment shows that there is 
a certain magnitude of the retinal image of two points 
below which they cannot be perceived as separate. This 
magnitude of the retinal image (b) or of the corresponding 
visual angle (iv) may be determined, since the distance between 
the threads and their remoteness from the eye are known. We 
find that the two images fuse to one at the moment when the 
distance between their retinal images has become so small that 
the eye has to turn only about i' in order to bring first 
one thread and then the other upon the same point of the 
retina. But that is the same magnitude as we discovered above 
to be that of the just perceptible movement of the eyeball. It 
follows, therefore, that the resting eye apprehends the distances 
of objects in space with the same degree of accuracy as that 
with which it apprehends its own movements under the most 
favourable conditions, i.e., when the movement begins with 
the visual axes parallel. The limit which it can attain to in 
the cognition of spatial distance is identical with the limit 
set to its apprehension of the sensations of its own move- 

The dependency of spatial apprehension upon sensations of 
movement, which we have inferred from these fundamental 
experiments, is confirmed by many other of the phenomena of 
vision. The muscles of the eye are on the whole symmetrically 
arranged. Thus one muscle (a\ the rectus extermis, turns the 
eye outwards, and another (b), the rectus interims, turns it inwards 
(Fig. 21). The two muscles differ but little in their dimensions, 
and both lie in a horizontal plane, which passes through the 
centre of the eyeball. Their position is, therefore, the most 
advantageous possible for the movements which they are to 
bring about. This complete similarity of conditions renders it 
obvious that sensations of movement occasioned by equally 
extensive revolutions will be of approximately equal intensity, 
whether these revolutions be made inwards or outwards. \Ye 
find the same correspondence as regards movement upwards and 
downwards. The eye is chiefly moved upwards by means of a 

Influence of Ociilar Movement on Spatial Vision 

single muscle (c], the rectus superior, 
which runs obliquely forwards in the 
upper part of the socket of the eye, and 
is affixed to the upper part of the eye- 
ball, a little outwards from the middle. 
Its action is assisted by the operation 
of another muscle, which is hidden by 
the eyeball in our figure. This muscle, 
the obliquus inferior, runs in the lower 
portion of the socket, from before and 
within backwards and outwards, con- 
necting with the posterior surface of 
the eyeball. Equally symmetrical in 
their arrangement are the muscles by which the downward 
movements are mediated. The operation of the muscle lying 
opposite to c on the lower side of the eyeball, the rcctns 
inferior, is aided by a muscle, d, the obliquus superior, which 
runs forwards and inwards, and pulls on the upper surface 
of the eyeball. Owing again to the symmetrical distribu- 
tion of the muscles, the effort of movement is approximately 
the same whether we turn the eye up or down. On the ether 
hand, there is a very considerable difference between the ar- 
rangement of the muscles which turn the eye outwards or 
inwards, and that of the muscles which turn it up or down. It 
similarity were required in this relation also, the muscles would 
have to be so placed that the rectus superior (c), which moves 
the eye upwards, and the rectus inferior, on the other side of the 
eyeball, which moves it downwards, should be inserted at the 
point where they would best subserve the movement which they 
are to effect. This is, however, as our figure shows, not actually 
the case. The direction of c is somewhat more oblique than 
that of a and b. With an equal expenditure of effort, then, the 
former muscle would move the eye a less distance upwards than 
either one of the latter pair would turn it in or outward.- 
this reason it is assisted by a second muscle. So that the cilurt 
necessary to produce a movement up or down is in general 
greater than that required for an equally extensive movement 
outwards or inwards, and, accordingly, the movement-sensations 
are more intensive ; and we must expect to find that distance in 

156 Lectures on Human and Animal Psychology 

a vertical direction will appear greater than the same distance 
in a horizontal direction. This is true as a matter of fact. If 
we draw a cross with equal arms, it will appear 
longer in the vertical direction (Fig. 22) ; while 
in other figures, such as squares or rectangles, the 
vertical distances are similarly overestimated. 


FIG. 22. 

These differences in the estimation of vertical 

and horizontal distances are the most important, but they are 
not the only errors made in measurements by the eye. Smaller 
differences of a similar character may be observed between the 
upper and lower half of a vertical line, and between the inner 
and outer portions of a horizontal line. So that, strictly speak- 
ing, no one of the four arms of the cross in Fig. 22 appears 
exactly equal to any other. These lesser differences also corre- 
spond in every case to asymmetry in the arrangement of the 
muscles. When we remarked above that the two muscles a and 
b, which move the eye out and in, differ but little in their 
dimensions, it was, of course, implied that they are not completely 
similar. As a matter of fact, b, the rectus interior, is somewhat 
more strongly developed than a, probably because the converg- 
ing movements of the visual axes predominate in all cases 
where we are occupied with what is near at hand, i.e., are fixat- 
ing near objects : so that b is exercised more than a. It may 
be observed, accordingly, that the external half of an exactly 
bisected horizontal line appears longer than the inner half ; the 
weaker muscle requires a stronger effort to produce a like move- 
ment, and the greater effort is accompanied by a more intensive 
muscle-sensation. To realise this apparent inequality, it is, of 
course, necessary to close one eye. For the outside for the 
right eye is the inside for the left ; binocular vision destroys the 
inequality. A similar difference, and one which does not dis- 
appear in binocular vision, is that between the upper and lower 
half of the field of vision. If we look closely at the cross in 
Fig 22, we see that the upper half of the vertical line appears 
somewhat longer than the lower. And to this difference, again, 
there corresponds an asymmetry of muscular distribution. The 
muscles which pull the eye down are more strongly developed 

Geometrical Optical Illusions \ 5 7 

than those which move it up, probably for the same reason 
which we found to hold in the case of the internal and external 
recti. Since the visual axes are usually directed somewhat 
downwards, and this is especially true when we are fixating 
near objects, the muscles which move the eye below the horizon 
get the more exercise, and an upward movement consequently 
involves a greater strain than an equally extensive movement 

These visual effects of asymmetrical muscular distribution on 
the eyeball allow us accurately to predict other anomalies in the 
estimation of distances, which can be experimentally demon- 
strated. You know that we are more tired if we walk a distance 
in many short steps than if we take longer and fewer ones. The 
same holds of the eye. In passing over an uninterrupted path, 
it moves with less effort than over an equal distance which is 
frequently subdivided. If we bisect a straight line, then, and 
divide up one half into numerous smaller sections, the sub- 
divided portion appears considerably longer than the other 
This experiment may, of course, be varied in all manner of ways. 
A subdivided angle appears larger than tiie same angle when 
open; a plane figure appears larger when divided up into numer- 
ous smaller areas than one which is objectively equal to it, but left 
undivided, etc. These phenomena, which can be best observed 
in geometrical figures, have been designated geometrical optical 
illusions. They are all convincing proofs of the co-operation of 
sensations of ocular movement in the act of spatial vision. 


The spatial perceptions of the sense of touch differ in many 
respects from those of the sense of sight. The difference may 
be partly due to the fact that in the normal development of 
our mental life the eye ranks as a far more perfect instrument 
than the skin, and that its particular development appears to 
precede that of the more delicate perceptions of the sense of 
touch. This does not mean, of course, that the two processes 
are sharply separated ; they rather cross one other, each in- 
fluencing and assisting the other. But at least for man and 
the higher animals vision is the earlier activity, so that ihe >ense 

I5& Lectures on H:inian and Animal Psychology 

of touch is rather guided and educated by sight than vice 

If we remember that the pressure-sensations of the skin are 
always influenced by vision, we shall see that the local relations 
which attach to them must be primarily visual But the sensation^ 
of touch are of such a character as to be able to throw off this 
influence to a certain extent. For the skin, as for tae eye, the 
peculiar property of sensation which depends upon the locality 
of impression varies from place to place. So that this locality 
may in time be recognised by the local colouring of the sensa- 
tion itself, without its being necessary to call in the eye to assist 
in the determination. When once the eye has settled the relation 
of the locality to its local colouring, we are able to refer a 
definite sensation to its right place upon the cutaneous sensitive 

It follows from this that the spatial discrimination of im- 
pressions will no longer depend upon movements, or upon the 
vividness and comparability of their accompanying sensations, 
but simply and solely upon the greater or less difference in the 
iocal colouring of sensations. If two contiguous portions of the 
skin differ indefinitely little in this respect, we shall not be able 
to distinguish the sensations proceeding from them. We shall 
only apprehend the impressions as spatially different when they 
affect portions of the skin whose sensational character is really 
different. And it is plain that this limit is not a fixed and un- 
changeable one, but that by close attention to our sensations 
we shall become able to distinguish between impressions lying 
nearer and nearer together. It is in this way that the great 
influence of practice observable in experiments of this kind finds 
its natural explanation. 

In the same manner, the differences in power of discrimination, 
which we find existing at various points on the surface of the 
skin, will depend on the fineness with which the local sensation- 
differences are graduated. These divergences are really very- 
large. On the finger-tips we can plainly distinguish as separate 
two impressions, e.g., t\vo compass points, the distance between 
which is only one millimetre ; while upon the skin of the back 
the distance must be 60 millimetres. So that the entire skin 
may be regarded as a graduated system of sensitive points 

Spatial Perceptions of the Sense of '1 ouch 159 

But these points are not arranged uniformly upon it in order of 
sensitivity, but are at various distances from each other, and 
variously distributed. Besides the natural character of th- 
cutaneous sensations, their control by the eye may contribute- 
somewhat to this graduation. Not all portions of the skin are 
equally subject to visual control ; many, like the skin of the 
back, lie entirely beyond it: others, such as the hand and 
fingers, are peculiarly subject to it. It must also be remembered 
that all portions of the skin do not naturally get a like 
amount of practice. It is again the hands, and especially the 
tips of the fingers, which are most constantly exercised ; and 
after them come the lips and tongue. On account of this 
natural difference in amount of practice, the further develop- 
ment of the spatial discrimination of the skin, which we attain 
to by voluntary practice, differs considerably for different parts 
of the organ. On the finger-tips, e.g., it is quite small, on the 
upper and lower arm strikingly large ; the power of discrimina- 
tion may be doubled or even quadrupled in the space of a few 
hours. The advantage of such practice, it is true, quickly dis- 
appears : after twenty-four hours it has perceptibly decreased ; 
after a few weeks or months it has entirely vanished. lint the 
result is not limited to the portion of the skin directly exercised. 
If, e.g., the fineness of apprehension has been doubled on the 
back of the right hand, the sensibility of the left will have in- 
creased by an equal amount, although that hand has not been 
exercised at all. The same result is obtained from all sym- 
metrical portions of the skin. At the same time, the effects of 
practice never extend beyond these symmetrical portions. \\\- 
practising the right lower arm or right cheek, we cannot i.elp 
practising at the same time the left lower arm and the .e;t 
cheek ; but there is no practice of the upper arm, or bre.i^t. or 
forehead. This peculiar result must be explained by reterence 
to the psychological processes involved in practice. In practis- 
ing, we learn to attend to sensation-differences which beiore 


escaped our notice. Now the local character of sensations of 
symmetrical portions of the skin is very similar. If, therefore, 
we have learned to attend to smaller sensation-differences upon 
the one side, we shall also have learned to do the same for the 
corresponding differences on the other. Especially with respect 

160 Lectiires on Human and Animal Psychology 

to right and left, there is complete correspondence in fineness of 
graduation and in the rapidity with which the local colouring 
alters from point to point. The case is, of course, different 
when we are dealing with asymmetrical places ; the sensations 
and their gradations are so different, that experience gained at 
one place can never be applied to another. Or at most, this 
previous knowledge can only be valuable because the attention 
in general has been rendered more keen by practice. 

\Ve have seen that the smallest noticeable difference on the 
skin is probably not determined by reference to sensations of 
movement at all in the case of the normal seeing individual, but 
is simply the result of the discrimination of local sensation- 
differences. In the same way, our judgment of the increase or 
decrease of the spatial distance between impressions of touch 
will depend solely upon the knowledge which we possess of the 
position of each impression in terms of the local colouring of its 
sensation, or, more correctly, upon the permanent associations 
into which the two are brought. But this knowledge was 
acquired with the assistance of the sense of sight. We judge 
whether a distance on the skin is longer or shorter from the 
memorial image of the position of the stimulated part which its 
sensation arouses in us. This memorial image is independent 
of the movement required to pass over the distance ; it is con- 
ditioned solely by the idea which the sense of sight has helped 
us to form of every portion of the skin as determined by its 
peculiar sensational character. And it is an obvious inference 
that the discrimination of spatial distances, whether large or 
.small, always remains unchanged so long as the sensibility of 
the cutaneous surface itself remains the same. This is the result 
which we actually obtain by experiment. If a distance of 1 1 
millimetres is just noticeably different from one of 10 milli- 
metres, we can also distinguish 21 from 20, and 31 from 
30 ; in short, for cutaneous sensibility in general, it is not the 
relative, but the absolute, just noticeable difference of distance 
which is constant. Exceptions to this rule are explicable from 
the fact that in long distances the fineness of our discrimination 
of neighbouring cutaneous points is considerably altered. 

Accidental and Congenital Blindness 16 1 


The normal development of the sense of touch, then, comes 
later than that of the sense of sight ; so that the measure which 
it applies to space is obtained from visual perceptions, and not 
at all or only secondarily from sensations of movement in the 
limbs. For this reason the mechanism of the touch reflexes and 
the laws of its development will not possess the great importance 
which attaches to them in the development of vision. Their 
influence must necessarily be diminished to the extent, that is, 
to which it is destroyed by the predominance 01 the sense of 

But this destruction is only partial. Every influence which 
makes against that of the sense of sight increases the independ- 
ence of touch, and helps to develope it to an extent which is 
never attained under ordinary conditions. Accidental blindness 
shows us striking alterations in this respect : the muscles become 
much more responsive ; the least tactual stimulus arouses move- 
ments which bring the external object into contact with different 
parts of the skin, and particularly with the most sensitive portions. 
And much greater still is the part played by tactual movements 
in those rare instances where the dominant influence of the 
sense of sight has been absent from the very beginning of mental 
development, in cases of congenital blindness. 

The congenitally blind are forced to construct their entire 
spatial world from the perceptions of the sense of touch. And 
they do this with marvellous completeness. That sense which 
remains throughout the normal life on a low plane of develop- 
ment attains a perfection which in fineness of discrimination 
may at least be compared with that of indirect vision, the vision 
of the lateral portions of the retina. In one respect only must 
the skin necessarily remain inferior to the eye: it requires im- 
mediate contact with its impressions. 

How now will the congenitally blind acquire ideas of distance 
in space or of spatially extended objects? They have at their 
disposal simply pressure-sensations from the skin, and sensations 
of movement from the exploring limbs. From these alone they 
must construct their perceptions of space. The means to this 
construction is obviously to be found, as it is in visual perception, 


1 62 Lectures on Human and Animal Psychology 

in the association of the two sensational series by the uniform 
working of the reflex mechanism. But of course this latter 
requires a much more complete development in the blind than 
in the seeing. First of all, each limb is brought into reflex con- 
nection with some definite portion of the skin. The local differ- 
ences of sensation are in consequence associated with definite 
sensations of movement ; so that there exists for each of these 
provinces of the skin some central point (although this may per- 
haps be variable) to which all neighbouring sensations are referred. 
Then, further, the separate portions of the skin are brought into 
relation with one another ; and so, by the interconnection of 
originally diverse sensational systems, the whole mass of cuta- 
neous sensations is united into a single system. This inter- 
connection must necessarily tend to be effected whenever the 
separate limbs come into contact with one another. For in this 
way there will be obtained a certain measure, however imperfect, 
of the distance between the separate organs of touch and their 

The course of this development will undoubtedly require a 
longer time than the education of the visual sense. The latter 
was completed in a single act ; but here there is required a great 
number of successive acts, whose capacity to unite at all in a 
common effect is simply due to the fact that they are all of a 
similar nature. That is, the process which gave rise to the 
space-perception of the sense of sight must be many times 
repeated for the sense of touch. Now just as we normally fixate 
with the yellow spot anything that we wish to apprehend clearly, 
so will the blind be compelled by the great difference in the 
sensibility of the various parts of the skin to make exclusive 
perceptional use of those portions which are capable of the 
finest discrimination. The parts of the organ of touch which 
possess this capacity in a pre-eminent degree are the hands. The 
blind are constantly practising their hands in touch, and even 
more in movement. Touch-sensations alone can obviously 
never suffice for the exact apprehension of spatial relations. The 
reason for this is that if .the parts of an object do not lie 
exactly in the same plane, the cutaneous pressure-sense is un- 
able to give any account of them. Hence the slight tactual 
movements of the hands, and especially of the fingers, which in 

Why are not Visual Objects Inverted? 163 

the blind are wonderfully active, are of very great importance. 
By their means the spatial properties of objects arc more 
accurately apprehended, partly through successive contact with 
the parts of the organ of touch which are capable of finest dis- 
crimination, partly owing to the continual connection of sensa- 
tions of pressure and of movement. But we always find that 
the blind do not apprehend even fairly simple spatial relations 
with anything like the rapidity witli which the perceptions of 
sight enable us to obtain an adequate idea of the most compli- 
cated figure. Their sensations of touch and movement have to 
construct the object gradually for them out of its parts. 


Thus the slow and imperfect development of the spatial per- 
ceptions of the accidentally and congenitally blind confirms our 
assumption that the sense of sight normally outruns that of 
touch. In holding this view, we are in conflict with the opinion 
which was generally current in the older psychology, and is not 
yet entirely abandoned, the opinion that the sense of sight is 
more probably educated by the sense of touch. What we grasp 
with our hands, it was said, is mure certain to our sense-percep- 
tion than what affects us at a great distance. It was forgotten 
that both objects alike make an impression upon the sensory 
nerves, and that these, in the absence of correlated psychical 
processes, can have nothing to say regarding the origin of the 

But there was an especial circumstance which seemed to sup- 
port the view that the sense of touch was necessary for the 
development of that of sight. We see objects in their natural 
position, and not inverted. But the images which external 
objects produce upon the retina are reversed. The eye is an 
optical apparatus composed of a series of curved surfaces, which 
cast upon the retina a miniature image of all objects lying within 
the field of vision. The spatial relations of this image, however, 
are exactly the reverse of those of the object itself: if the latter 
stands upon its feet, the retinal image stands upon its head, and 
vice versa. So long, therefore, as it was supposed that the act 
of vision is concluded with the formation of the retinal image, 
our vision of objects the right way up necessarily remained a 

164 Lectures on Human and Animal Psychology 

paradox. But what does the mind know of the retinal image ? 
We have only learned of its existence and its inverted position 
as regards the object from the physicist and physiologist. In 
order to be able to perceive this image as it really is, we should 
have to assume another eye behind the retina. And, as a matter 
of fact, this hypothesis has now and again been considered a 
probable one. It was never said, of course, that there is a real 
second eye ; but it was supposed that when the image affected 
the mind it was again inverted by it, just as it would be by a 
second eye, apropos of which an ingenious philosopher has 
remarked, not unjustly, that, instead of ascribing to the mind 
this perpetual business of inversion, it would be much simpler to 
stand it on its own head, so that its inversion might set right- 
again the inverted world imaged on the retina. 

From the standpoint of our own investigation of spatial vision 
this difficulty is capable of a very simple solution. It is merely 
as a series of locally-coloured sensations that the retinal image 
affects our mind. Only by the movement-sensations of the eye 
does the mind learn to connect these into a spatial order. But 
what do the movement-sensations tell us about the position of 
objects ? As the eye moves it passes from point to point of an 
external object. In moving round its centre from above down- 
wards, it passes over an object from top to bottom. It brings 
all the parts of its retinal image successively upon the spot of 

clearest vision. Now when the visible 
portion a of the eyeball moves down- 
ward in front, the yellow spot g at the 
back will be turned upwards (Fig. 23) ; 
as the front points fixates the different 
FlG> 2 , parts of the object, the point g tra- 

verses the retinal image in precisely 

the same way. So that, if the position of objects in space is 
inferred from movement, the retinal image must be inverted, 
since only where this is the case is it possible for the movement 
to correspond with the actual position of the objects. So far 
from being a paradox, the inverted retinal image is necessary for 
vision. The retinal image must have been upside down, even if 
the laws of the refraction of light in the eye had not rendered 
the inversion physically necessary. 

Why arc not Visual Objects Inverted? 165 

Of course the further question may be raised, how \vu know 
that we are moving the eye up or down. Are not 'up' and 
'down' relative notions, which presuppose the perceiving subject 
and his position in space ? It is really just because up and 
down are only relative that we are able to introduce order into 
the world of spatial vision. If we had perception of absolute 
direction up and down, we should be obliged to think that cither 
by day or by night, as the case might be, we stood upon our 
heads ; that would follow necessarily from the rotation of the 
earth. The reason that we do not think so is, that we make 
ourselves the central point in all space-references. Up and down, 
like right and left, are terms which only have a meaning when 
referred to ourselves. In distinguishing an upper and a lower 
in our spatial perceptions, we make continual reference to our own 
body: we call that 'down 'which for our eye lies in the same 
direction as our feet ; we call everything ' up ' which lies in the 
same direction as our head. 

There still remains one objection, which seems to tell against 
the influence of movement upon spatial sense-perception, which 
we have already recognised and indeed proved to exist in 
numerous cases. Do we really always move our eyes, it might 
be asked, when we wish to see things spatially ? Must we 
actually turn the eyeball up or down in order to know what is 
above and what is below ? By no means. Without moving our 
eyes in the least, we can apprehend objects as spatially ex- 
tended, and assign to each its own spatial position. How shall 
we attempt to meet this objection ? We might, as has been 
sometimes done, point to the great rapidity of the ocular move- 
ments, and our consequent inability to observe them. It might 
be assumed that though we think our eye is at rest, it is in 
reality executing very rapid movements. But we cannot escape 
the difficulty in this way : the rapidity of the muscular move- 
ments is by no means so great as we should be obliged to 
assume that it was on this hypothesis. And, on the other 
hand, we are able experimentally to reduce the duration of a 
light-impression so greatly as to entirely exclude the possibility 
of eye-movement during its operation ; e.g., by illuminating 
instantaneously with the electric spark. Under these conditions 
objects are still seen spatially. There can be no doubt, there- 

1 66 Lectures on Hujiian and Animal Psychology 

fore, that movements are not requisite for every single spatial 

But there is another point that must not be lost sight of. We 
must always distinguish mental processes from mental products. 
The latter may depend upon a capacity acquired in the course 
of previous development. It is not necessary that what was at 
first a factor in the formation of our space-perceptions, and is still 
operative to perfect and refine them, should on that account be 
a persistent and inevitable condition of all vision. The child, 
taught by its mother to make the first step, learns in time to 
run alone. Why should there not also be conditions of vision 
\vhich are operative solely, or at least principally, in the first 
stages of the development of this sense ? 

As a matter of fact, we have already discovered conditions of 
this kind. The relative position of the sensitive points of the 
retina is determined by a series of intensively graduated move- 
ment-sensations, standing in associative connection with corre- 
lative, locally coloured light-sensations. If the impressions 
once experienced are given a second time, these points can be 
recognised by their local colouring. So that if two impressions 
affect two retinal points which were on a previous occasion 
separated from one another by a movement-sensation of definite 
intensity, we shall be able, after frequent repetition of the whole 
process, to distinguish them without the actual occurrence of 
the movement and its attendant sensation. When once the 
local sensation-differences have obtained from movement-sensa- 
tions the measure of the distance which separates them, they 
retain this measure in independence of its source. A definite 
place-reference is attached to the local colouring, behind which 
its true character as qualitative property of sensation entirely 
disappears. We imagine that we perceive directly the locality 
of an impression, while in reality we are only perceiving a 
peculiarity of the sensation, and from it recognising the locality. 
And when we have extended our power of spatial discrimination 
by practice, we think that our capacity for the apprehension of 
spatial differences has been directly increased, whereas in point 
of fact it is only our ability to discriminate small sensation- 
differences which has improved. What is true for sight in this 
regard holds also for touch, only that the latter (even when it 

Remarks upon the Theory of Space-construction 167 

has attained an unusually high degree of development) stands 
in constant need of further assistance from movement-sensations 
on account of the less definite character of its local sensation- 
attributes. So that the name 'sense of feeling,' which is some- 
times used for touch, is significant. Originally we ' fee) for ' 
objects with the eye as we do with the hand. But the hand 
remains simply an organ of 'feeling,' not only because it must 
come into actual contact with the objects which it is to perceive 
while the retinal image is produced by the action of light at a 
distance, but also because it goes on to 'feel over' after contact 
has taken place; and a complete perception is only gained by 
the combination of the two kinds of sensation, pressure and 


I have attempted to describe the phenomena of spatial sense- 
perception in such a way that the theory which is to explain 
and co-ordinate them results of itself. The theory which I 
have given you is directly suggested by the facts, and does not 
attempt to go beyond them. But at the conclusion of our 
considerations, we must not omit to state that there are both 
physiologists and psychologists who still believe that they can 
dispense entirely with any such explanation of the arrangement 
of our visual and tactual sensations, or, at least, think that one 
of the factors discussed above suffices to explain all the facts. 
In the former case, it is assumed, as was uniformly done in the 
older physiology, that the spatial arrangement of our ideas is 
given directly in the arrangement of the parts of the retinal 
image ; or, as it is put to-day with more show of learning, 
though without any real improvement in the lorm of expression, 
that every sensation of the two senses with which we are he-re 
dealing possesses from the first a certain spatial quality, 
we cannot deny that this would he the most convenient 
hypothesis possible. But it is equally undeniable that it is 
totally unable to take account of all the factors which we have 
found to exercise a determining influence upon our apprehen- 
sion and estimation of spatial magnitudes. Where the attempt 
has been made to do justice to the factors in question from this 

1 68 Lectures on Hitman and Animal Psychology 

point of view, it has been found necessary to set up a number 
of artificial and complicated secondary hypotheses, some of 
which even go the length of self-contradiction. These may, 
perhaps, possess a certain value for the logician, as warning 
examples of how hypotheses should not be constructed ; but 
they are absolutely useless to the psychologist. 

The case is different when an attempt is made to furnish a 
theoretical explanation of space-construction in terms of one of 
the influences which determine the act of perception, to the 
neglect of the others. It has often been thought possible to set 
up a theory in terms of movement and movement-sensations 
alone, and either entirely to neglect the local sensation-qualities 
of the retina and skin, or to regard them as functioning in 
entire independence of movement, and as being, like the latter, 
sufficient in themselves for an adequate explanation of the facts. 
The first of these views inevitably leads to the conclusion that 
muscle-sensations as such possess a special space-quality ; the 
latter ascribes this quality to the retinal sensations alone, or to 
both of these sensation series. So that there is indirectly 
implied a return to the view that the extensive idea in general 
neither demands nor is capable of any psychological explana- 
tion. But it is not enough to grant that the spatial arrangement 
of impressions is determined on the one hand by movements, 
and on the other by properties attaching to the resting sense- 
orcan and connected with the olace of the stimulus. Ex- 

o *- 

perience shows that these two influences are so intimately con- 
nected that neither of them is operative without the other. The 
principal proof of this is found in the fact that effects which 
can only be explained from the laws of ocular movement still 
persist when the eye is at rest ; cf. the case of momentary 
illumination by the electric spark. The above-mentioned 
illusions with regard to the relative lengths of horizontal and 
vertical lines and other similar phenomena do not disappear 
when movement is prevented, although they may sometimes be 
less striking, 

If, following Lotze, we call every constituent of sensation 
which may be of influence upon the act of spatial ideation a 
Leal sign, the theories which hold that space-perceptions have 
been generated by psychological processes, and are neither 

Remarks upon tlic Theory of Space-coiistructivn 169 

given a priori nor result from a special quality of sensation, 
may be distinguished as the theory of simple and the theory of 
complex local signs. The first assumes either local signs of 
movement-sensations, or local signs of the skin, or both, with- 
out, however, in the latter case admitting the interaction of the 
two. The theory of complex local signs, on the other hand, 
regards the extensive idea as the mental resultant of intensively 
graduated local signs of the movement-sensation and qualita- 
tively graduated local signs of the sensor)' surface. Space- 
perception depends on the uniform association of these two 
sensation series, although the members of one of them need 
only be reproduced in order to be effective. This is especially 
true of the intensive series, whose terms are so intimately con- 
nected with those of the other, qualitative series, that every pair 
of definitely different local signs will be invariably associated 
with the movement-sensation corresponding to the passage of 
the o reran over the distance between them. 




WE have now shown at some length how the mind comes 
to arrange visual impressions in spatial order upon a 
plane surface. But our knowledge of the formation of the field 
of vision has given us no idea either of the nature of external 
objects, or of the visible parts of our own body. The impres- 
sions, though spatially disposed, have not as yet been brought 
into those relations in virtue of which they are arranged as 
separate ideas, each apprehended as a whole of definite spatial 
form. How does this separation come about ? How do we 
pass from the spatial perception which leaves its objects side by 
side without distinction or difference to the idea of objects 
which are spatially separate ? 

First of all, it is plainly the boundary lines of objects which 
separate them from one another, and further divide up a single 
object into parts. They afford a definite resting-place to the 
fixating eye. Whenever we have a series of objects suddenly 
presented to us, the eye is arrested by the sharpest boundary 
lines. It thus learns the rough outlines of objects first, and 
from these passes by degrees to the finer delineations of their 
parts. This influence of boundary lines on the movement and 
fixation of the eye may be easily proved by experiment. We 
hang, e.g., before a white surface a number of vertical black 
threads of different diameter, and allow an observer to look- 
through a tube towards the screen in such a way that the threads 
lie in his field of vision. Supposing that the observer knows 
nothing beforehand of the arrangement and nature of the 

The Separation of Visual Ideas \ 7 r 

threads, he will be sure to say when questioned that he *aw 
the thickest thread first, and the others afterwards, for the 
most part in the order in which their distinctness brings them 
to consciousness. And a little attention will lead him t<> dis- 
cover that in the first moment of looking through the tube the 
eye turned by a kind of mechanical necessity to the sharpest 
outline in the visual field, and then, after clearly apprehending 
this, directed itself successively upon the others in the order of 
their attraction. This order remains constant if the threads are 
hung at different distances, except, of course, that the influence 
of distance upon the apparent thickness of the threads must be 
taken into account. If two threads of the same diameter arc- 
hung at convenient distances for vision, the nearer one is 
apprehended first. But if their diameters are unequal, that will 
be first seen which makes the stronger impression upon the eye ; 
so that the boundary lines appearing in our field of vision 
determine on the one hand the movements of the eye, the 
image of the boundary line being brought upon the place oi 
clearest vision, and on the other that process within the eye 
whereby it adapts itself to the distance of the objects viewed. 
This internal process of adaptation for near objects is also a 
muscular movement attended by sensation. And the sensation 
furnishes a measure of the amount of adaptation : the nearer 
the object we are looking at, the greater the convexity of the 
crystalline lens under the action of the intra-ocular muse 

This tendency of the eye to fixate distinct points, or boundary 
lines, can only be explained by reference to a mechanism similar 
to that concerned in reflex action. Indeed, it seems to be a 
justifiable assumption that the relation of movements of tin- 
eyeball and of the muscles of accommodation to boundary line- 
and points is nothing but a further development of retle 
present in the eye from the very first. During the first days of 
a child's life, every light-impression produces an ocular move- 
ment, which brings its image to the place of clearer 
But as the retina is continually affected by uniformly diffuse 
light, that which is distinct and definitely bounded will very 
soon be separated out from this indefinite chaos of light- 
impressions, forming as it does a special stimulus, 
different from its uniform surroundings. To such a stimulus 

172 Lectures on Human and Animal Psychology 

the eye turns ; and when several are present, to each in succes- 
sion, the order of fixation being determined in every case by 
intensity, i.e., by the degree in which each stimulus differs from 
its surroundings. Even when the sense is completely developed, 
visual apprehension occurs with the mechanical necessity 
peculiar to reflex movements. And though we may voluntarily 
counteract this constraining influence, yet we are always falling 
again under its sway whenever the unexpectedness of an 
impression or some other particular reason renders an act of 
will impossible. 

The influence of boundary lines and of points is modified by 
a third factor, which is dependent upon the same conditions 
and is operative in the same way as these, namely, the movements 
of objects, by which their position relatively to one another and 
to the perceiving subject is changed. Since every object which 
is marked off from its surroundings by boundary lines forms a 
permanent whole, however its surroundings vary, it becomes 
the object of a particular idea. If objects at rest are appre- 
hended as similar units, that is only because this characteristic 
of limitation points to a separation from their surroundings such 
as is matter of immediate observation in every case of move- 
ment. In the series of separate ideas obtained from the original 
fields of vision in this way, through the medium of movement 
and boundary line, the first place is taken by the body of the 
perceiving subject, whose permanence gives it a preference over 
all other objects, while, besides that, as the substrate of all 
sensations and perceptions, it affords the universal centre cf 
relation for the spatial arrangement of the entire external world. 

The sense of sight, then, comes to apprehend objects separately 
in space by means of changes in their position and correspond- 
ing changes in boundary line. Another and a final motive to 
this spatial separation of objects consists in their reference to 
points in space at different distances from the eye. We can 
show, even in fully developed vision, how the idea of spatial 
depth originates, since the idea is of comparatively late growth. 
This is proved conclusively by the experience of the con- 

Ideas of Depth. !-- o 

genitally blind who have been restored to sight !>y a sur-ical 
operation. In such cases we find a certain power of orienta- 
tion acquired during blindness by the aid, as we may suppose, 
of the light-shimmer which is never entirely absent ; but there 
is no cognition of distance. Remote objects are not infrequent ly 
regarded as lying close at hand : the patient shrinks back from 
contact with them. We may observe the same phenomenon in 
the first months of childhood : the baby will reach for the moon 
or for the objects seen in the street through a third-story 

The development of the ideas of depth is primarily conditioned 
by ocular movements. We let our eye range from nearer objects 
to more remote, and the path over which it travels gives us a 
measure of the distance of the objects which we have successively 
fixated. For a movement-sensation is associated with every 
movement, and its intensity increases with the extent of the 
movement. When the relative distances of objects are to be 
measured, they must not, of course, cover one another. And, 
moreover, their bases must be visible. If this is not the case, 
we may quite well estimate objects at different distances from 
us as situated close beside each other. You may convince 
yourselves of this by holding before the lower half of the eye 
a small piece of paper, and so covering the lower part of the 
objects looked at. If the difference of their distances is small, 
they are generally regarded as at the same distance ; if it is 
large, you note, indeed, that one object is nearer, and the other 
more remote, but you have no approximate idea of the distance 
between them. That you notice any difference at all in these 
instances is due to the accommodation of your eye for near and 
far. Since adaptation also depends upon a muscular move- 
ment, we have a rough measure of the focussing of the eye in 
the accompanying muscle-sensation. At the same time we 
are obviously less accustomed to attend to this mechanism. 
Usually we do not employ it in measurement but make use of 
the movements of the eyeball, which are much more accurate, 
and have a far wider range. 

When we pass with the eye from the base of one object to 
that of another we usually begin with the nearer. If we arc 
estimating the total distance of any object from ourselves. \vc 

174 Lectures on Human and Animal Psychology 

begin from our own feet. So that the foot is the most original 
and natural unit for measuring distances : the length of the 
foot is the first spatial distance which occurs to us. Now when 
we are passing from nearer to more remote objects our eye 
moves from below upwards. Suppose we are standing at a 
(Fig. 24), and one eye, <?, is directed successively upon the more 
and more remote points, b, c, etc. Dur- 
ing this process it turns upwards ; the 
visual axis passes gradually from a po- 
sition where it is directed vertically 
downwards towards the horizontal, un- 
til, finally, when the object is very 

fi ^ -^ remote, it completely attains this latter 

FIG. 24. position. This movement is not confined 

to the eye : the head moves as well, es- 
pecially for objects lying directly below us, and so assists the 
ocular movement. For these head-movements, again, we have 
a measure in sensations of movement ; so that the result is the 
same, however the movements which carry the fixating eye 
from point to point are brought about. 

Since in these movements both head and eye turn from below 
upwards, remote objects seem to lie higher than near ones, and 
the horizon which bounds our view is at the same height as the 
eye. If the earth were a perfectly plane surface, we should all 
imagine ourselves to be standing in a depression ; and the 
surrounding landscape would appear to rise uniformly to the 
horizon. This phenomenon is, of course, modified in various ways 
by all sorts of inequalities in the earth's surface, and also in part 
by the spherical form of the earth. Since, again, distances 
which are objectively equal require a lesser eye-movement in 
passing over them the farther they are from us, more remote 
objects seem to lie closer together than those which are nearer. 
So that we are frequently unable to cognise a difference of 
distance in the former case which is quite obvious to us in the 
latter. If you look at the angles I, 2, which are subtended by 
the equal distances a b, b c (Fig. 24), you will see that these 
angles, which give a direct measurement of the magnitude of 
the ocular movement, become smaller and smaller, and will at 
last entirely disappear. But if we take a more elevated position, 

/(teas of Depth 175 

so that our eye h at o\ our field of vision is at once extended, 
and remote distances become visible which were previously 
hidden from our view. Near distances appear, on the other 
hand, relatively smaller than they were before. So that if we 
climb a mountain, or ascend in a balloon, all objects, near and 
remote alike, appear closer to us. 

Objects situated at different distances from the observer pre- 
sent differences not only in the relative position of their bases, 
but in a whole number of other properties. Originally these 
are turned to account only when associated with judgments of 
distance in terms of eye-movement, but later they may become 
independent indications of distance. To this class belong pri- 
marily the shadows cast by objects. Their direction and magni- 
tude are dependent on the position of the source of light with 
regard to the object, and of the object in relation to the point 
of view of the observer. Then, again, increase of distance means 
gradual decrease in clearness of boundary lines. And the more 
remote an object, the paler will be its colouring, which will 
also vary in quality according to the absorption of light by the 
atmosphere. All these factors together make up those con- 
stituents of perspective in drawing and painting which enable 
the artist, by proper distribution of contours, light and shade, 
and colour-tones, to produce upon a plane surface the illusion 
of actual tridimensional relations. 

When the distances are very great, the apparent magnitude of 
the objects comes into play as a further factor. It furnishes the 
most obvious standard of measurement when we are comparing 
quite distant objects, and in cases where the above-mentioned 
factors of perspective are absent is our sole criterion for the 
estimation of distance in the third dimension. If we compare 
a tree which is ten feet off with another whose distance is a 
hundred feet, the former appears larger, even though we know 
that the two are objectively equal. The magnitude of an object 
which is directly given in perception we term its apparent mag- 
nitude. Whenever we have learned from repeated experiences 
the actual magnitude of an object, its apparent size is employed 
as a measure of its distance from us. This is not the result of 
an act of reflection on our part, that is never involved in the 
process of perception, but is due to the direct association of an 

176 Lectures on Hitman and Animal Psychology 

idea of distance with the impression. The development of this 
idea, however, is dependent upon experience ; and we must 
therefore explain it itself as the result of an association between 
the distance of an object whose real size is known and the 
apparent magnitude of that object. The apparent magnitude 
of a person approaching from a distance will therefore excite 
directly the idea of his distance, because in many previous in- 
stances we have associated the idea of this particular distance 
with this particular magnitude by means of other characteristics 
directly given, and especially by moving the eyes from our own 
feet to his. 


The apparent magnitude of an object and the magnitude of 
its retinal image are usually regarded as directly proportional. 
The obvious reasons in support of this view are that both de- 
crease with distance, and that plainly the magnitude of the 
retinal image must be the principal condition of our having any 
idea at all of an object's apparent magnitude. If some one is 
approaching us from a distance, his image upon our retina and 
his apparent size (i.e., the size which we ascribe to him in idea) 
increase at the same time. But since this idea of apparent 
magnitude is the product of numerous associations, some of 
which are quite complex in character, we ought not to expect 
to find any constant relation between the two values, that 
of the retinal image (physiological) and that of the idea (psy- 
chological). This presumption is confirmed by experience. For 
we find that while the magnitude of the retinal image remains 
constant, or, what is the same thing, that of the visual angle 
formed by drawing lines of vision to the boundaries of the 
object, the apparent magnitude may be extraordinarily different, 
owing to its determination by the other factors which enter 

o * 

into the association. Foremost among these stands the idea 
of distance suggested by other characteristics of the object ; and, 
in the second place, there comes into account the idea of the 
magnitude of similar objects. 

The most striking instance of our seeing the same object at 
the same distance sometimes as larger, sometimes smaller, is 

Apparent Magnitude and Distance 177 

the one afforded by the sun and moon. At the distance which 
separates the sun from the earth the si/e of its image cannot 
differ at morning, noon, or evening ; its magnitude remains equal 
at all times of the day. But when the sun is at the zenith it 
appears smaller than when it is on the horizon at rising or set- 
ting. This is explained in the following way. We form a definite 
idea of the distance of the sun, though, of course, the idea is 
very far from the truth. The sky seems to our eye a solid arch, 
which rests upon the earth at the horizon, and closes down upon 
the nearest mountains or the towers of the nearest town. To 
frame an idea of the distance of the sun at the zenith, we have 
at most only a tower or a mountain as our standard ; to get 
an idea of the distance of the horizon, we make use of every 
object within our field of vision. Between ourselves and the line 
of the horizon we see large numbers of trees, fields, villages, and 
towns : and a distance which contains so many objects must 
of course be very large. So we come to imagine that the 
horizon is farther off than the zenith ; the arch of the .sky which 
rests upon the earth is not semicircular, but is rather shaped 
like a very convex watch-glass. But if our retinal image is 
equally large whether that which we are looking at is near by 
or far off, the magnitude of the object regarded must be dif- 
ferent in the two cases. The remoter object looks to be actually 
larger, just because it seems to be of the same size as the nearer 
one. It is as though a man on a steeple should appear as 
large as one by our side ; we could not but imagine that the 
former was a giant. Before we form an idea of the magnitiuii 
of an object we always consider the distance at which we viev 
it. Quite frequently we mistake the distance. But though we 
may have convinced ourselves of this error a hundred times, we 
cannot free our perception from it, so stable are tiie associations 
through which it has arisen. Our perception of the size of the 
sun rests upon two wrong ideas : in the fh>t place, we suppose 
that it is not much farther off from the eye than the nearest 
mountain peak or the top of the neighb Hiring church spire ; 
and secondly, we imagine that it is sometimes nearer, sometimes 
more remote, according as it stands at the zenith or approaches 
the horizon. We need not be astronomers or physicists to know 
that both these notions are false. But however well we may 


ijS Lectures on Hii))ian and Aniuial Psychology 

know this, and however sure we may be that our distance from 
the sun does not become alternately greater and less, we still 
make the same mistake, the astronomers and physicists among 
us no less than the ordinary man. 

Our perception of objects is, therefore, always dependent 
upon their distance ; not, however, upon their actual distance, 
but upon the distance as we imagine it. If we could obtain a. 
perception of the actual distance of the sun and the moon, they 
would appear infinitely large to us. On the other hand, when 
\ve try our best to imagine them quite close, they appear smaller 
than usual. If we look at the moon through a tube, or through 
the closed hand, seeing nothing but that portion of the sky 
where the moon is situated, it will seem no larger than a half- 
crown, whereas it generally looks about as large as a plate. 
The simple explanation of this fact is that we do not now 
localise the moon somewhere behind the trees which fill the 
foreground of our normal field of vision, but close behind the 
tube or closed hand. In the same way, when we look at the 
moon through an ordinary telescope, it seems smaller, and not 
larger, than usual, though the telescope magnifies, and we can 
see by its aid a number of things upon the surface of the moon 
which are invisible to the naked eye. That is also because the 
moon is not localised at a distance, but at the end of the tele- 
scope. The same thing happens when we direct our telescope 
upon distant mountain peaks : we see their outlines more clearly ; 
we observe details which the unaided eye could not distinguish ; 
and yet we note that on the whole the mountains do not appear 
larger, but smaller. In these cases the magnitude of the retinal 
image of the moon or of the mountain is increased, yet we see 
the objects themselves as smaller. 

But we have not even yet completed our account of the 
influences which are here at work. If we look at a man on the 
top of a tower, he does not appear nearly so small as he should 
do in accordance with our idea of his distance. When we look 
at the mirror on the opposite wall, we estimate its distance 
pretty accurately. But we see it larger than we really ought to, 
if we compare the size of its retinal image with those of other 
and nearer objects. Clearly, the fact that we already know the 
size of the man and the mirror is here of importance. We have 

Apparent Magnitude and Distance \ 79 

seen men close at hand thousands of times, so that we know 
certainly that there never was a man only a millimetre high, nor 
a drawing-room mirror only two centimetres square This ex- 
perience exerts an influence upon our perception, and serves to 
modify the idea which we should otherwise have formed regard- 
ing the distance of the objects we are looking at. This cor- 
rection is, as you know, not complete : the man on the roof is 
much smaller to view than the man by our side, and the mirror 
on the wall twenty feet away a little smaller than it is when we 
stand directly before it. There is a kind of conflict between the 
fact that the object viewed is at a distance, and must, therefore, 
appear smaller, and the fact that we are acquainted with its 
true magnitude. Both sides are, as a matter of fact, right in 
this controversy ; but since it is not possible to grant the claims 
of both at the same time, we follow the example of that most 
excellent judge who decided all lawsuits involving money by 
dividing the sum between the two parties in the suit. 

Our perception, then, can only determine the true magnitude 
of objects where this true magnitude is actually known to us ; 
and this knowledge must come from direct and often repeated 
experience. However sure we may be that the moon is im- 
measurably larger than a plate, we shall not on that account see 
it a whit larger. \Ye are convinced that a magnifying-glass 
does not make the objects seen through it any larger; and yet 
they continue to be larger for our vision. The sun at midday, 
we are certain, is not smaller than it is in the morning, yet as 
we look at it it appears smaller. Vision requires to be con- 
vinced in a quite different way. No assertion on the part of 
other people, no speculation or calculation, is of influence in 
determining our perception, but only an association of ideas 
repeated over and over again. Isolated experiences, therefore, 
make no impression upon our minds. From a window in my 
room I look directly upon a neighbouring church tower. The 
face of the church clock appears about as large as that of a 
moderately large clock which hangs upon my wall. The ball of 
the steeple looks about as large as the button of a flag-staff. A 
little while ago the clock face and steeple knob were taken 
down for repairs, and lay upon the street. To my astonish- 
ment, I saw that the former was as large as a church door, and 

i So Lectures oti Human and Animal Psychology 

the latter as large as a waggon- wheel. Now the two are in 
their places again and look to me just as they did before, al- 
though I have learned their true size. The workman upon the 
roof does not seem so much smaller than he actually is, because 
I have observed the size of my fellow-men hundreds of times. 
But the ball of a steeple and a church clock are not objects of 
every-day experience. The button of a flag-staff and the clock 
on the wall are much more familiar. And so I think of the 
steeple knob as the flag-staff button, and the church clock as a 
wall clock. Even that idea seems exaggerated, if I compare 
these things with objects more immediately around me. For I 
can just cover the ball of the steeple with the head of a pin, and 
the tower clock with my watch, if I hold these at a little dis- 
tance from my eyes. If it were not altogether too improbable 
that the steeple should carry a watch and have a pin's head as 
its ball, I should perhaps imagine that that was the actual state 
of affairs. 

We see, then, that our perception of things in space is extra- 
ordinarily variable ; that it is conditioned by a number of 
influences which by no means emanate from the objects them- 
selves ; that we take into account the apparent magnitude of 
objects, their distance from us, and finally our experiences of 
the same or similar objects in other connections. How, then, 
can we assert that our perception is determined by the objects 
outside us ? All these influences are not found in the objects, 
but in ourselves. It is we who involuntarily and unwittingly 
alter the phenomena, in terms of the ideational elements which 
are already present in consciousness, ready at a moment's 
notice to form associations. And this whole variability of our 
world of perception depends primarily on the idea of depth, 
which gives to the spatial arrangement of the visual field its 
property of being apprehended as at different distances from 
the observer. Such a property must necessarily open a wide 
field for the play of the most diverse subjective and objective 
influences upon our spatial ideas. 

But although the arrangement of objects in terms of spatial 
distance, under the operation of these influences, must always 
remain imperfect and incapable of exact measurement, we must 
not forget that through it alone do we obtain a final form for 

Apparent Magnitude and Distance i8r 

our world of ideas. With the reference of objects to clitT'.-rent 
distances in space, the world of perception is placed outside of 
us, and is differentiated into an infinite diversity of content. 
Although the spatial relations which we ascribe to external 
objects may at the outset often be incomplete and deceptive, 
still the decisive step has been taken with the very introduction 
of those relations. The ceaseless activity of our sense-percep- 
tion is constantly at work in the endeavour to perfect our ideas. 
It furnishes us with new ideational groups, and corrects the 
most serious errors in those already acquired. All the sense> 
co-operate in this work, each revising and supplementing the 
others. But it is primarily the common action of the two co- 
ordinate organs of vision to which we owe the greatest part of 
our ideational development. There are no other organs which 
so directly supplement and correct each other's perceptions, and 
which thus give so great an impulse to the fusion of separate 
perceptions into a single idea, as the two eyes 



I "HE two eyes may oe compared to two sentinels who, 

viewing the world from different standpoints, impart their 

experience to each other, and so complete in idea a common 

picture, uniting in itself all that each observer has seen 


It is not very long since the fact was discovered that bino 
cular vision was different from monocular. The early physio- 
logists thought that the image of an object produced in one eye 
was not different from that apprehended with both. And it 
was accordingly supposed that the two eyes were simply equi- 
valent to a single eye, a conclusion which found an apparent 
confirmation in the anatomical structure of the optic nerves. 
At a certain point in their course after leaving the brain, de- 
cussation occurs. At this point the nerve-fibres are closely 
interlaced ; and then again two nerve-trunks appear, one ex- 
tending to either eye. It was supposed that at the point of 
decussation and interlacing of the two nerves the fibres were 
divided. Each fibre, from whichever side of the brain it came, 
was thought to divide in such a way that one of its parts ran 
to each eye, and within each eye to correspondingly situated 
retinal points. It was in 1840 that the English physicist 
Wheatstone proved that the images cast upon the retina of 
each eye are very frequently dissimilar, without there being 
occasioned any disturbance of vision. If we hold an object 

close before us, and shut first one eye and then the other, we 


Binocular Vision i&; 


see it a little differently in the two cases. Suppose, e^., we 
hold our hand between the eyes, at a little distance from them. 
so that the surface of the hand is at right angles to the face ; one 
eye sees only the hack, and the other the palm, of the hand. 
If the point of anatomical decussation were really a place of 
division, and if the images cast on the two eyes were directly 
intermixed in the brain, simultaneous binocular vision would 
only give us a confused picture. For on a portion of one retina 
there is represented part of the back of the hand, on the corre- 
sponding region of the other part of the palm. These two 
images would therefore be superimposed in the common act <>f 
vision, and that would render any clear apprehension impossible. 
But observation by no means confirms this view of the matter. 
It is rather the fact that we see the hand more perfectly with 
two eyes than with only one. It is not merely that we see- 
simultaneously much that in monocular vision could only be 
apprehended successively ; but we perceive directly that the 
hand is not a picture painted on a surface, but has extension in 
the third dimension. The same test may be repeated with the- 
most diverse objects ; we shall always find that the apprehen- 
sion of the third dimension of objects is intimately connected 
with simultaneous binocular vision. If we are using simply 
one eye, we are often unable to decide whether the fixated 
object is tridimensional, or merely a drawing upon a plane sur- 
face. So that in monocular vision illusions are possible in this 
connection ; drawings in perspective and light and shade in 
particular may give a very strong impression of the third 
dimension. If the object is near to us, the illusion vanishes at 
once so soon as the second eye is opened. Although, there- 
fore, a perception of the third dimension may be gainec: 
a single eye, it is always less complete, instantaneous. an<i 
immediate than that given in binocular vision. As a rule we are 
only able in monocular vision to attain to a perception of the 
third dimension of objects by degrees, and in terms of the move- 
ments made by the eye in passing from a nearer point to one 
more remote, i.e., by a series of acts following one another in 

If, then, the immediate idea of depth is always cornice! 
with simultaneous binocular vision, it seems obvious to say that 

184 Lectures on Human and Animal Psychology 

we see objects in this way more adequately just because the 
images cast upon the t\vo eyes are different. We have a direct 
perception of extension in the third dimension because the two 
eyes look at things from different points of view. And this 
fact is, moreover, confirmed by observation. When we remove 
the object farther and farther from the eye, the perception of 
depth disappears. But with increase of distance the difference 
between the retinal images grows less. And at last, when the 
object is so far off that the distance between the two eyes is 
practically zero in comparison with it, the two images are pre- 
cisely alike, and fall upon correspondingly situated portions of 
the two retinae. If, e.g., we hold a sheet of paper close before 
the eyes, so that the right eye sees one side and the left the 
other, we get a clear idea of its extension in the third dimension. 
But if it is removed farther and farther off, we come to see less 
and less of the two sides ; and at last perceive nothing but the 
front edge, which is just the same for one eye as it is for the 
other. In other words, perception of depth and difference 
between retinal images always run parallel to each other. 

If this difference between the retinal images of the two eyes 
is the cause of the perception of depth, it is obvious that this 
perception may be induced without any actual vision of a 
tridimensional object, simply by presenting to the eyes retinal 
images having differences similar to those produced by the per- 
ception of such an object. If, that is, we cast upon one retina 
an image which looks like the back of the hand seen obliquely, 
and upon the other an image which resembles the palm of the 
hand seen under similar conditions, an idea of extension in the 
third dimension will arise in our minds, although the images 
employed are simply drawings upon a plane surface. The 
retinal images are precisely the same as they would have been 
had we looked at an actual hand ; the result as well must there- 
fore remain unchanged. 

It is quite easy to test this. And it is best to take for the 
purpose objects of fairly simple form. Suppose that we are 
holding before our eyes a truncated cone with circular base, the 
apex turned towards the face. First, we close the right eye, 
and draw an exact picture of the cone ; now we close the left, 
and make a similar drawing. The two pictures are different, 

because the right eye sees parts of the object which the left 
does not, and vice versa. The left eye sec-, the cone appn.xi 
mately as A, the right us B (Fig. 25). Neither of these view, 
as mere drawing furnishes an)- incentive to an idea of the third 
dimension. The most that we can do is by an effort of imagi- 
nation to see the small inside circle at will either nearer or more 
remote than the larger out.->ide one. But if we let A affect the 
left eye as though it were an image proceeding from an actual 
cone, and B act similarly on the right eye, we have ju>t as 
definite an idea of tridimensionality as we gain by observing 
the cone itself. 

It will not, of course, do to view the two drawings with both 
eyes directed quite at random. \Ve must look at them in a 
way which corresponds to tue formation of images by actual 

Fin. 25. 

objects. The left eye must fixate the smaller circle in A, the 
ri^ht the smaller circle in B. Only under these condition* are 

o . . 

the two images in the eyes related as they would 

fixated the top of a real truncated cone. But the experiment is 

not altogether easy. We are accustomed to direct both eyes 

upon one and the same point. Here we have to fixate a 

different point with each, the tup of A with the left, the top of 

B with the right. Only long and continued practice can en 

us to command our eye-movements to the extent required lor 

independent fixation by either eye. Normally the inoveni 

of the two eyes are completely concurrent. 

themselves are determined by external impressions ; an- 

probable that these were also originally instrumental 

ing about the functional concurrence. For, as we have 

is! law of the reflex mechanism of each eye that out 

'are i- 

1 86 Lectures on Human and Animal Psychology 

always attracted by distinct points or boundary lines, and 
moves from one to the other of these according to the intensity 
of the impression to which they give rise. Since both eyes 
follow the same law, their movements must necessarily be 
closely interconnected ; the point which leads one eye to fixate 
it will also arrest the other. Thus there arises an impulse to 
common fixation on the part of the two eyes, which can only 
be overcome by practice. 


To obviate this difficulty, which confined observation to a 
few practised individuals, Wheatstone constructed the stereoscope. 
By means of this instrument any one may readily obtain an 
idea of the third dimension from representations upon a plane 
surface. The ordinary form of the stereoscope is that given it 
by Brewster. It contains two small angled prisms, behind 
which at a little distance are placed the drawings to be com- 
bined. In free vision the eyes must have their axes parallel in 
order to fixate the drawings b simultaneously. But if the 
n prisms/ are introduced, and their refract- 

ing angles turned towards each other, the 
rays coming from the drawings b will be 
diverted in such a way that these fall 
upon the place of clearest vision and 
adjacent parts of the retinas, although 
the two eyes do not fixate the drawings 
b, but the point F. The necessary result 
then is that the inner circles of A and B 
(Fig. 25) affect coincident points of the 
two retinae, and the remaining portions of 
the figure exhibit precisely the same differences in the retinal 
image as arise when we directly observe a real object of similar 

The following are the simplest stereoscopic experiments. 
The perception of depth will arise if two vertical lines at 
different distances from one another are presented to each eye 
in the stereoscope ; e.g., the lines a b to the left eye, the lines 

77/6' Stereoscope 187 

c d to the right. We obtain in this way a common image <f 
two lines, the first of which, I, is due to the fusion of a am: 
c, the second, 2, to the fusion of b 
and d. The former lies in the plane 
of the paper, the second at some 
little distance behind it. This is in 
accordance with what would nor- 
mally be the case. When we fixate 
binocularly two lines the right one 

i* IG* 7* 

of which is farther off than the left, 

the horizontal distance between the two lines in the retinal image 

of the right eye is necessarily greater than in that of the left. 

The idea of depth arises in the same way when we present 
to either eye a line drawn somewhat obliquely, and make the 
inclinations of the two lines a little different. If the lines / and 
r, which fall upon the left and right eye respectively, have tin- 
inclinations indicated in Fig. 28, 
we obtain a common image, s, 
which extends into the third 
dimension, its upper extremity 
being more remote than its 
lower. If, on the other hand, 
the lines are inclined as in FIG. 28. 

Fig. 29, we obtain a common 
image whose lower extremity is more remote than the upper. 

These two cases in which oblique lines are differently inclined 
or the horizontal distance between vertical lines is difierent, arc 
realised again and again under the conditions of tridimen^ional 
vision with the naked eye, and also in the stereoscope, 
constitute the two fundamental ex- 
periments of stereoscopic vision. The 
vertical or oblique lines need not be 
straight ; the result is the same if they 

o * 

are somewhat curved. All stereoscopic 

vision depends ultimately upon the 

combination of these two 

experiments. On the other hand, we 

can never obtain the idea of depth if hori/.onta. line 

different distances are presented. This is very easily explained 

.'88 Lectures on Human and Animal Psychology 

when we remember that no such condition of tridimensional 
vision exists in nature. We may turn and twist an object as 
we will ; its boundary lines are always either vertical or oblique. 
The facts of stereoscopic vision prove indisputably that the 
two eyes perceive independently of each other, and that their 
perceptions are only secondarily combined in a common idea. 
Any other view as to the causes of stereoscopic phenomena 
becomes involved in inevitable contradictions. It is, e.g., abso- 
lutely impossible to maintain that the two eyes are really only 
one ; that every nerve-fibre divides into two branches, which run 
to exactly corresponding points of the two retinae. If this were 
so, the common image which we obtained from our truncated 

cone would be of the character indi- 
cated in Fig. 30, where those parts of 
the drawing which do not fall upon 
corresponding retinal points simply 
cover one another ; but there is no 
hint of the origin of an idea of a 
simple tridimensional object. 

If we admit, as the phenomena in- 
evitably compel us to do, that the two 

eyes are separate organs of vision, which sense independently of 
L-ach other, we can only look for the fusion of the two visual 
perceptions in some psychological process. As a matter of fact, 
the phenomena themselves lead to this conclusion. We see 
that the idea of depth only arises when the two images exactly 
correspond to the views which we have of an actual body in 
space ; and we find that the direct perception of the third 
dimension always implies binocular vision. Suppose now that 
the two plane "pictures were laid separately before you, and you 
were told that they were two projections of one and the same 
object, what inference would you draw as to the nature of this 
object ? You would say, of course, the object is extended in 
three dimensions ; and would even possess a tolerably correct 
idea as to the extent of its third dimension, and perhaps be able 
to construct an exact model of the whole object. If the per- 
ceptions of the two eyes are originally two separate things, it 
must be by an essentially similar method of procedure that we 
come to fuse these senarate areal imacres in the common idea of 

'Ike Stereoscope i>.g 

an object extended in the third dimension. We, to<>, must con- 
struct our idea of the model of the object from its areal projec- 
tions. The only difference is that we do not do this consciously, 
but unconsciously and involuntarily, by an act of sensational 
association : so that it is only the result, the idea of the object 
itself, which appears in consciousness. 

The necessity of fusing perceptions in the idea of a single 
object rests in part in the infinite number of these perceptions. 
Again and again there are presented to our two eyes corre- 
sponding and supplementary areal projections of tridimenMonal 
objects. Always and invariably must we perceive such objects 
from the different points of view of the external world obtained 
through our two eyes. But this constant motive to connect the 
two partial ideas is only half the matter. Another and stronger 
influence is found in the mental endeavour which dominates all 
perceptive processes, the effort to obtain a permanent associa- 
tion of simultaneous ideas and ideational elements. We have 
found this endeavour operative in the processes of areal and 
tridimensional perception of which we have already spoken. 
There can be no doubt that the fusion of the two visual images 
is the result of an act of mental association. But we '. 
still to determine more precisely how this association takes 

When dealing with the formation of the perceptions of the 
single eye, we found that sensations of movement hirnish a 
measure of the spatial distance of separate points in the visu 
field. Similarly in the binocular idea of depth it is scnsatio 
of movement which furnish our primary measure of spatial 
tance. If the common field of vision contains a single bri-iit 
point, the reflex mechanism which governs the relation 
ocular movements to the yellow spot brings about the 
of this point by both eyes. Its image fails on the yellou- -o-j 
the place of clearest vision ; i.e., it is the point of inters: 
the visual axes. If other bright points appear in the common 
field, they are successively apprehended in the order in win 
their intensity stimulates the tendency of the eye to men- 
there results a successive fixation of the distinct p > 
boundary lines present in the field of vision. But there \ 
necessarily arise at once important differences between 

Lectures on Human and Animal Psychology 

various cases in which the eyes mark out an object in this way 
point after point. If the points which the eye passes over lie on 
a plane surface, the image of the points which are no longer 
fixated, whose image, that is, falls not on the yellow spot, but 
on the lateral portions of the retina, still affects retinal points 
of approximately coincident position in the two eyes. With 
this coincidence of position there is also given a certain simi- 
larity of the peculiar sensational colouring, dependent upon the 
place of the impression. If, on the other hand, the points suc- 
cessively fixated lie at different distances from the eye, the 
image of the point which is no longer fixated does not fall upon 
points of coincident position and analogous sensational charac- 
ter in the two retinas. And this divergence will be greater the 

o o 

greater the distance in the third dimension which separates the 
points. There will necessarily be an essential difference in the 
Association of the two visual images in this case ; and the two 
series of actual experiences, corresponding to the perceptions of 
surface and of depth, will be clearly distinguished from one 


We must, however, ask how this distinction can give rise to 
the peculiar idea of a tridimensional object as opposed to an 
areal object. On this point different opinions may still be en- 
tertained. It is certain, say many authorities, that only images 
which fall upon coincident retinal points are seen as simple; all 
others are seen as double. We possess, therefore, a sure criterion 
of the presence or absence of the third spatial dimension in the 
presence or absence of double images. And the distance which 
separates tnese, i.e., the magnitude of their deviation from a 
coincident retinal position, allows us to infer directly the 
magnitude of extension in the third dimension. So that the 
.perception of depth consists simply in the neglect of double 
images ; and the idea of depth arises more clearly, the more 
there is to neglect in order to attain to the perception of a 
single object. 

This view cannot, however, stand the test of experiment. If 

7 hcory of Stereoscopic I'ision 191 

Fig. 31 is introduced into the stereoscope, the left eye receives 
the image A, the right eye that of B. The lines i and 2 fill 
upon corresponding parts of the retina, 
lines I and 3 upon differently situ- 
ated portions. The result is that the 
two heavily drawn lines i and 3 are fused 
to a single idea, the line arising from 
their fusion giving a clear perception of 
depth, while the fainter line 2 crosses it 
in the plane of the paper ; that is, our i < -.. 31. 

vision has combined into one the t\vo 

lines which fall upon differently situated portions of the retin;e, 
while the two lines which fall upon corresponding parts arc- 
perceived separately. It follows necessarily from this that the 
idea of depth cannot arise from the apprehension and subse- 
quent neglect of double images. Were that the case, it would 
not be possible for the images of I and 2, which are cast upon a 
series of corresponding points, to appear separately as double 
images. This experiment also proves therefore that the forma- 
tion of a visual perception is an ideational act based upon an 
association of two visual perceptions, and determined not only 
by the position of the retinal images, but also by other proper- 
ties which these images possess. The two strongly drawn lines 
force themselves first of all upon perception, and they alone can 
be referred to a single object when the images of the two eyes 
are compared ; while the object itself must extend in the third 
dimension of space to the position occupied by the images. 

The idea of depth is, therefore, not produced by disregarding 
or by intensively weakening the separate perceptions in the 
common act of vision, but rather by the clear apprehension of 
them and their subsequent combination, this primary associa- 
tion beinsf further associated with other similar ideas. I he 


differences of the two retinal images are by no means to be 
disregarded as worthless inaccuracies. They give u>, on the 
contrary, an extraordinarily exact measure of the spatial 
qualities of external objects. And the inference is unavoidable 
that if we perceive these spatial properties through theditfennc 
of retinal images, these must themselves be given in p.-rceptior 
Avith their characteristic spatial differences. 

192 Lectures on Human and Animal Psychology 

But some doubt may still remain regarding the manner in 
which these differences of the retinal images, whose comparison 
gives us the idea of depth, are themselves apprehended and 
elaborated into that idea. We must set out from the fact that 
the movement-sensations of the eyeball which inform us of the 
spatial relations of the areal field of vision furnished also, at 
least, originally, a measure of distance in the third dimension. 
The hypothesis then appears probable that the idea of depth has 
been generated by movement. We have already discussed the 
signification of eye-movement for the estimation of distance. 
If we are fixating an object binocularly, any increase or 
decrease of its distance from us is very distinctly perceived by 
means of the movements of convergence or divergence, which 
are made by both eyes in keeping the object constantly fixated. 
We become conscious of these movements by movement-sensa- 
tions, and by these latter we measure the approach or removal 
of the objects. If a spatially extended object lies before us, it 
simply presents to us simultaneously what is perceived succes- 
sively when an object moves. At the same time, though the 
tridimensional object lies before us as a whole, we can only 
clearly apprehend some portion of it at any single moment. In 
this case also, then, we pass gradually by ocular movements of 
convergence and divergence from nearer to more remote points, 
or from more remote to nearer. In this way we perceive what 
is nearer or more remote in the object, just as we observe the 
changes in the position of a single point when it is in motion. 

It can hardly be doubted that the idea of depth originally 
arose in this way by a succession of sensations and perception^. 
But it is a different question whether it continues to arise in the 
same manner, whether each single idea which has been gradually 
acquired by both eyes continues to be formed by a series of sue-' 
cessive acts. We have already discussed a similar question in 
our investigation of the perception of plane surfaces. There, too, 
movements play an important part. But we saw that move- 
ments are by no means continually operative in every single 
perception, but that the resting eye is itself able to see things in 
space, and possesses a fairly accurate measure of spatial exten- 
sion. And we found that it was the presence of a local colour- 
ing in visual sensations which enabled the eye to dispense with 

theory of Stereoscopic V 'i si on K,: 

the unremitting action of these movements. These local sii^ 
are permanent attributes, which, when once their relation to 
sensations of movement has been discovered, suffice to brim: 
sensations into the extensive form. 

And the idea of depth, which in binocular vision corms in to 
complete our simple apprehension of visual space, may also arise 
when the eye is wholly unmoved. It often seems to appear 
at the very moment when the eye is affected by an impression 
of light, so that there would be far too little time for its forma 
tion from a number of successive perceptions separated by 
movements. This can be shown very beautifully and con- 
clusively if an observer is allowed to look into a stereoscope 
in the dark, and the stereoscopic pictures are then suddenly 
illuminated by an electric spark. The duration of the electric 
spark is so small that there is no possibility of eye-movement^ 
during the time of its occurrence. Nevertheless, if the picture-* 
are sufficiently simple, there arises immediately after their illu- 
mination by the spark a clear idea of extension in the thin' 

The idea of depth, then, can arise in an extremely short time, 
and certainly requires no series of movements for its production. 
In its case also, that is to say, there must be something among 
the peculiarities of vision with the resting eye by means of which 
the sense of sight can be freed from the conditions orginaily 
imposed upon it. This something cannot be anything else than 
that whereby our vision of plane surfaces has also been to a 
certain extent freed from the co-operation of movements which 
orginally were so necessary. Here again the local character 
of sensations, which serves to determine their arrangement upon 
the retina to which they belong, gives the signal, so to spe; 
at which the mind constructs the spatial extension of each 
particular retinal image. From the differences which it finds 
in these images, it measures the extension of objects in the third 
dimension of space. Just as in the areal field of vision the falling 
of coincident images upon portions of the retina endowed with a 
practically common sensational character gives us an indication 
of the extension of the object in a single plane, the excitation 
of portions of the retina whose sensations are not of similiar 
character serves as a sign for extension in the third dimension. 


194 Lectures on Human and Animal Psychology 

Our measurement of spatial distance corresponding to a certain 
difference between sensations was, as you know, originally made 
in terms of movement ; but after the measure has once been 
obtained, the indissoluble association of the two sensation-series 
(sensations of movement and local sensation-qualities) renders 
it possible for the first series in certain cases to disappear while 
the measure of spatial distance still remains unaffected. At the 
same time, observation shows that the connection of the two 
series cannot permanently be disturbed without causing a dis- 
turbance in spatial vision, and this disturbance can then only be 
gradually eliminated by a new serial association term after 
term. If, therefore, the eye is freed from the determining 
influence of movements in particular perceptions of space, its 
freedom is still nothing absolute ; but now and then the renewed 
control by movement will be found necessary. Only in this 
way is the firm association of the two sensational series, which 
are brought into connection with each other by the persistent 
operation of conditions seated in the sense-organ, preserved in 
undisturbed integrity. 



THE fusion of the two retinal images to a single idea is 
only a particular instance of a general law of the forma- 

tion of ideas. In the visual idea which comes from the two 
eyes we do not discover any trace of the perception of one eye 
as distinct from that of the other ; but we blend them at once 
into a single and indissoluble perception. In this sense it is 
true that the two eyes constitute only a single organ of vision. 
That they are really like two independent observers, regarding 
things from two different points of view, and that we become 
acquainted with the properties of objects only by combining the 
result of these observations, are facts which we do not remark ; 
there is given in consciousness simply the result of this combi- 
nation. That is to say, it is not until the two perceptions ot 
binocular vision have fused that we have an iJta ; or, in other 
words, the fact that the two eyes unite to perform a common 
function is a necessary result of the nature of the mental pro- 
cesses of association. 

This connection of the separate elements of an association to 
a single idea, which is inevitably conditioned by the nature of 
the mind, is also furthered by the laws of external perception. 
Our external perceptions are of such a character, that they can 
only be referred to an object corresponding to the idea which 
has been formed of it. So that we can raise the further ques- 
tion : how does the ideational activity behave in the presence 
of impressions which cannot be referred to one and the 

196 Lectures on Human and Animal Psychology 

spatially extended object ? These conditions, of course, are 
never realised in nature; but we can by means of the stereoscope 
present to the two eyes impressions of the kind supposed. It 
is just as easy to put into the stereoscope different images, chosen 
arbitrarily and at random, as it is to bring before the two eyes 
in this way the planary projections of one and the same object. 
What does the mind do with these perceptions, which it is unable 
to combine into a single idea ? 

There are scarcely any observations which serve to throw 
more light upon the nature of our ideational activity than these 
very experiments, in which something is presented to the organs 
of sense which is irreconcilable with the laws of their normal 
functioning, and which bewilders the mind, as it were, with the 
problem of removing the contradiction involved between con- 
flicting perceptions. One general law may be formulated : 
however great the difference between the separate perceptions 
offered to the two eyes, a separate simultaneous apprehension 
of them is never possible. What happens is either a combina- 
tion of the separate perceptions, on the analogy of stereoscopic 
vision proper, or an alternate apprehension of this or that retinal 

Wherever the two pictures have a certain resemblance to one 
another, whenever, i.e., their difference does not too glaringly 
exceed the differences which occur in nature, they are fused in 
a single idea. And even if their differences only correspond 
remotely to those shown in the pictures of a tridimensional 
object, the idea of depth will still arise at once. The minor 
differences are neglected, and the images interpreted in terms 
of that class of actual objects which they most closely resemble. 

But more than this, figures which cannot be combined at all 
to give an idea of depth fuse to a single idea, if they possess a 
certain similarity by means of which they can be readily appre- 
hended as pictures of the same object. If, e.g., there are placed 
in the stereoscope two circles of nearly the same size, there 
results the idea of a single circle of mean diameter. In the same 
way, if two horizontal lines whose vertical distance from one 
/- another is a little different are presented to each eye, we have as 
a result the idea of two lines at the mean distance. Now neither 
horizontal lines nor circles of different sizes can give us the idea 

Combination f ])ijjer<nt ln;a^* 1,7 

of depth. Ho\v docs it happen that their combination is none 
the less possible ? \Ve mu>t not forget here that then- may In- 
differences between the retinal images of the two eyes even in 
the absence of the conditions necessary for the perception of 
depth. If, e.g., we hold the figure of a circle very near our eyes. 
but a good deal to one side, so that the figure is nearer one eye 
than the other, the retinal image of the nearer eye is greater 
than that of the more remote, since the magnitude of the retinal 
image is directly dependent upon the distance of the object per- 
ceived. In this case there are in the two eyes retinal images of 
different magnitude ; and yet, when we fixate the circle, we sec 
it singly. It is just the same with the two horizontal lines, or 
with any other figures. So that the condition of vision obtained 
by placing in the stereoscope two figures of somewhat different 
size is not essentially different from that which sometimes occurs 
in normal vision. It is, of course, true that in reality we never 
get images of different magnitudes when we fixate an object 
lying directly before us, as is the case in the stereoscopic experi- 
ment. But that is a secondary circumstance which we may 
neglect, because when in normal vision we are estimating the 
magnitude of objects lying very far to one side of us we still do 
not pay any regard to their different distances from the two eyes. 

Quite different phenomena make their appearance if entirely 
different objects are presented to the two eyes. If we place in 
the stereoscope two pictures representing objects taken quite at 
random, we observe a curious alternation of ideas. \Yc neither 
perceive two pictures simultaneously and separately. nor<io the} 
fuse together ; but first one and then the other m.iki 
appearance. It frequently happens that one picture appears by 
itself for a time ; then various portions of th-.* other force 
selves to the front, and then suddenly the second picture- 
can be observed. We notice as a universal rule that there is 
never any simultaneous superposition of the parts which belong 
to one picture upon corresponding parts of the other ; and 
no complex perception, consisting of parts both of the first and 
of the second image, can hold its place as a permanent idea. 

Lectures on Human and Animal Psychology 

Such a composite picture is always a transitional stage from one 
image to the other. And this transition, or alternation, between 
two perceptions forcing themselves upon our consciousness at 
the same time, is very readily occasioned by external influences. 
In this connection the movement of the eyes is of special 
importance. As we move these organs one of them may fixate 
some sharply drawn boundary line within the first picture, while 
the second is directed upon some less prominent portion of the 
second picture. In this way there arises a tendency for the 
former to predominate in the resultant idea. But if we again 
move the eyes, and this point of fixation changes, the second 
image may come to prevail in exactly the same way. First of 
all there enters into the idea that portion of the picture which 
forces itself upon our notice with special intensity, and this 
portion then brings all the rest of the image with it. 

FIG. 32. 

For the observation of these phenomena we may employ quite 
complicated drawings ; but they can be just as well illustrated 
by means of simple figures, e.g., by letters of different form. If 
we present a U to one eye and a W to the other, or &J to one 
and an S to the other, there is never any fusion to a single idea. 
Frequently we see only one letter ; then it breaks up into parts, 
and parts of the other letter are added to it ; and, finally, this 
latter alone is present in our idea. So that there is no perma- 
nence of any one image, but a perpetual alternation, a breaking 
down and building up of images ; and the eye is greatly fatigued 
by this alternation of images which it cannot control. 

If, on the other hand, the two letters have no conflicting charac- 
teristics, they may be united in a fairly permanent idea. Thus we 
can unite an E and an F, or an L and an F, and in both cases 

Reflection and Lustre; Theory oj Lustre 

obtain the idea of an E. But still the perception is not quite ^o 
steady as is the image of a real /: perceived in vision. 
When portions of the two images perceived are superinip.M;,i 
we observe a curious fluctuation in idea. For sonic little- 
distance near the boundary the contour is always altogether 
interrupted ; and this distance varies from greater to smaller. 
A similar interruption of boundary lines may be observed in 
the combination of pictures the lines of which cross one another. 
If we present to one eye two horizontal lines, A, and to the other 
two vertical lines, B, each pair being separated by a moderate 
distance (Fig. 32), we obtain a total image in which the lines in 
one direction are interrupted by those in the other. It may be 
either the horizontal lines, as in C, or the vertical lines, which are 
interrupted. This again depends in most cases on the movement 
of the eyes. If the point of fixation travels in a vertical direction, 
the vertical lines are seen as continuous, and similarly with 
the horizontal lines. It seems as if there is a tendency to see 
objects in binocular vision as extending in the third dimension, 
which has arisen from long habit, and which is manifesting itself 
in these experiments. This tendency is realised so far as the 
nature of the retinal images permits ; and as a result we simply 
see one image behind the other. But the resultant idea is not 
completely explained by this supposition. How is it possible for 
us wholly to ignore this one set of boundary lines, which is so 
definitely presented to us ? How does it happen that portions 
of one retinal image completely disappear? 


To understand these phenomena, we must familiarise ourselvc^ 
with a series of facts, which may be observed both in monocular 
and in binocular vision, and which are of not less importance lor 
the right understanding of the formation of ideas than th< 
which we have already discussed. 

It is a well-known fact that we may see reflected in the 
polished surface of a table the ceiling, furniture, and \vindo\ 
the room in which it stands. And you all know that not only 
are the outlines of the reflected objects perfectly distinct, but tin 
colours are given quite truly. Natural as this observation 
appears, it cannot be directly explained in terms of scn>arion 

2OO Lectures on Human and AmmaL Psychology 

For if the colour of the table is dark brown, one would have 
thought that the white window, mixing with this dark brown, 
would produce some shade of light brown. But that is not the 
case. The colour of the objects reflected in the table is wholly 
unaltered, while at the same time the colour of the table itself is 
distinctly seen. \Ye are not able, of course, to apprehend clearly 
the colour of the table and the colour of the reflected images 
with absolute simultaneity. But we can accurately observe 
the two colours in succession without being disturbed by the 
mixture of light-impressions on the retina. 

Suppose we lay a coloured object, a, upon a uniform colourless 
surface, and hold above it a plate of glass, g, placing beside the 

glass plate a second object, b, of a 
different colour from the first upon a 
r similar background (Fig. 33). By 
looking through the glass plate we see 
the object a directly, and in addition 
the mirrored image b' of b. That is, 
our experiment has exactly repro- 
duced by artificial means the con- 
ditions which are present in the 
reflection of objects in a polished 
table ; we see an object, a, of a definite colour, e.g., red, and a 
reflected image, b', of an object, b, which is also of a definite 
colour, e.g., white. The result in the two cases is exactly the 
same. The image b' is not pale red, but quite unmistakably 
pure white ; and if the attention is directed to the object a, it 
does not either appear as pale red, but we distinctly cognise it 
as a pure red colour. We are, therefore, able to separate out 
and to consider in isolation either of the two coloured impres- 
sions, in spite of the fact of their intermixture upon the retina. 

But this simple experiment is more instructive than observa- 
tions of polished tables or other reflecting objects. For in it we 
can vary the conditions at will, and so gain more exact informa- 
tion regarding the causes of the phenomena. If we turn the sup- 
port of b in such a way that it comes to the position c, where it 
forms with g the same angle as that formed by the latter with 
the support of the object a. the mirrored image of b falls exactly 
upon the place where the object a is seen. But now that this 

Reflection ami Lustre ; Theory oj Lustre. 201 

happens, the two images are referred to one and the same distance 
in space, and, consequently, fuse together. The result is a 
colour-mixture ; the combined image of a and b appear* pale 
red, so far as the two are superimposed. 

The separation of the colours may be prevented in still other 
ways. If the coloured objects a and b are not definitely limited, 
but are so large that there is no clear perception of their 
dimensions, we obtain a mixed sensation, just as in the previous 
case where the mirrored image and the image of direct vision 
were localised in the same place. But the separation will occur 
at once if we draw lines to mark out a smaller figure upon each 
of the coloured surfaces. These boundary lines compel us 
mentally to assign to each figure its definite distance. And 
since the distances of the two figures are clearly apprehended as 
different, there arises the idea of a separation of the two 
images, with their entire sensation content. 

o * 

We see here, then, the ideational activity working to effect a 
disjunction of impressions such as can never take place in the 
sphere of visual sensation proper. In sensation the impressions 
are mixed, however different the objects from which they pro- 
ceed. But since in the idea every impression is referred to i 
object, there is ascribed to each its own amount of participation 
in the mixture. Thus the idea corrects, as it were, what is 
reported by sensation. 

Under certain circumstances, once more, an object mas- 
appear to reflect when seen with both eyes, while it does not do 
so for monocular vision. If in Fig. 34 we fixate the ob 
with the left eye / alone, we see it in 
its natural character. If. on the other 
hand, we look at it with the right eye 
r, we see the mirrored image b 
behind it. When this image is very 
bright and covers the whole of a, it 
may happen that the latter is com- 
pletely ignored, so that the right eye 
sees only b ', and the left eye only a. 
As a consequence, there arises 
single idea of a reflecting object, and 


ith that the clear discrimination of 

2O2 Lectures on Human and Animal Psychology 

the object and the image mirrored behind it. Here we have 
obviously a case analogous to those already discussed under the 
head of stereoscopic experiments. Where the position of the 
object corresponds to that of the reflected image, the former is 
neglected, just as those portions of one of the stereoscopic 
images which were covered by lines of the other image were 
ignored. Since our observations of reflecting objects around us 
have accustomed us to neglect more or less extensive portions of 
an image, we carry this habit of neglecting certain elements into 
cases where the objects seen cannot naturally, and without a 
forced interpretation, be brought under the head of reflectors. 
But this is the sole form of combination whereby the two separate 
perceptions can be fused to a single idea. 

The phenomena of reflection, which occur both in free and in 
stereoscopic vision, are closely allied to those of another class 
which are important as throwing considerable light upon the 
nature of the ideational activity, the phenomena of lustre. Lustre 
and reflection pass into each other without any very distinct 
boundary. Since the phenomenon of reflection depends on an 
ideational activity, we may also infer that lustre will be referable 
to some mode of ideation. At the same time the popular view 
is opposed to such a conclusion. According to it, lustre is at 
least something given directly in sensation, if it is not some 
quality attaching to the lustrous body as such. But very 
simple observations will serve to convince us of the falsity 
of this opinion. 

We found that when the furniture of a room is mirrored in the 
surface of a polished table we are able, despite the mixture of 
colours which ensues, to analyse our sensation into its con- 
stituents, and that in this way we always cognise the reflected 
objects and the reflecting table in their proper colours. But 
our cognition of the mirrored objects is only quite clear when 
the mirroring surface is very uniformly coloured, so that we 
can abstract from this uniform colour of the table surface at the 
points where the mirrored images are visible. A good mirror, 
however coloured, always shows us the reflected objects just as 
they would be if directly fixated. This is not the case if the 
mirror is differently coloured in different parts, or if dark and 
light places alternate on the polished table. Even though each 

Reflection and Lustre ; Theory of Lustn 203 

portion of the mirroring surface reflects with equal cleanup, the 
reflected object is not clearly seen. Why? Plainly because it 
is difficult in such a case to restrict our attention to the a: pre- 
hension of a single object. On the one hand, the attention is 
attracted by the boundary lines of the differently coloured 
portions of the mirroring surface, and on the other by those of 
the reflected object. And this equally strong attraction by 
different impressions brings about a conflict of ideas which 
prevents any permanent or clear apprehension. We cannot see 
the mirrored images clearly for the mirroring object, nor the 
mirroring object for the mirrored images. In other cases where 
a plurality of ideas is simultaneously presented it is still p<>s.-,iblc 
to apprehend each particular one distinctly by bringing it in 
its turn singly before consciousness. Here that is impossible. 
For the same sense-organ gives us simultaneously impres- 
sions which belong to two different ideas. And, moreover, 
the two ideas are of approximately equal intensity, so that 
neither the suppression of one by the other nor their alternation 
is possible. 

The correctness of this account of the origin of lustre may be 
confirmed in various ways by experiment. The phenomenon of 
mirroring obtained when we produce, by means of a plate of 
glass, a reflected image behind the place where the object of 
direct vision lies, can be readily and immediately transformed 
into a phenomenon of lustre by taking the two objects in the 
experiments described above (the mirrored object and that 
directly fixated) in such a way that the ideas which they arouse 
are of equal intensities. Pure mirroring occurs most readily 
when the object seen directly is dark, and the mirrored image 
bright, and when the former is uniform over its whole smt.Ke. 
the latter sharply delineated and obviously situated at a definite 
distance behind the real or apparent mirroring surface. \\ hcr- 
ever the boundary lines of the reflected image are vague, and 
our judgment of its distance accordingly uncertain, or where the 
boundary lines of the directly fixated object are prominent and 
interfere with those of the reflected image, mirroring passes over 
into lustre. 

From this it is clear that the plvnomenon of lustre will 
very readily in binocular \ision, when one eye sees only 

204 Lectures on Human and Animal Psychology 

object, the other only the reflected image. In this case we know 
quite well that we have before us two different things, an object 
and an image which is mirrored in the object. In no other way 
can it come about that the two eyes perceive different colours. 
But we have no idea of the distance of the image behind the 
object. We do not even know which of the two perceptions is 
referable to object, and which to image. And so a very strong 
lustre may be produced by placing in the stereoscope a strip of 
coloured paper for one eye and a strip of the same size and 
form, but of a different colour, for the other. Green and yellow, 
blue and red, or any colours that are sufficiently different, give 
an extremely vivid lustre. And in like manner we get lustre 
by employing very different degrees of brightness of the same 
colour. Strongest of all is the lustre obtained by the com- 
bination of black and white. In this case we do not see a 
black and a white surface, or a white surface through a black 
one, but we obtain the same single impression as when looking 
at lustrous graphite or a lustrous metal, except that the lustre 
is usually stronger than that which we ordinarily find in natural 

Our every-day experience teaches us that wherever we see 
lustre there is no possibility of a clear apprehension of the 
objects seen. Too much or too diffuse lustre is therefore un- 
pleasant to the e}/e, though the luminosity of the object be not 
nearly strong enough to affect us unpleasantly. A lustrous 
stimulus is only pleasant for vision when it occurs at rare 
intervals, and allows the sense-organ to recover itself in the 
meantime by turning to impressions of the ordinary visual 
character. Otherwise lustre dazzles us. And this disturbance 
of vision (which may even affect sensation) is again of a mental 
or psychophysical character. It makes its appearance wher- 
ever there is a conflict of ideas which press upon consciousness 
with equal intensity. We have observed its analogue in the 
stereoscopic experiments, where pictures differing so greatly 
that they could not be combined in a single idea were presented 
to tiie two eyes. In both cases we are only dealing with a par- 
ticular consequence of the principle of ideational unity, which 
we shall have to refer to again in our discussion of consciousness 
and of the connections of ideas in consciousness. While this 

Phenomena of Suppression in Muscular Vision 2u 5 

principle, in the normal course of our mental lite, merely condi- 
tions a steady alternation of particular ideas in .succes- 
sion, it leads to such peculiar phenomena as lustre and ideatiuiiai 
rivalry, when this normal alternation is prevented, whether by 
the striving of two ideas for apprehension at the same time, or 
by the refusal of a simultaneous plurality of ideas to be resolved 
into its elements. 


In addition to lustre and ideational rivalry, there exists yet 
another form of the apprehension of binocular perceptions. 1 1 
the perceptions of the two eyes do not press upon consciousness 
with equal intensity, but one of them has a considerable pre- 
ponderance, for some reason lying in the nature of the external 
impressions, this predominant impression alone becomes an 
idea, while the other is completely ignored. Here again stereo- 
scopic experiments enable us artificially to reproduce the condi- 
tions of the phenomenon. We can do so most simply by 
employing coloured objects of definite outline. If we place a 
black background in the stereoscope and lay on this a white 
square as an object to be perceived by one eye, we obtain, not 
an idea of white and black mixed, despite the tact that the other 
eye sees nothing but black, but we imagine that both eyes see ,t 
white square on a black ground ; and the white is as intensive 
as that of the object seen with the first eye. That is. the per- 
ception of one eye completely suppresses that of the other. 
The explanation of this is obviously to be found in the tact that 
the definitely outlined white object, contrasting sharply with its 
background, possesses a far greater intensity for ideation than 
does the uniformly black surface. The same phenomenon may 
therefore be observed when we lay a black square upon a white 
background, or in general if we present to one eye only a >quare 
of any colour we choose, placed upon a differently coloured 

In like manner one perception may be completely changed 
by the presence of another if a coloured object of like form and 
dimensions is presented to either eye, each, however, contrasting 
at a different degree of intensity with the differently 

206 Lectures on Human and Animal Psychology 

background upon which it lies. Suppose, e.g., that we lay upon 
a white background a dark red object for the right eye, and a 
dark green object for the left. The perception of the right eye 
entirely suppresses that of the left : we see only the red object, 
and nothing at all of the green one. If, on the other hand, 
we had taken a black background instead of a white one, we 
should see only the green object, and nothing of the red. The 
reason plainly is, that dark red contrasts with white more 
strongly than does bright green, while this latter in its turn con- 
trasts more strongly with black. That colour which stands out 
more clearly from the background is more intense for ideation, 
and so we perceive it alone, and entirely ignore the other. On 
a grey background we get the idea of a very lustrous object 
seen in greenish light. In this case both perceptions come to 
consciousness, because they are of approximately equal inten- 
sity, i.e., stand out with equal distinctness from the background. 
But, as we have seen, this simultaneous presentation of two 
different ideas always gives rise to lustre. 

It may sometimes be observed that these phenomena of 
suppression do not extend to the entire image, but are restricted 
to one part of it. This is especially likely to occur when one 
retinal image possesses a much greater extension than the other. 
i / If, e.g., we present 

to one eye a white 
circular surface, /, 
and to the other a 
black circle with a 
small white spot at 
the centre, we shall 
FIG. 35. see this last in the 

common image as a bright spot surrounded by a very dark 
border, which becomes brighter and brighter as we approach 
the periphery, and at last almost entirely white. In this case it 
is clear that the image r entirely suppresses the image / at the 
centre, but conversely is itself suppressed by the latter towards 
the periphery, while between these two portions of the common 
image there are continuous transition-stages. The following 
experiment is of a similar character. To the eye /a uniform 
surface, e.g., blue, is presented, and to the eye r two coloured 

Phenomena of Suppression in liinocutar I'i.ion 207 

surfaces, which are joined in the median line, e.g., green and red. 
In the common image we see at the centre, where green and 
red meet, simply these two colours, while towards the outside 
they are intermixed with a bluish colour-tone. 

But these two last experiments can only partially be sub- 
sumed under the phenomena of suppression. This, it is true, 
plays a part in them, inasmuch as a portion of one of the perceived 
images preponderates over the other, and thereby causes the 
latter to disappear entirely. But the preponderance is in their 
case limited to one part of the image, while in uth ,r parts of it 
it may not infrequently happen, on the contrary, that the image 
of the other eye is the predominant one. This fact seems to be 
almost directly opposed to the laws of ideational activity. The 
fact of the unity of the idea we have seen to be firmly estab- 
lished. It is no contradiction of this that one perception per- 
manently suppresses the other, or that two perceptions alternate 
in succession ; but that each of the two perceptions should be 
partially apprehended, and so appear in ideation in the form of 
a mixed image, this seems to be scarcely in harmony with our 
law. We have, however, already become acquainted with a 
whole class of phenomena in which also two perceptions may 
combine to form a single idea : these, as you know, are the 
phenomena of lustre and reflection. In the case of lustre two 
ideas are presented to us, which we do not succeed in keeping 
separate. In mirroring we effect this separation, and we can 
therefore either alternate between the idea of the rellected and 
that of the reflecting object, or we can unite both of them in a 
total idea. When we look at the image in a mirror, we usually 
embrace both image and mirror in a single idea ; the mirror is 
the frame which surrounds the picture. Now in the experiment? 
which we have just been describing we have obviously the same 
conditions. Besides the considerable intensification of one por- 
tion of a visual perception, the idea of reflection also exerts 
some influence. And therefore that especially prominent |>or- 
tion of the perception of one eye is alone given in- ideation at 
the place which it occupies in the common image, at the other 
parts of which ideation has free play, and so tends to apprehend 
the image of the other eye as a mirror in which the first is seen 

Still the conditions in this artificial experiment do not entirely 

208 Lectures on Hitman and Animal Psychology 


correspond to what takes place in nature. In nature, too, it 
may happen that we see only the mirror with one eye, and only 
the mirrored object with the other. We need simply hold the 
mirror close to our eyes and cause the mirrored image to fall 
very much to one side. But there are many conditions which 
we can introduce into the experiment which are never realised 
in nature. Suppose, e.g., that we place a large blue and a small 
yellow object in the stereoscope for combination in binocular 

vision, both lying on a red 
background (Fig. 36) ; we have 
a common image, in which the 
yellow is seen surrounded by 
blue. So far there is nothing 
out of the common, for it may 
also happen in nature that we 
see a yellow object reflected in 
a blue mirror. But where this 

happens we must necessarily see the mirror with the same eye 
which sees also the mirrored object. For if we are viewing a 
small object in a very large mirror, it may certainly happen that 
this image is only visible to one eye, but the mirror itself is 
never visible to one eye only, least of all to the one which is 
not looking at the mirrored image. The conditions of the 
experiment here, then, are not in accordance with nature. How 
does the eye cope with its perplexity ? Since the right eye sees 
yellow upon a red background, and the left blue on the same, 
there is occasioned simply the idea that a yellow object on a red 
background is mirrored in a blue object ; that is, not only the 
yellow object, but also the red background which immediately 
surrounds it, are comprehended in the resultant image. Farther 
towards the lateral parts of the image, however, there arises in 
ideation the perception of a blue surface distinctly outlined 
against the red ; and so the blue sensation gradually comes to 
predominate. Thus we obtain as a final image that of a large 
blue square upon a red background, in the middle of which is a 
small yellow square surrounded by a fringe that shows a deep 
red on the inside, but becomes more and more tinged with blue 
towards the outside. 

All these phenomena, which can be varied in many other 

Phenomena oj Suppression in Jiuiocnlar I 'i.^n>i: 2^(> 

ways, show that a single idea is alw.iys formed from the 
tions of binocular vision, and that this always takes pLce 
analogy of the conditions of vision found in nature. The pro- 
cess by which the two visual perceptions unite to form one id--a 
depends, therefore, upon the formation of numerous association*, 
some acting in harmony with, and some in opposition to, one 
another. In this latter case, phenomena of suppression, or idca- 
tional rivalry, make their appearance. The separate perceptions 
of the two eyes are themselves composed of sensations which 
combine with each other in a quite different way from the li_;ht- 
impressions which fall upon one retinal point in the same eye. 
We have rather to regard the binocular visual ide i .is a /, 
resultant of the originally separate perceptions of the /.v 




THE mental phenomena with which we have been con- 
cerned hitherto have represented stages in one and the 
same great process. We have seen that ideas are derived from 
sensations in the regular course of development, and that both 
alike have a single end, knowledge of the external world. But 
we have purposely neglected one very important side of our 
mental life. We never actually find a mind which apprehends 
things without joy or sorrow, and contemplates them with ab- 
solute indifference. In cognising objects we feel ourselves 
attracted to or repelled from them, or incited to the performance 
of some kind of action, according to their nature. We can, 
therefore, comprehend all those phenomena which are not in- 
cluded in the ideational process under the two words 'feeling' and 
'will.' Feeling and conation always accompany our sensations 
and ideas ; they determine our actions, and it is mainly from 
them that our whole mental life receives its bias and stamp of 

Feeling and will are closely interconnected. And both are 
again connected with ideas. The separation between these 
processes is one that exists only in psychological abstraction, 
and has no basis in reality. Feeling passes over into impulse, 
impulse into voluntary action, and voluntary action has reference 
to objects which are given to us as ideas. 

In ordinary language we employ the word 'feeling' in 
various senses. We call hunger and thirst 'feelings' ; we speak 
of the ' feeling ' of pain, and of ' feeling ' external objects 

The Feelings 

21 I 

with our hands; we call love and hate, joy and sorrow, h->pe 
and anxiety, 'feelings'; we talk of our 'feeling' for the 
beautiful and the ugly, and even of ' feeling' that something is 
true, or honourable, or virtuous. What is our justification for 
bringing under one and the same concept mental processes 
which are so diverse in their nature, and belong to such different 
stages of development ? It may be mere chance ; language 
may somehow have come to apply the same name to a number 
of totally different phenomena. Or it may be purpose ; perhaps 
these processes possess something in common, notwithstanding 
all their divergences. 

As a matter of fact, there is one point in which all ' feelings ' 
agree, however different they may be in other respects : they 
all imply a condition of the feeling subject, an affection or 
activity of the self. Feelings are always subjective, while the 
idea always has an objective reference. Even when the matter 
of ideation is some one of our own conscious processes, thi> 
is regarded objectively. So that the term 'sense of feeling 
has been reserved for the sense whose impressions are most 
obviously connected with subjective states of pleasure and un- 
pleasantness. What language implies in calling joy and care, 
love and hatred, 'feelings,' is simply that they are apprehended 
as exclusively subjective states, and not as properties of objects 
outside us. 

The attempt has sometimes been made to limit the meanm., 
of the term 'feeling,' on the ground that the processe 
hended under it are entirely too diverse in character, 
particular it has been thought necessary, from the psych- >1 
standpoint, to strike out from the category of feeling a'.i those 
subjective excitations which are directly connected witii >en- 
sations. Hunger, thirst, bodily pain, impressions of touch in 
general, are, it is said, sensations ; they are accompame- 
physical processes in the nervous system : but feeling is a pu 
mental condition. The term should, therefore, be 
mental states which are independent of bodily affections and 
arise solely from some kind of reciprocal action among ideas. 
But as soon as we give up the reference of feeling to a >ubjc-c 
tive condition of pleasure-pain, or of some similar pair of sub- 
jective opposites, we have no reason for uniting affective states 

212 Lectures on Human and Animal Psychology 

in general in a common class. If, on the other hand, we retain 
this principle of classification, we cannot exclude from 'feel- 
ing' the sense-feelings attending the simple sensations. It 
seems to have been regarded as a difficulty that one and the 
same simple process should be called both sensation and feel- 
ing. But it has been forgotten that joy and sorrow, hope and 
anxiety, and all the other ' feelings ' are really states of mind 
which are affective only so far as they have reference to the 
feeling subject ; while in other respects they depend upon ideas 
which objectively regarded are entirely empty of feeling. The 
only difference is, that in the case of these more complex feel- 
ings we attach a greater value to the feeling as such, and 
therefore give a separate name to each of its particular forms. 
A very complex ideational association may give rise to a feel- 
ing which is simple and uniform throughout. We are, therefore, 
inclined to substitute the subjective result, the feeling, for those 
various and complex processes which serve as its objective 
background. There is no need for this in the case of the simple 
sense-feeling. Its objective substrate is a sensation which is 
equally simple with it, and more readily discriminated by virtue 
of its relation to an external object. 

If the feeling is characterised merely by its relation to the 
feeling subject, it is clear that the distinction between it and 
the sensation or idea cannot be in any sense original. The 
simple sense-feeling in particular is contained in the sensation ; 
and it is just as incorrect to say, ' Feelings alone are primitive,' 
as to say, 'We have at first simply sensations.' The ultimate 
fact is, that we sense and feel. The logical separation of 
feeling from sensation can only come about after we have dis- 
tinguished subject from object. Then, and then only, is the 
elementary process of sensation analysed into a subjective factor, 
the feeling, and into an objective factor, the sensation. The 
sense-feeling may in this way be considered as an integral 
element of the sensation itself; and for that reason it is also 
termed the affective tone of sensation. 

Since the aim of these lectures is to give a general description 
of the elementary constituents of our mental life before dealing 
with their connection in consciousness, and the complex pro- 
cesses to which this connection gives rise, it will be best for us 

Sense-feelings 2 1 * 

to confine our attention here to tiie ^nsc-j^iin^. The Minpii- 
city of the conditions upon which they depend make-, it e.t x v 
to examine them without any detailed reference to their con- 
nection in consciousness. For the same reason they wi.i give 
us the most efficient aid in our endeavour to understand the 
general nature of feeling and its relation to will. But we shall 
be obliged occasionally to glance at the higher feelings, 
especially the intellectual and aesthetic. There are certain 
problems to which these furnish more definite answers, just 
because of the more complex associations which they involve. 


There are some sense-organs which require quite intensive 
stimulation for the production of a sense-feeling possessing any 
considerable degree of strength. This is especially true of the 
eye and the ear. A moderate visual or auditory stimulus seems 
to give rise to scarcely anything except the objective sensation ; 
a moderate light-impression, e.g., is referred simply and solely 
to the external illuminating object. But it must be admitted 
that careful introspection enables us to recognise a certain 
affective tone even in the sensations produced by weak visual 
and auditory stimuli. We notice this especially when the im- 
pressions are given without direct reference to external object." 
with clearly defined boundaries. Thus the ditierent spectral 
colours, red, green, blue, etc., as well as white ami black, have 
characteristic, if weak, feelings attaching to them ; while every 
single musical clang is also attended by an affective tone, deter- 
mined in each case by tonal pitch and clang-character. But 
the intensity of these feelings is very slight. Their importance 
is, however, increased by the fact that they enter as in; 
elements into the (esthetic feelings, which are more ei.i 
affective states, connected with entire ideationai complexes. 
On the other hand, a dazzling light or a deafen 
directly occasion a feeling of pain, behind which the 
significance of the sensation may m u.s turn entirely disapp 
These intensive stimuli disturb the normal fund 
or"-an is, theretore, the subjective factor which comes 

& ' 

most predominantly into consciousness 

214 -Lectures on HIUULDI and Animal Psychology 

Cutaneous sensations are also as a rule referred wholly to 
external impressions, so long as they are not painful. But there 
are certain stimuli which, although truly tactual, produce sen- 
sations with very intensive affective tones. Stimuli of quite 
weak intensity, which only lightly touch the surface of the 
skin, excite tickling or itching. Both of these may also arise 
independently of external impressions. They are always charac- 
terised by a tendency to diffusion of effect. The mechanism 
of the matter probably is that weak tactual stimuli call into 
reflex activity the unstriated muscles which lie directly beneath 
the skin, and to which cutaneous movement is due. When 
these muscles are contracted, we have also the peculiar feeling 
of shivering accompanying the muscle-sensation proper ; and 
this readily combines with the feeling of tickling. The reflex 
excitation of the cutaneous muscles frequently extends to 
other muscle groups ; and when the excitability is great, may 
lead to general reflex convulsions, which are greatly exhausting 
to the organism. 

A fairly low degree of temperature has an analogous result. 
If we let a cold stimulus of slight intensity act upon the skin, 
we have first a sensation of cold, i.e., the cold is perceived as 
an alteration in the condition of excitation of the cutaneous 
organs, then the smaller cutaneous muscles are reflexly excited, 
and so the feeling of shivering is occasioned. The same effect 
may be produced by the action of internal causes in producing 
a sudden loss of heat, and consequently a cutaneous sensation 
of cold. This happens in fever chills, where the effect is much 
intensified by the abnormal reflex excitability of the cutaneous 
muscles. Very high or very low temperatures, finally, have the 
same result as very strong pressure-stimuli : they do not pro- 
duce sensations of heat or cold, but only severe pain. And 
it is of the very nature of pain that its character is always 
constant : a prick, a grinding pressure, intense heat, and freezing 
cold all excite pain of the same strongly affective quality. 

It is different with impressions of smell and taste. Even at 
weak intensities, these are accompanied by distinct feelings of 
pleasure and unpleasantness. And the feeling is so closely 
fused with the sensation that it seems impossible to think of 
the two as even temporarily dissociated. It is only the fact 

Sense-feelings 2 i r 

that there are impressions which arc relatively free from fcelin ; 
which leads us to believe that the quality and intensity of the 
feeling depend upon other conditions than those of the sensa- 
tion. It is, perhaps, scarcely necessary to point out how 
important the strong .effective tone of the sensations of .smeii 
and taste is for our physical life. Here, more than in any of 
the other senses, the feelings of pleasure and pain serve a^ sub- 
jective indications of the impressions which \ve should seek 
and of those which we should avoid. Thes indication, may, of 
course, occasionally lead us astray; but the adaptation of 
natural unperverted feeling to the beneficial or injurious charac- 
ter of stimuli is on the whole marvellously complete. 

But peripherally excited sensations do not constitute: the 
only material for the formation of sense-feelings. There are a 
large number of sensations which are not caused by external 
impressions, and which do not imply the apprehension of ex- 
ternal objects, but which may in all other respects be co-ordi- 
nated with the sensations from the sense-organs proper. To th:- 
group belong, in the first place, the muscle-sensations, which we 
have already described in our inquiry into the processes con- 
cerned in perception. Moderate muscular exercise is connected 
with a more or less distinct feeling of pleasure; while ex- 
haustion, excessive effort, or a pathological condition of the 
muscles not only alters the quality of muscular sensation, but 
also brings with it a very intensive feeling of unpleasantnes 
Secondly, there belong to this class sensations from the various 
tissues and organs of our body. These are generally of > 
intensity, and therefore easily overlooked. Hut under special 
conditions, especially where the state of the particular or,;an 
or tissue is pathological, they may become so intensive 
the feelings connected with them dominate consciousne 
almost exclusively, and produce general discomfort throughout 
the organism. These sensations, therefore, are only to 
us at their highest degree of intensity, i.e., when they have be- 
come painful. But the affective character of pain is. as we have 
already remarked, essentially the same in every case. For that 
reason, the specific differences of the org.mic feelings are not 
generally apprehended by us. Nevertheless, observation sho\ 
that such specific differences really exist. Language use 

2i 6 Lectures on Human and Animal Psychology 

different terms to denote the pains coming from different 
organs. We speak of ' stabbing ' and ' gnawing ' pains in our 
bones, of ' pricking ' pains in the porous serous membranes, 
and of 'burning' pains in the mucous membrane. Here, 
as in the external sense-organs, pain is simply the sen- 
sation raised to its highest degree of intensity ; and that 
peculiar property of the pain which is dependent upon the 
structure of the organ is prefigured in the pure sensation. This 
is especially noticeable in intermittent pains. There are times 
when sensation is present which cannot be called painful ; and 
during these intermissions the peculiar sensation-colouring 
which gives its special character to the ensuing pain is usually 
not at all affected. 

We may, therefore, regard these sensations from the bodily 
tissues and organs as originally equal in value to those from 
the organs of special sense. Gradually, however, these latter 
acquire a paramount position, through their importance for the 
development of ideas ; while the great majority of the organic 
sensations pass unnoticed until their unusual intensity in a 
particular case announces some important change in the con- 
dition of the organism, of which consciousness is obliged to take 
account. That is to say, in proportion as sensations, originally 
O r iven as undifferentiated, are divided up into sensations of 
special sense (relatively free from affective tone) and organic 
sensations (strong in affective tone), there is a tendency for this 
latter group gradually to disappear from consciousness. Hence 
the entire life and action of the child is determined by the 
sense-feelings ; while the more developed and better furnished 
the mind becomes, the more independent is it of their domina- 
tion, and the greater is its measure of success in permanently 
repressing the weaker sense-feelings and at least temporarily 
subduing the stronger ones. The only exception is the hypo- 
chondriac, who delights in observing his own bodily symptoms 
and states. By dwelling anxiously upon all those weak sensa- 
tions which pass unnoticed by the normal consciousness, he 
gains a great deal of training in the apprehension of his sense- 
feelings. The physician will often laugh at his 'pains' and 
'aches' as illusions ; but generally they are real enough. The 
abnormality of the hypochondriac does not consist in his per- 

Common Feeling and tlic Otli, r Total /v t 


ception of feelings which have no existence, but in his uiv.inct 
apprehension of, ami anxious reil .-ction upon, feelmg> which a 
healthy man usually takes no notice of. 

To the specifically organic feelings belong also those of 
hunger, thirst, and shortness of breath. They, too. arc c ,n- 
nected with sensations which normally recur with moderate in- 
tensity at definite intervals, but which will increase more .m,i 
more in affective intensity, if their demands are not satisfied. 
Hunger, thirst, and respiratory excitations are sensations cen- 
trally aroused, but peripherally localised, thirst in the mucous 
membrane of the palate and throat ; hunger in the stomach ; t he- 
respiratory sensations in the respiratory organs, and especially 
in the muscles of the chest, which subserve the process of 


Our general condition of bodily comfort or discomfort is de- 
pendent upon the mass of organic sensations which are always 
present, but in varying degrees of affective intensity. The 
totality of feelings acting upon consciousness at a given moment 
is termed the common feeling. This has been defined in term> 
of its origin as the sum total of simultaneously present organic 
feelings 6f whatever quality. Hut the definition overlook.- the 
fact that our state of feeling is alway> qualitatively single. \Ve 
can never be moved simultaneously by a number of diverse and 
independent feelings ; they combine to form a resultant possess- 
ing the character of a feeling of definite quality and intensity. 
At a given moment we feel 'well,' or 'ill,' or indifferently. If we 
should ever say, in speaking of our general bodily condition, 
that we are feeling at once 'well' and 'ill,' it may ,d\va\s be 
proved by introspection that we have been uniting sufi-t-su: , 
feelings in the single judgment. Hut this is the result ot re- 

O *r> J \r* 

flection upon our leelings ; the fusion is not given n: the ! 

This qualitative unity of feeling seems to cone 
ideational unit)- of our consciousness. The scn-ations 
at a given moment bv external and internal .-timnli are not 

o * 

perceived as a mere medley of impiessions, but are associated to 
form ideas, which are then brought into spatial and temporal 

218 Lectures on Human and Animal Psy etiology 

relations with each other. In the same way all the particular 
feelings are united in one total feeling, into which each enters 
as a constituent factor. But the analogy cannot be carried any 
farther ; there is an important difference between ths two pro- 
cesses. We can prove that an idea is a compound process by 
analysing it into sensations. A clang, a compound clang, and 
a visual object are single, but not simple, mental facts. We 
can analyse each of them into a number of simple sensation- 
elements. But no feeling is capable of analysis in introspection, 
whether it is connected, like the sense-feeling, with a single 
sensation, or, like the elementary aesthetic, intellectual, and 
moral feelings, with a complex ideational group. 

This simplicity of feeling, together with its subjective charac- 
ter, which makes it impossible for us to refer it in every case to 
external objects, as we do with sensations and ideas, is doubtless 
the cause of that ' obscurity ' which has been so often empha- 
sised, and which consists simply in the indefinable nature of 
affective quality. It is this ' obscurity ' which has led to the 
attempt to substitute for the impossible definition an enumera- 
tion of the objective conditions under which feeling arises, and 
a description of the relations obtaining between the ideas pre- 
sent at the moment of its appearance. These supply the only 
means at our disposal for the production in others of feelings 
similar to those which we experience ourselves under particular 
circumstances ; and they are perfectly justifiable so long as we 
do not mistake them for an account of feeling itself. But 
psychology has fallen into this very error again and again. 
That is, it has ' explained ' feeling by reflecting upon the ideas 
among which it arises, and by which it is attended. Sense- 
feeling has been defined as a furtherance or inhibition of our 
bodily well-being, or even as a direct cognition of the useful- 
ness or danger of sense-stimuli. ^Esthetic feeling has been said 
to consist in the idea of definite mathematical proportions, 
moral feeling in reflection upon the useful or hurtful conse- 
quences of our actions, and so on. Leaving out of account the 
objections to these theories on other grounds, we see that they 
are all overthrown by the consideration that feeling is not it- 
self an intellectual process at all, although it is always con- 
nected with intellectual processes 

Common Feeling and the Other Total l-'ccling, 219 

Every feeling is a qualitatively simple* and undeounp,,sable 
mental state. This fact does not, of course, exclude the possi- 
bility of there being in consciousness several simultaneous ic-ei- 
ings. Only, these simultaneous feelings always combine in a 
total feeling which possesses a unitary character, and cannot, 
therefore, be regarded simply as the sum of the original par- 
ticular feelings. The oscillatory and the discordant feelings are, 
perhaps, the most instructive examples of these complex affec- 
tive states. In the first group, opposing feelings alternate with 
each other in rapid succession. Hut there is also a continuous 
modification of one affective phase by the other, so that a new 
feeling with a characteristic quality of its own arises alongside 
of the primary changing feelings. Its quality is, of course, de- 
pendent upon those of the original feelings; but it can;, -t be 
analysed into them. Its intensity is constantly altering, so that 
at one moment the primary feelings, at another this new ]> -r- 
manent feeling which is characteristic of affective o.iciliaii- .n, 
predominates in consciousness. Affective discordancy is directly 
derived from affective oscillation. It occurs when the oscilla- 
tions of feeling follow each other very quickly, and the suca 
feelings themselves are strongly opposed. \Ve have an ex- 
ample of this from the sense-feelings in tickliiv.;, and fr- m 
the intellectual feelings in doubt ; while the dissonance of 
two clangs may be taken to exemplify it in the field of tin.- 
elementary aesthetic feelings. 

The statement that doubt is a compound of the feeling 
acquiescence and repugnance is certainly a true description 
of the alternating affective states which go to constitute tin- 
entire mental process. But there >,cems to be present in ad 
tion a resultant total feeling directly corresponding to the 
sension in the emotional condition. There may be moments ot 
doubt when neither the feeling of acquiescence nor the !e 
of repugnance is in consciousness at all ; and thes<. 
possess a unique affective character which does not a. near t 
analysable into either of the other two feelings whici: 
it from time to time; but it may continue to exist a 
of them. At such moments, therefore, there exis: 
jng S) those of acquiescence and repugnance and the total f 
resulting from the two, but qualitatively different from tin 

22O Lectures on Human and Animal Psychology 

Doubt always involves a strong opposition between the con- 
stituent feelings. In the feelings of tickling and of dissonance, 
which are formally related to it as ' discordant ' feelings, although 
their ideational connections are entirely different, the affective 
state is more homogeneous. In tickling, which is due to a con- 
tinued weak cutaneous stimulation, we can clearly distinguish 
two original feelings, either of which may predominate accord- 
ing to circumstances, a pleasurable feeling, which probably 
accompanies the weak touch-sensations ; and a painful feeling, 
which appears to be connected with reflexly excited muscle- 
sensations, to which, e.g., those of the diaphragm belong. If the 
tickling is slight, the direct effect of the stimulus, and therefore 
the pleasurable feeling, is most prominent ; if it is more inten- 
sive, the reflex effect, and therefore the unpleasant feeling, gains 
the upper hand. The specific total feeling of tickling is the 
resultant of these two It, again, may be perceived with especial 
clearness when the two opposing factors are of approximately 
equal intensity. But in general, and perhaps as a result of the 
strength of the two factors, the total feeling in tickling is of 
relatively slight intensity. The opposite holds of the dissonance 
of two clangs, when we can always distinguish the feelings which 
a> - e connected with the separate clangs from the total feeling of 
dissonance itself. As the dissonance increases, the total feeling 
prevails more and more over those excited by the separate 

We see, then, that the total feelings arise from the union of 
particular feelings, but that they constitute new and simple 
feelings of definite quality, quite distinct from their constituents. 
Plainly the common feeling must be regarded as belonging to 
this group. We do not intend to denote by the phrase the 
medley of miscellaneous separate feelings present in conscious- 
ness at a given moment, but rather a new feeling to which they 
give rise, and whose quality they all help to determine. The 
entire sum of separate feelings combines to form a complex 
unity, the trend of which finds its expression in the resultant 
total feeling. Similar total feelings with accompanying par- 
ticular feelings constitute the 'higher' intellectual, aesthetic, and 
moral feelings. In all these cases, every particular feeling and 
every total feeling have their own characteristic quality, in 

Relation oj Feeling to Idea 

22 I 

virtue of which they stand in relations of agreement an-! ,11* 
agreement to other feelings, though they arc" never ana'.ys.djle 
into them. So that nothing can be more erroneous than tin- 
opinion sometimes held that the entire world of feeling is com- 
posed of a certain sum of elementary feelings, perhap. >cnse- 
feelings, of approximately constant quality. The essential 
characteristic of feeling, especially of the higher feelings is 
rather an inexhaustible wealth of qualities ; new qualities arise 
from the mutual influences of simultaneous feelings, and from 
the induction of present by antecedent feelings. And to tiii- 
we must add that the worth of the feelings constantly increase- 
as the relations in which they stand become m<> re complex, for 
it is these relations which determine the influence of any par- 
ticular feeling upon our entire mental life. 


Finally, the existence of total feelings, and especially of those 
which reflect an oscillatory or discordant affective state, iead< 
us to an important fact, without mention of which everything 
that we have said about feeling, and especially about the origin 
of resultant feelings, would be incomplete. \Ye have considered 
feeling primarily as a process which accompanies ion 
Since we have been occupied hitherto with the anaK -is of iue,. 
they naturally suggest themselves as points of departure for on: 
investigation of feeling. But ideas are not the only mental 
processes, even if we abstract from feelings and the other sub- 
jective processes connected with them. All the changes which 
occur in a given ideational content are as such also inent..'. 
processes characterised by a particular rapidity and manner o! 
occurrence, and, like the ideas themselves, connecte i with leei 
ings. So that even from the standpoint of an objective obser- 
vation, which only takes account of the ideational, and not of 
the affective, side of mind, we are obliged to distinguish tlu-> 
processes of change in ideational content from the ideas 
are altered. Introspection of such changes is ais< 
called 'ideation.' We are told, t \., to ' lorm an idea ' of some 
change ; that a content is appearing or vanishing, or that con- 
tents are passing through consciousnes 

222 Lectures on Human and Animal Psychology 

rapidity. It is scarcely necessary to remark that we are not 
really dealing here with two different things, ideas and changes 
in their condition and arrangement. The analogy with physical 
bodies and their changes of position which is usually thought of 
is entirely misleading. The ideas themselves are, as you know, 
not unchanging objects, but processes, occurrences, whose exist- 
ence is necessarily bound up with that of the changes supposed 
to take place in them. If, e.g., an idea disappears, that means 
that the mental process which we call an idea ceases to exist. 
So that when we speak of 'ideas with which feelings are con- 
nected ' our language is at least subject to misinterpretation. 
We should rather say: all ideational processes, whatever their 
nature, whether they consist of the idea of an external object, 
or of some internal change in this idea, are at the same time 
affective processes. Affective discordancy furnishes an obvious 
proof of this : in doubt and in dissonance the resulting feeling 
is determined to a far greater extent by the characteristic 
alternation of ideas than by the nature of the ideas themselves. 
The total feelings in particular are always essentially dependent 
on some peculiarity of the alternation and succession of ideas. 
We shall return again to this point in our consideration of the 
emotions, of which total feelings are important constituents. 




IN the previous discussion we took as our starting-point the 
fact that the affective side of consciousness at any moment 
seems to exhibit a unitary interconnection similar to that 
presented by its ideational contents. But further investigation 
convinced us that the affective unity of consciousness differs in 
important respects from its ideational unity. The latter appears 
to be external, in the sense that the particular ideas are united 
into a whole, more especially by the spatial relations in which 
they stand to each other, without the constituents of this whole 
being necessarily brought into any internal relation. In feeling 
it is quite different. It is true that several qualitatively different 
feelings may exist side by side, but they always give n.>>e also to 
a total feeling which endows the entire group of separate feelings 
with an internal coherence. 

We shall best understand this internal unity of feeling if we 
look somewhat more closely at the connection ol _/<r/;v, 
will, a subject to which we reterred in a general way at the 
beginning of the last lecture. This connection may be regarded 
from two points of view. In the first place, feeling is only 
thinkable as a mental state of a being endowed 
pleasurable and unpleasurable feelings tend to direct the 
of the will. Whether or not they pass over into actual volition 
is determined by internal and external condition^. Hut without 
the capacity to will the alternatives could not po-sibly exist. 
Secondly, will is an internal process, distinguished from other 

224 Lectures on Human and Animal Psychology 

mental activities through the tact that in it \ve are conscious of 
definite motives. But motives are always accompanied by feel- 
ings, and tiie feelings further appear to us as those elements of the 
motive which contain the real reason for the activity. Without 
the excitation which feeling furnishes we should never will any- 
thing. A mind which contemplated things with entire indiffer- 
ence as ' pure intelligence ' could never possibly be roused by 
them to volition or action. Feeling, therefore, presupposes will.. 
and will feeling. In a concrete voluntary action the two are 
not different processes, but part-phenomena of one and the same 
process, which begins with an affective excitation, and passes 
over into an act of will. It often happens, however, that the 
final term in this series is wanting: the intensity of a feeling 
may become lessened, or it may be displaced by another feeling 
without leading to an act of will. We may therefore divide 
feelings at once into two groups, those which form constituents 
of a voluntary act and those from which no definite volition 
results. The latter class, again, contains different degrees. If 
the subjective condition is, and remains, simply a pleasant or un- 
pleasant mood, we speak of a feeling proper. When there is 
added to this a definite tendency towards a willed result, we 
term the internal process an effort or an impulse. If in this 
effort we are further conscious of some inhibition, which prevents 
it from passing over directly into volition, we call it a desire. 

It is in the doctrine of feeling and will more than anywhere 
else that psychology still wears the fetters of the old faculty- 
theory. And so it has usually taken a radically false view of 
these intimately connected part-processes, regarding each con- 
stituent as an independently existing whole, which might in- 
cidentally, but need not necessarily, exert an influence upon the 
constituents of the other. Thus first of all feeling was con- 
sidered apart from its connection with will, and then desire was 
treated as a separate process, sometimes found in connection 
with feeling. Further, impulse was opposed to desire proper as 
an obscure desire, in which the subject is not conscious of the 
desired object ; or, perhaps, as a lower desire, referring exclusively 

Dei'dopnient of II ill 


to the needs of sense. (That is why many psychologists li "id 
that impulses only exist among animals.) And fln.nly tin -e 
processes are still further supplemented by the postulation ol 
will as an entirely new and independent faculty, whose function 
it is to choose between the various objects of desire, or in cert. tin 
circumstances to act in accordance with purely intellectual 
motives and in opposition to impulses and desires. According 
to this theory, that is, will consists in the capacity for J fit- 
choice. Choice in this sense presupposes the possibility f 
decision between various objects of desire, and even of deci 
against the desired object on the ground of purely rational con- 
siderations. It was therefore supposed that desire ir a condition 
which precedes volition, and that at least in many cases this 
latter is only the realisation of desire in action. 

We must pronounce this theory a purely imaginary construc- 
tion from beginning to end. It has taken its facts from every 
possible source except an unprejudiced introspection. Feeling 
is not independent of volition, as alleged ; impulse is not a 
process which can be distinguished from will, still less opposed 
to it ; and desire is not the uniform antecedent of will, but 
rather a process which only appears in consciousness when 
some inhibition of voluntary activity prevents the realisation of 
volition proper. Finally, to define the will as the capacity of 
choice is to render any explanation of it impossible from the 
outset. Such a capacity presupposes volition as its antecedent 
condition. If we could not will without choice,/.*., as directly 
determined by internal motives, a volition involving ch 
would necessarily remain impossible. 

This confusion of volition and choice brings another error ;n 
its train. Will is supposed to arise from all sorts of involunt. 
activities. Generally this view is applied exclusively to c: 
voluntary acts, which many psychologists regard as the 
ones. Both the human and animal body, it 
originally, before the appearance of will, the 
ments of the most diverse character. These 
part purposive, owing to the ideological 
with motor fibres in the central organs. Thus a -tin 
caused pain would give rise to a lefle.x movement of c 
resulting in the removal of the stimulus. It is further supi 


226 Lectures on Hitman and Animal Psychology 

that the mind perceives the purposiveness of these reflex re- 
actions, and so the thought arises in it that it might possibly 
undertake similar movements itself, and attain the same pur- 
posive result. The next time that the stimulus approaches, 
therefore, the mind will be on the ajert to execute the defensive 
movement, and so remove the stimulus before it has any painful 
consequences. The most remarkable results of this kind of re- 
flection are those obtained as regards locomotion. It may per- 
haps happen that the body gives a sudden spring, in response to 
a strong reflex stimulus. ' Eureka ! ' says the mind to itself. 
' Why should I not cause my body to spring when this un- 
desirable stimulus is not there?' But when the will has once 
discovered that its voluntary muscles enable it to do almost any- 
thing it wishes, it, and not the reflex, is master. The reflex has 
played its part, and is henceforth restricted to the sphere of the 
absolutely necessary. 

You will not find, of course, that this description is in literal 
agreement with that given in any of the works treating of the 
development of voluntary acts from reflex movements, but in 
substance there is no difference. You will even find such ex- 
pressions as ' The mind takes note of this and that,' or ' It now 
executes movements voluntarily which it formerly observed 
taking place in the body involuntarily.' And indeed there is ho 
reason why the mind should not act in this way, if it were ' pure 
intelligence,' as these writers palpably assume, or even if it 
only had at its disposal a small number of feelings to occupy 
its leisure moments. 

But the matter assumes a different aspect when we look at it 
without preconceptions, and refrain from reading into the facts 
of observation notions and reflections which exist only in our 
own minds. In the first place, there is not the slightest con- 
firmation to be found for the assertion that the lower animals, 
and children in the early days of life, are merely reflex machines, 
which make certain movements with mechanical certainty as 
soon as we press the spring. Even such of the protozoa as un- 
doubtedly belong to the animal kingdom give plain evidence of 
voluntary movement. The chick just out of the shell executes 
movements which are in great part at least of the nature of 
voluntary actions. No one will of course deny that reflex 

Development of Will 227 

movements may also be observed from the fir-t especially 
among the more complexly organised animals. We have our- 
selves referred to the reflex movements of the eye and the 
organs of touch, and the part which they probably play in the 
formation of our space-perceptions (pp. 126 ff.). It must not, 
however, be forgotten that these purposive reflexes have become 
possible through an organisation acquired in the course of count- 
less generations. What are the conditions which have been 
operative during this development to increasingly modify the 
organisation of the nervous system, so that the movements 
which constitute its mechanical response to external stimuli may 
be as well adapted as possible to subserve the immediate ends 
of the life of the organism ? There is only one intelligible 
answer to this question. It consists in a reference to those 
processes which even during the individual life mediate the for- 
mation of purposive reflex and automatic movements, to the 
processes underlying practice. Practice always implies that an 
action which at first was performed voluntarily has gradually 
become reflex and automatic. Thus when the child learns to 
walk, the taking of each single step is accompanied by a consider- 
able effort of will ; but after a time and by slow degrees it be- 
comes able to initiate a whole series of movements without 
attending to their execution in detail. In the same way, we 
learn to play the pianoforte or to execute other complicated 
movements of the hands by frequent repetition of particular and 
connected acts, and their consequent transformation into a chain 
of effects which follow each other with mechanical certainty 
when once the appropriate impulse has been -given. Now the 
modifications which the nervous system undergoes during the 
life of the individual in consequence of the mechanising of these 
practised movements must naturally, like all other modificatu 
of the same kind, be summated and intensified in the course ot 
generations. The purposive character of the reflexes beo>: 
then readily intelligible, if we regard them as resulting from the 
voluntary action of previous generations ; while, on the < 
the view which sees in them the starting-point ot the wills 
development fails to explain their existence and purposivcness, 
and is further in disagreement with the results of objective and 
subjective observation, with objective : for the observation of 

228 Lectures on Human and Animal Psychology 

animals, and especially of the lower forms of animal life, never 
goes to prove the primitive character of reflexes which the 
theory assumes; and with subjective: for it remains com- 
pletely unintelligible how a decision of will can arise from purely 
intellectual processes. Introspection invariably points to feeling 
as the antecedent of will ; but feeling, as we saw above, is not 
separable from it, since it always implies a certain tendency to 
will in one way or the other. 

Moreover, this theory usually occupies itself exclusively with 
external voluntary actions, and entirely overlooks the fact that 
there is an internal volition manifesting itself only in the form 
of conscious processes. We direct our attention voluntarily 
upon any object which appears in our field of vision ; we are 
clearly conscious of an effort of will in trying to recall a word or 
a fact that we have forgotten. We voluntarily turn our thoughts 
in different directions in selecting out of a number of ideas 
which are passing through consciousness those which have most 
direct reference to the general trend of our thinking at the time. 
It is quite impossible to derive these internal voluntary processes 
from external voluntary actions. It is surely obvious that the 
contrary is true, that every external voluntary act presupposes 
an internal volition. Before we voluntarily execute any particular 
movement, we must have formed the decision to make it. And 
this decision is an internal voluntary process. Internal voluntary 
actions, then, are possible without external ; but external acts 
always require antecedent internal ones. 

External voluntary actions, therefore, presuppose as their con- 
dition internal volitional processes. And in like manner reflec- 
tion and choice between various possible actions (which are 
usually and wrongly held to be the essence of will) imply the 
pre-existence of simple voluntary actions. In these latter some 
definite object, whether it be something external or an internal 
idea, is willed, without any reflection or choice at all. Choice is 
nothing but a complex volun'.ary process. At first several 
motives to will are present simultaneously. Later some one of 
these, which accords with the decision we have formed, gains the 

Simple and Complex Voluntary Acts 22^ 

predominance over the others. If this predominance is 
sive enough to allow one definitely directed volition to prevail 
over the others, but not strong enough to give rise to an extern . 
voluntary action, we have simply a desire. If the inhibitions 
due to conflicting voluntary impulses arc gradually overonu.-. 
desire subsequently passes over into a voluntary action. This 
explains the fact that desire may exist in two forms, first as 
the state of mind preliminary to a voluntary action, and secondly 
as a permanent conscious process which does not give rise to 
any such action. If in the latter case there is connected with 
Mie desire the idea that it cannot be realised either for the 
present or at all, we have what is called a u'is/t. Desire, then, is 
mainly a matter of affection and conation, while in wish there is 
present besides these an intellectual process of considerable- 
intensity. But the popular view that desire is the uniform and 
necessary antecedent of volition is the result of the erroneous 
doctrine of will which we have just been discussing, and is entirely 
without foundation. It follows from the conditions which we 
have enumerated that desire may be present in the mind before 
a voluntary act occurs, but it is not indispensable ; indeed, it is 
probably absent oftener than it is present. Even in complex 
voluntary processes the action may take place before the state 
of desire has had time to develop. And in simple processes 
the possibility of desire is altogether excluded, since the internal 
voluntary action gives rise directly to the external without find- 
ing in consciousness any resistance to be overcome. Feelings, 
of course, occupy an entirely different position. They are 
always present alike in simple and complex conative processes, 
the only difference being that they are more complicated in the 
latter case than in the former. Before volition is realise 1 in 
action, the tendency of will is known, and this tendency is 
simply a matter of feeling. Feeling, therefore, is not a process 
different from volition, but simply a constituent of the comp'.i 
voluntary process. It is only because we have 
experience of feelings from which no voluntary acts at is 
we are able to separate the two processes, 
altogether impossible : voluntary activity always presupposes 
an antecedent voluntary tendency, i.e., a feeling. 

What is it that must be ad-led to feeling in order that a 

230 Lectures on Human and Animal Psychology 

volition may result ? This question has really been answered in 
what has just been said. The tendency to will contained in the 
feeling passes into a voluntary activity of the same direction. 
What exactly are we to understand by this activity, which, 
together with feeling,constitutes the chief characteristic of volition ? 
The concept of activity contains two factors, in the first place, 
activity implies a process or change in the given condition of an 
object, and, secondly, the reference of this change to some 
subject as its immediate cause. Thus in the physical sciences we 
speak of the chemical action of the electric current, of the 
mechanical action of wind and water, etc. The chemical 
decomposition of a liquid into its constituents, the movement of a 
mill-wheel, etc., are in these instances the observed changes ; the 
electric current, the moving water, and air are the subjects to 
which these changes are referred. So that we may ask in this 
matter of voluntary activity, what is the change that occurs, and 
what is the subject which we postulate to explain it ? In the first 
place, the change is always an alteration in our states of 
consciousness : an idea may arise which was not previously 
present, or an existing one may disappear ; or, again, the change 
may consist in an obscure idea becoming clearer, or a clearer one 
more obscure, etc. These ideational processes are further always 
connected in consciousness with various feelings and emotions. 
In external voluntary actions the changes which refer to move- 
ments of the body play the most important part. If we 
abstract from the active subject, it is muscle-sensations, and 
perceptions of movements and their results, which form the 
chief conscious constituents of an external voluntary action ; and 
all or some of them have now and again been regarded as the 
exclusive characteristics of volition. But it is surely evident 
that they do not exhaust the psychological analysis of will ; 
every one of the changes in ideational content to which volition 
may give rise can under certain circumstances occur inde- 
pendently of it. Ideas which are brought to consciousness by 
voluntary recollection may also crop up through involuntary 
association ; and muscle-sensations may be produced by re- 
flexes, or, as you know, by means of external and artificial 
stimulation of the muscles. What must be added to all this, 
therefore, is the reference to an active subject, which introspec- 

Simple and Complex Voluntary 


tion teaches us to regard as the direct cause of th<: id. -.tti 
changes. But what is this 'active subject '? The mo->t obvious 
answer appears to be : the willing subject is our own self. Hut 
that answer does not in any way assist our psychological an.dysi*. 
For what, again, is this 'self which we are led to look upon a-, 
the author of our voluntary actions? When we examine :t 
closely, we see that it is only another expression for the old 
phrase ' willing subject;' We perceive changes in our conscious 
content, and refer them to a single subject ; then we go on to 
name these changes ' voluntary actions,' and the subject brought 
in to explain them our 'self.' The only means of determining 
more exactly the nature of the 'self is to analyse out wh it we 
regard as the cause of our voluntary action in each particular 

Now the willing self is usually regarded as the immediate 
cause of voluntary actions, but by no means as their finai and 
only condition. We suppose that the will is determined by 
motives. We assume, of course, that a motive caniKt 1 
effective without a willing self; but, on the other hand, we regard 
it as equally obvious from the facts of our immediate internal 
experience that a willing self cannot act without motives. The 
connection between motive and will is, therefore, just as 
necessary as that between will and active subject. A retlex. or 
a passive movement which some external force compels us to 
make, is not conditioned by motives, although they have cause* 
of their own just as certainly as voluntary actions. Motives arc 
therefore causes of volition ; and since volition always arises from 
internal processes, it is at once clear that they must be / 
psychical causes. 

Now what is a motive ? It is customary to make a di-tin 
between simple and complex motives, and to comprehend under 
the latter rubric complex groupings of motives, where tin 
stituents may to some extent operate in ditierent 
But in giving an account of the particular cause 
mine volition, we shall only recognise as <, 
those which give it a definite direction, and wh 
forces, incapable of further analysis. 
is a particular idea icith an ^ fee tire to> 
since feeling is itself simply a definite voluntary tendency, tl 

252 Lectures on Human and Animal Psychology 

combination of idea and feeling in motives only means that an 
idea becomes a motive as soon as it solicits the will. Hence it 
is tautological to say that only ideas with a strong affective tone 
can operate as motives, since it is just the affective tone of an 
idea which gives it the power of acting as a motive. 

Nevertheless, introspection can show the conditions in virtue of 
which some ideas become motives and others do not. These 
conditions are of two kinds, they consist partly in the imme- 
diate attributes of sense-impressions, partly in the nature of our 
previous conscious experiences. All those attributes of sensation 
which endow it with a vivid affective tone serve also to make 
the impression effective as a motive to will. In this case it 
generally happens that the impression, with its strong affective 
tone, is the only motive present in consciousness : the voluntary 
action is a simple, or, as it is ordinarily expressed, an impulsive, 
action. There can be no doubt that the majority of the actions 
of animals are of this character. But impulses make up a large 
part of human action also, and especially in the earlier stages of 
its development. All sense-impulses are simply tendencies to 
will connected with definite sensations ; i.e., they are feelings 
which have a strong tendency to pass over into actual volition. 

But in course of time the mind acquires various dispositions 
toward the renewal of previous ideas which are themselves 
connected with definite voluntary tendencies. An external 
stimulus will not any longer simply call out the impulse corre- 
sponding to it ; but this impulse will increasingly tend to influence 
and be influenced by the dispositions already existing in the 
mind. These, again, may be transformed into conscious motives 
to will either by the external impression or by secondary 
influences. So that the chief motive of actual volition is hence- 
forth not some particular sense-impression which happens to be 
there, but the entire trend of consciousness as determined by 
its previous experiences. This trend or disposition does not, of 
course, come directly to consciousness as such. We can only 
give an account of those dispositions which enter into the con- 
flict of motives in virtue of their perception as ideas to which a 
strong affective tone attaches. And even of these many remain 
so indistinct, that though they may be factors in the resultant 
total feeling, and therefore in the act of will which finally results, 

Psychological Elements in i'olunlarv Ait ion 2,3 

there is still no clear perception on our own part of th'-u imie;>c:r. 
dent existence. On the other hand, we can know notum.; .-.h.u- 
ever of the influence which may be exerted by the dispositions 
that never become realised in idea at all upon the changes in 
our ideational content, and so upon the final act of will. The 
links which join the actual current processes with the pa-t 
history of consciousness simply serve to bring out with unnii- 
takable clearness the general fact that the determining ground 
of action has not been any single impression, nor any particular 
motive, whether called up by association or arising ' of itself,' 
but the entire trend or tendency of the mind, which has its roots 
in the original nature of consciousness and the accumulated 
experience of the mental life. A more or less intensive feeling 
is connected with this general idea, and becomes an essential 
element in the common feeling of the moment. The action which 
results from this plurality of conflicting motives we call a ioinfu-x 
voluntary action or a volitional action. It possesses two distin- 
guishing marks in consciousness, first, the feeling of a dfci*ion, 
preceding the action and based upon the connection of the 
present impression with past experiences ; and secondly, the 
idea of the voluntary act as determined by a choice between differ- 
ent and conflicting motives. Either one of these characteristics 
may be more or less distinct. The clearness of the perception of 
either usually stands in inverse ratio to that of the other. The 
feeling of decision is predominant where the voluntary act 
occurs at once and with complete certainty ; the feeling of 
choice prevails where there is a long preliminary conflict of 

These facts make it obvious that simple voluntary actions arc 
the necessary presuppositions of the more complex 
the former the impression does not cause the action 
its effect depends upon the state of consciousness 
But as this effect is relatively simple, the directly 
is the principal motive in the decision, and other mot 
no appreciable significance as compared 


If we bring together once more all the 

voluntary action, we see that it consists, in tlu nrst place. 

234 Lectures on Human and Animal Psychology 

feeling, in which the tendency of the will is manifested; secondly, 
in a change in ideational content, which may be accompanied by 
an external effect mediated by the organs of movement ; and 
thirdly, in the general idea of the dependence of this change upon 
the whole trend of consciousness. This last, like all secondary 
ideas, finds its principal expression in a feeling, which partly 
precedes the decision of will (in the form of the above-mentioned 
feeling indicating the volitional tendency), and partly accom- 
panies it. To these three constituents must be further added 
the feelings which arise subsequently as a result of the internal 
and external effects of the action, but which exert no influ- 
ence upon its performance. 

One very important attribute of volition, which affects all the 
elements of voluntary action which we have here cited, is its 
unity. Despite the conflict of motives and the oscillations of 
feeling conditioned by it, the voluntary act itself at any given 
moment must be single and unitary. This fact is the basis of 
the unity of the self. By a hysteron proteron which often recurs 
in psychology we tend to regard the latter as the cause of the 
unity of volition. But, as a matter of fact, what we call our 
' self is simply this unity of volition plus the univocal control 
of our mental life which it renders possible. Furthermore, this 
unity of volition enables us to explain directly another fact to 
which we have already referred, the fact that the feelings of 
each moment unite in a single total feeling, whatever oppositions 
may exist between them. This total feeling is the resultant 
volitional tendency. And it is just as impossible for it to be 
resolved into a number of independently coexisting feelings as 
it would be for us to will several different things simultaneously. 

In virtue of the attributes reviewed in these lectures, feeling 
and will react upon the ideational side of our mental life ; and 
thus help to determine the entire content of what we call, by an 
arbitrary distinction, but one which is of service in the analysis 
of the facts, consciousness. Now that we have described the 
various constituents which go to make up the mental life, we 
will turn for a time to the phenomena which result from the 
combination of all of them. 




WHAT is 'consciousness'? Much attention has been 
devoted to this question in modern times both by 
philosophers and psychologists. There could be no doubt that 
the word denoted some phase or aspect of our mental life, and 
was not identical with any of the other concepts, like 'idea,' 
' feeling,' ' will/ etc., which we apply to particular mental pri- 
cesses and states. So that the view naturally suggested itself 
that consciousness is a special mental condition, requiring to be 
defined by certain characteristic marks. And the feeling that it 
was necessary to oppose to consciousness an unconscious mental 
existence promoted this opinion. Ideas, affective proces.* 
may vanish and then again appear. It is therefore inferred 
that after leaving consciousness they have continued to exist 
in an unconscious state, and at times return to their former 

From this point of view, nothing is more natural than t<- 
think of consciousness as a kind of stage upon which our iiii-a-* 
are the actors, appearing, withdrawing behind the 
coming on again when their cue is given. And the notion has 
become so popular that many philosophers and psycho! 
sider it much more interesting to learn what takes place behind 
the scenes, in unconsciousness, than what occurs in consciousn 
Every-day experience, it is supposed, has made the latter familiar 
to us; but we know nothing of the unconscious, and to 
something about it would be a really interesting addition to 

236 Lcctni-es on Human and Animal Psychology 

Nevertheless this comparison of consciousness to a stage is 
entirely misleading. The stage remains when the actors have 
left it ; it has an existence of its own, which is not dependent 
upon them. But consciousness does not continue to exist when 
the processes of which we are conscious have passed away ; it 
changes constantly with their changes, and is not anything 
which can be distinguished from them. When the actor has 
k-ft the stage, we know that he is somewhere else. But when 
an idea has disappeared from consciousness we know nothing 
at all about it. Strictly speaking, it is not correct to say that 
it subsequently returns. For the same idea never returns. A 
subsequent idea may be more or less similar to an earlier one ; 
but it is probably never exactly the same. Sometimes it has 
constituents which the earlier idea had not ; sometimes certain 
of those which belonged to the latter are lacking in it. There 
is scarcely any view which has been a greater source of error in 
psychology than that which regards ideas as imperishable objects 
which may appear and disappear, press and jostle each other, 
objects to which, it is true, additions are at times made through 
tne action of the senses, but which, when once they have come 
into being, are only distinguished by the variation in their 
distribution in consciousness and unconsciousness, or at most, 
by the different degrees of clearness which they possess in con- 
sciousness. As a matter of fact, ideas, like all other mental 
experiences, are not objects, but processes, occurrences. The idea 
which we refer back to a previous one, when we apprehend it 
as similar to that, is no more the earlier idea itself than the word 
which we write or the picture which we draw is identical with 
the same word which we wrote previously or the similar draw- 
ing which we made sometime ago. Indeed, you will see, if you 
consider the complex conditions under which our inner experi- 
ence arises, that nothing like the same degree of similarity 
between the earlier and the later product can be expected here as 
may be found under certain circumstances in the field of external 
actions like writing and drawing. The circumstance that new 
processes exhibit relations and similarity to others previously 
existing, can no more prove the continued existence of the idea 
as such, than it can be inferred from the similarity of the move- 
ment of the pen in writing a definite word now to that involved 

Ike Concept oj LOH.M loiisiu** 

on a former occasion, that this movement has continued to e.\M 
in an invisible form from the time it was first made, and ha 
simply become visible again when we have written the 
anew. If ideas are not imperishable facts, but transitory pro- 
cesses which recur in more or less altered form, the whole <.f tin's 
hypothetical structure falls to the ground. And at the same 
time the unconscious loses the significance ascribed to it as an 
especial kind of mental existence, which, though not itself c.n- 
sciousness, might at any rate enable us to determine the character- 
istics or conditions which must attach to the objects of mind in 
order that they may become conscious. 

In the same way, all attempts to define consciousness as a 
particular mental fact co-ordinate with our other internal experi- 
ences have proved fruitless. It is obvious that those u ho w u!d 
regard it as the capacity of internal observation, as a kind of 
'internal sense/ commit in this analogy an error similar to that 
involved in its comparison to a stage. The perceiving organ 
and the perceived object are two different things ; consciou:,ne 
and conscious process are not. The activity of observation, ot 
attention, is of course found among what we call conscious pro 
cesses. But it is just one conscious fact, co-ordinate with tin- 
rest, a fact which presupposes the existence of consciousness. 
not one which renders consciousness possible. The same criti- 
cism applies to yet another explanation which is -,onietiines 
given. We distinguish in consciousness, it is said, a whol- 
number of ideas. Therefore consciousness must possess the capa- 
city of discrimination ; the word must be equivalent to discriiri 
nating activity. But here again the question arises whether t In- 
discrimination of processes directly perceived is the antec 
condition of these processes, or whether it is not rather a ir- 
to which they are essential. In the first place, the < 
be there to be distinguished. The child runs together anum er 
of separate objects into a single idea, where tin 
consciousness keeps them separate. Discrimination, t 
observation, consists in processes which presuppose < 
ness, and which consequently cannot cons! 
And consciousness itself is not a particular mental proi 
ordinate with others ; it consists entirely in th< 
have internal experiences that we pc 

238 Lectures on Human and Animal Psychology 

feelings, and voluntary impulses. We are conscious of all these 
processes in having them ; we are not conscious of them when 
we do not have them. Such expressions as ' the limen of con- 
sciousness,' ' appearance in and disappearance from conscious- 
ness,' etc., are pictorial ways of speaking, useful for the brief 
characterisation of certain facts of internal experience, but 
never to be regarded as a description of these facts. What 
really takes place in the raising of an idea above the limen of 
consciousness is, that something occurs which had not occurred 
previously. And what really happens when an idea disappears 
from it is that some process ceases which has hitherto been in 
progress. In like manner we must think of the range of con- 
sciousness as denoting simply the sum of mental processes 
existing at a given moment. 

Although, therefore, consciousness is not an especial kind of 
reality co-ordinate with the particular facts of consciousness, 
modern psychology still finds the concept indispensable. We 
must have a collective expression for the whole number of 
mental experiences, given either simultaneously or successively. 
As simply denoting the existence of internal experiences, while 
leaving their nature altogether undetermined, the concept is 
especially serviceable for the treatment of the interconnection 
of the mental facts, of all those processes with which we have 
already become acquainted in isolation. It has no meaning 
apart from its reference to this interconnection of simultaneous 
and successive mental processes ; and the problem of conscious- 
ness consists in determining how the particular phenomena are 
interrelated, and how their relations and connections again 
combine to form the totality of mental life. For the sake of 
simplicity in treatment, it will be convenient to confine ourselves 
at first to the ideational side of consciousness, and then, when 
we have discussed the problem just formulated from this point 
of view, to supplement our results by reference to the affective 
and conative elements. This is of course the plan which we 
have followed in our analysis of particular mental processes. It 
will, however, soon become evident that in dealing with the 
interconnections of mental processes we cannot carry our 
abstraction through to the end, since the affective side of our 
mental life constantly exercises a determining influence upon 

Condition of Ideas in Consciousness 239 

the combinations and relations of ideas. In certain, 
therefore, we shall not be able to avoid at least a passing rcier- 
ence to the affective and conative factors. 

The first question which may be raised within the limits of 
the conditions laid down runs of course as follows : luw many 
ideas may be present in consciousness at a ^ircn moment / The 
content of this question is not quite so precise a-^ its wording 
seems to imply. The estimation of the number of constituents 
which a whole contains is, naturally, dependent upon we 
regard as the constituent unit. Now, even if we neglect the 
continual change among ideas, their combinations in conscious- 
ness are enormously complex. So that it may easily be a matter 
of doubt whether some given portion of conscious content is to 
be considered as an independent idea, or 'as a part of a more 
comprehensive idea. We may here dispense for a time with 
any final theoretical solution of this difficult preliminary problem. 
For our present purpose it will suffice if we can furnish a practical 
criterion. We shall accordingly regard an idea as separate and 
independent when it is not connected by customary association 
with other ideas simultaneously present. If, e.g., there is placed 
before the eye a number of letters in serial order (say, x r (>, 
we shall consider that each one forms an independent idea by 
itself, in spite of its spatial association with the rest. For, re- 
garded as a whole, letters form no new complex idea, capable 
of entering into definite connections with other ideas, 
other hand, we perceive four such letters as u< <> /- k, we 
not hold them to be independent ideas, at least, lor one- 
reads them as a word, but shall look on them as combined into 
a single complex idea. From these considerations there- 
two results, which should never be lost sight of in experiments 
made to determine the ideational range of consciousness 
we must always decide from the objective and subjective . 
ditions operative in each particular instance < 
pass as independent units and what not. It is, 
obvious that the same objective impression may in one case be 
apprehended as one idea.and in another as more than one.arcord- 

240 Lectures on Human and Animal Psychology 

ing to the subjective conditions involved. Secondly, the con- 
clusions derived from ideas of one kind will not necessarily hold 
for ideas of any other kind. In particular \ve may expect to 
and that the range of consciousness will be smaller for complex 
: deas than for relatively simple ones. 

When the question of the range of consciousness was first 
aised, these conditions were entirely overlooked, and the general 
nethod of investigation pursued was not one which could lead 
:o any certain results. Conclusions were either deduced from 
:ertain metaphysical postulates, e.g., that the mind, as a simple 
Deing, could only contain a single idea at a given moment, or 
;he investigations were based solely on introspection. Any 
3ne may convince himself of the fruitlessness of this latter pro- 
cedure by seriously asking himself the question : how many 
ideas do I now find in consciousness ? And at the same time 
the experiment shows him the reason why his efforts at an 
answer are without result. The question is scarcely raised be- 
fore the moment to which it refers has passed, nor can the 
following moment be fixed any more successfully. It thus be- 
comes quite impossible to distinguish what is simultaneously 
given at a particular moment from that which comes later. 
This defect of direct introspection, however, itself shows us how 
we should endeavour to supplement it by experiment. It is 
only necessary so to arrange the conditions of experimentation 
that the confusion of simultaneous impressions with successive 
is less easily possible. This we can do by momentarily present- 
ing a number of sense-impressions, which are capable of becom- 
ing independent ideas, at a given signal, and then trying to 
determine how many of them have been actually ideated. Xow 
it would be wrong to suppose that the running together of the 
momentary impression with subsequent ideas is here altogethei 
precluded. Suppose, e.g., that by a momentary illumination we 
present to the eye a number of visual objects. The perceptions 
of the first moment will naturally be supplemented by others of 
which we do not become conscious until later. You may easily 
convince yourselves of this by holding a book in the dark at a 
convenient distance from the eye and illuminating the room 
for an instant by an electric spark. Even if in the first moment 
you only cognised a single word, it may very well happen that 

Condition of Ideas in Cou^um^nc** 241 

you will subsequently, by the aid of memory, be able cle.-iriy to 
apprehend others. Indeed, what is read subsequently in this 
way is often more than what was recognised in the first instance. 
But these experiments, again, lead us to a further fact, u'nich 
shows that it is possible to draw valid inferences from them 
with regard to the condition of consciousness at a given moment. 
We can very clearly distinguish the image that has been gradually 
reconstructed on the basis of the original impression from the 
image which corresponds directly to it. This is due to the fact 
that the particular moment is not precisely like those which 
directly preceded or follow it : the sudden appearance and dis- 
appearance of the flash of light marks it off from them ; this 
distinction makes it less difficult for introspection to neglect or 
voluntarily to exclude the subsequent filling out of the original 
ideational image. Here, as everywhere, psychological experi- 
ment does not enable us to dispense with introspection, but, on tlu 
contrary, renders introspection possible by furnishing the condi- 
tions which it requires for exact observation. 

Experiments of this kind with momentary impressions may be 
made in any sense-department. But visual impressions are be- 
suited to the purpose, because they can most easily 
with a view to their apprehension as independent ideas. Th 
impression itself is, it is true, not entirely moment. u 
stimulation has a physiological after-effect. However, i;. 
impressions which pass very quickly this atter-etteci is s< 
that we may neglect it for our pre-ent purpose. Visual 
ments are made with the apparatus represented in I':. 
is intended for the demonstration of the ph 
audience. If you should merely wish to perform the ex 
upon yourselves, the dimensions could, ot curst 
smaller. The apparatus consists of a black- 
grooves in front of a black vertical board, some two metro 
and falling when the spring F is pressed. In the screen ti 
a square opening large enough to enclos- a lar-i 
objects that can be separately ideated, such a- 
alphabet. This opening is so placed that when the .- 
raised it shows simply the dark background, but .hiring i 
passes very quickly before the ob; 
instantly covers them again. On the part of the screen 1 

242 JLectureS on Human and Animal Psychology 

the square opening a small \vhite circle is so placed that before 
the fall of the screen it exactly covers the centre of the visual 
surface afterwards briefly exposed to view. This circle serves as 
a fixation-point to put the eye in the most favourable position 
for the apprehension of the impressions. A in the figure gives a 

lateral, and B a front, 
view of the apparatus. 
In A the screen is 
raised, and covers the 
objects which are to be 
seen ; in B it is repre- 
sented in the moment 
of falling before them, 
so that a number of 
impressions ( letters 
taken at random) have 
just become visible. 
If we imagine the 
motion continued, we 
shall have these hidden 
again the next moment 
by the upper, closed 
part of the screen. The 
size and distribution of 
the visual objects and 
the distance of the 
observer from the 
apparatus must be so 
chosen that all the 
separate letters which 
are to be seen shall 
fall within the region 
of clearest vision. It 
B - is of course true that 

in these experiments 

every visual object is, strictly speaking, visible not only for a 
single moment, but for a measurable, although relatively small 
portion of time ; and, further, that this time is not exactly the 
same for the different objects. In the apparatus represented in 

Condition of Ideas in Consciousness 24^ 

Fig. 37, the upper line is visible for 0-09", the 1-wer for GO; . 
and the middle for OX)8". These times arc. houcver, -o small in 
comparison with the much greater duration of the- after-image 
that for our present purpose they may be regarded as really 

Experiments of this kind show that four, and sometimes even 
five, disconnected impressions (letters, numerals, or lines of 
different direction) may be distinctly perceived. If the .separate 
impressions are so arranged that they enter into combination 
with one another in idea, the number become* three times as 
great. Thus we are able to cognise instantly two dissyllabic 
words of six letters each. 

But the result of such investigations is to call our attention to 
other phenomena, which render it obvious that we cannot really 
learn anything in this way regarding the total range of consciou>- 
ness. We notice that the letters, numerals, words, etc., which we 
clearly apprehend at the moment when the screen falls, by no 
means exhaust the conscious content of the moment. Ik-sides 
these impressions which are clearly apprehended, there are- 
present in consciousness others which are less distinct, or wholly 
indistinct. In addition to the four or five letters which you \\ere 
able to read, you would have noticed, e.^., some of which \ 
cognised only the approximate outline, and others about which 
you had only the quite indefinite idea that they uere present and 
were visual impressions. The experiments show, theretre, that 
this method can only enable us to determine the number of v 
and distinct ideas present in consciousness, and can L 
information of the entire number which it contains, 
of clear ideas for the sense of si-ht amounts to 4 or 
they are comparatively simple and familiar ; if they are complex 
the number varies from I to 3, according to the de-re., 
complexity. In the latter case the number of simple 
present in a clear ideational complex may be as many 
You will notice, further, that the impression falling dire, 
the yellow spot is usually more distinctly apprehended than any 
of the others. But this is by no means necessarily the . 
laterally seen objects may take precedence over the 
seen directly, especially if the attention is voluntarily directed 
upon them. 

244 Lectures on Huviaii and Annual Psychology 


So that, even although our first method has told us nothing of 
the actual range of consciousness, it is worth while to spend a 
little time in examining the results obtained. Apart from the 
information they afford us as to the number of ideas clearly 
apprehensible at a single moment, their indication of different 
degrees of ideational clearness is particularly worthy of remark. 
It is true that the distinction of clear and obscure ideas did not 
escape the keen observation of Leibniz, and since his time has 
scarcely been disputed. But ordinary introspection does not 
admit of such definite and direct determination of the relations 
of the different degrees of clearness as that afforded by the 
method of instantaneous impressions. The experimental method 
demonstrates, e.g., the correctness of Leibniz' hypothesis that 
there is no abrupt transition, but always a continuous gradation 
from each degree of ideational clearness to the next. In the 
experiment with momentary visual impressions given above, we 
distinguished three kinds of ideas in consciousness : the clear, 
the more obscure, where a partial discrimination is still possible, 
and the quite obscure, in which we only cognise the presence 
of some conscious content belonging to a definite sense-depart- 
ment. We must now understand that these ideas only differ in 
degree of clearness, and that all three degrees are connected by 
continuous gradations. For the two extremes, however, we may 
employ the terms which Leibniz introduced. We may term the 
/appearance of an obscure idea in consciousness a perception, and 
\the appearance of a clear one an agjiercgption. These two names 
must not be understood to carry with them any presuppositions, 
either metaphysical or psychological. They merely express a 
fact, for which (as is usual in science) we choose the name pro- 
posed by the investigator who first called attention to it. We 
leave out of account any assumption, any theory, derived from the 
observed facts which Leibniz and his successors may have con- 
nected with these terms. Notice only that the relation of clear 
to obscure ideas furnishes an obvious analogy to that of objects 
distinctly or indistinctly seen in the field of vision ; and that it 
is therefore natural to refer the distinction of perception and 
apperception to consciousness itself, just as in external vision we 

Clearness and Distinctness of Ideas 2.\ 5 

account for the different degrees of distinctness by refercno- to 
differences in acuteness of vision in different p.-rtions of the 
visual field. We may s.iy, then, that thc/v/vv/rvd ideas are t!..i> c 
which lie in the field of consciousness, while the apperuntd are 
situated at its fixation-point. 

Now what are we to understand by the clearness of an idea * 
The word, like all the names of psychological conceptsfhas been 
transferred from external objects to the consciou-. subject. We 
use the term 'clear' to denote luminous or transparent objects ; 
i.e., those which are themselves easily perceived, or which aid the 
sense of sight in perceiving others. When the word is applied 
to consciousness, therefore, it must express some similar cha- 
racteristic in introspection. AjjJckajs^clearjAiiicn Jt^ii appre- 
hended in introspection m_ore_ perfectly than others which^ in 
contraTITstinction to^it^ajre.. called .obscure. The only difference 
between the original and the transferred meaning of the word* 
is this : in the former case the property of clearness may belong 
to the object without reference to our perception of it. but in 
the latter the idea is only clear in so far as it is clearly perceived 
in consciousness; a difference which again has its source in tlv- 
fact that our perceptions of mental processes ami the mental 
processes themselves are completely identical. Ideas are only 
ideas in virtue of our perception of them. Internal perception 
(introspection) is simply the fact of internal experience itself ; 
and we are only looking at this experience from different points 
of view when we speak of it at one time as idea, feeling, etc , 
and at another sum it up as internal perception. 

We are apt to identify the clearness of ideas with th-.-ir 
distinctness, and to define one of these concepts by the 
'distinct' being 'what can be clearly cognised,' or 
'what we perceive distinctly.' Now it must be admitte. 
the two properties are generally found together. Hut they are 
not at all identical; each of them denotes a ditteivnt a-pe< 
or a different reason for, the advantage which a given idea 
possesses in consciousness. An idea is ' clear 
of its own properties, just as in the use of the 
reference pure water is termed 'clear.' and not ' distinct,' because 
it is transparent, so that any object that happens to be in it in. 
be seen through it. On_Jlie__ oilier, hand, an idea i> 

246 Lectures on Human and Animal Psychology 

' distinct' with reference to the jjejfiniteness with_which_jt_ is 

marked off from other, ideas. Thus an object lying in clear 

water is distinctly seen because it stands out sharply from its 
surroundings. Similarly a tone is clear when we can fully 
apprehend its peculiar quality ; it is distinct when definitely 
distinguishable from the other elements of a compound clang, or 
from other simultaneous sound-impressions. 

As applied to our ideas, then, clearness and distinctness 
denote properties which depend directly upon the activity of 
ideation, or, what is the same thing, of introspection. One and 
the same idea under the same objective conditions may be at 
one time clear, and at another more or less obscure. And for 
this reason we must be especially careful not to confuse the 
clearness of an idea with its intensity. That is simply dependent 
upon the intensity of the sensations which constitute it. The 
intensity of perceptual ideas is determined by the strength of 
the sense-stimuli, that of memorial ideas by other conditions, 
which have nothing to do with ideational clearness. At the 
same time intensity usually promotes clearness and distinctness. 
Other things equal, the more intensive idea is usually the clearer, 
and very weak ideas in particular are but seldom clear and 
distinct. Nevertheless it may happen, if the subjective condi- 
tions of perception so determine, that an intensive idea is obscure 
and indistinct, and a weak one clear and distinct. A very weak 
over-tone in a clang, e.g., may be heard clearly and distinctly, 
while the more intensive ground-tone is less clearly perceived, 
and a loud noise simultaneously given perhaps hardly noticed at 

It follows from all this that the clearness of ideas necessarily 
depends upon the condition of consciousness for the time being. 
Inasmuch as they help to determine this condition, the intensity 
of impressions and memorial images also exerts an influence upon 
the clearness and distinctness of ideas. But since the state of 
consciousness is certainly not entirely dependent upon those 
conditions, they are by no means the only factors of importance 
in the matter. Our final definition of clearness must, therefore, 
be that it is that property in virtue of ^cvliich an idea has an 
advantage over other ideas in introspection. But it is not difficult 
to see that this definition is only a description of the word 

Phenomena acco^tpaiiviii^ . //yVmyV/^// 247 

'clear.' As a matter of fact, it is just as impossible to d< -MM- the 
clearness of an idea as to define the intensity or quality >!' a 
sensation. We can distinguish these fundamental properties of 
our mental processes from each other by showing that undei 
definite conditions they vary separately and independently. 
But the differences could no more be brought home to one wlu, 
had not experienced them than can the di.-uinctioiis ul~ cuiour tu 
the congemtally blind. 


On the other hand, the becoming clear of an idea is regularly 
associated with other mental phenomena, which not only as.M^t 
the introspective discrimination of clear and obscure ideas, but 
also throw some light upon the subjective conditions of the 
processes which we have distinguished as perception and apper- 
ception. These phenomena are of two kinds, they consist 
partly of sensations, partly of feelings. The sensations which 
Accompany apperception belong to the category of muscle-sen- 
sations. They are especially noticeable in cases of external 
sense-perception. If we are directing our attention upon a 
particular tone, or a particular visual object, to the exclusion 
of other impressions of lu;ht and sound, we have in ear or 
eye definitely graduated muscle-sensations, which are probably 
referable to the tensor tympani and to the muscles subserving 
the accommodation and movement of the eye. The same sen- 
sations may be perceived, though less clearly, to accompany 
memorial ideation, at least when the ideas are vivid, 
object seen with the mind's eye is referred to a certain di.*.tanc 
from us, and we consequently accommodate the muse 
apparatus of our eye to it. The tones of a me! 
recall in memory may give rise to a tension in the ear as clearly 
perceptible as though they were real. Kven the fainter pk 
ideas which constitute abstract thinking are not wholly without 
this sense-accompaniment. When we are trying to remember 
a name or are pondering a difficult problem we notic 
presence of strain-sensations. These are partly 
eye, visual ideas being, as you know, predominant in conscious- 

248 Lectures on I'lninan and Animal Psychology 

ness, partly in the forehead and temples, where the muscles 
lying' directly beneath the skin, which play a part in mimetic 
movements, are strained to a degree more or less proportioned 
to the amount of internal effort. 

The connection of these muscular tensions with mimetic ex- 
pressive movements leads us directly to the second accompani- 
ment of the apperceptive process, to feeling. There could be 
no emotional expression without feeling. Feelings precede ap- 
perception proper, and continue to exist during the course of the 
process. They are different in the two cases, though their 
passage in consciousness is continuous, so that those which 
precede apperception and those which accompany it form an 
affective totality, which by the fact of its continuity resembles 
emotion (of which we have to speak later), and which indeed not 
infrequently becomes actually transformed into emotion. Oui 
perception of these attendant feelings, like our perception of the 
sensations discussed above, is most distinct when the clearness 
of their ideational substrates is very great, and especially when 
this clearness is mainly the result of the trend of consciousness, 
itself, and not of external conditions ; e.g., to put it again con- 
cretely, when we are voluntarily recalling an idea which we 
have previously had, or when we are expecting an impression, 
etc. Even when the condition is not one of expectation proper, 
the feeling preceding apperception is very closely related to that 
of expectation. The feeling accompanying the process, on the 
other hand, may be compared to that of satisfaction, to the re- 
laxation of a tension, or, again, if the expectation is not realised, 
to that of disappointment or failure. It is quite true that these 
feelings are only clearly perceptible under the special conditions 
which characterise expectancy, recollection, etc. But careful in- 
trospection seems to show that feelings of the same kind are 
never entirely absent where ideas which were formerly obscure 
become clearer, even although their intensity be much less and 
their quality exceedingly variable. At least, if there is any es- 
sential difference, it is only in the case of the antecedent feelings. 
Their period of duration, e.g., may be very much shortened 
(though they hardly ever entirely disappear) when the object of 
apperception is an external sense-impression, or vhen memorial 
images crop up unexpectedly. 



The whole circle of subjective processes connected with apper- 
ception we call attention. Attention contains three essential 
constituents : an increased clearness of ideas ; muscle sensations, 
which generally belong to the same modality as the ideas ; and 
feelings, which accompany and precede the idcational change. 
At the same time the concept of attention proper has n>. 
reference to the first of these processes, but only to the last two 
Apperception, therefore, denotes the objective change set up in 
ideational content, attention the subjective sensations and feel- 
ings which accompany this change or prepare- the way lor it. 
Both processes belong together, as parts of a single psychical 
event. It may happen in certain instances that the objective 
effect is distinct, while the subjective aspect of the process does 
not attain a liminal intensity. Or it may happen, as when an 
expectation is unfulfilled, that the subjective constituent attains 
a great intensity, while the objective is entirely overshadowed. 
But these are only extreme cases of a series which, like ah mental 
series, contains terms arranged without break or interruption. 
Attention in the wider sense is not, and this is the important 
point, a special activity, existing alongside ot its three consti- 
tuent factors, something not to be sensed or felt, but itsei 1 " 
productive of sensations and feelings. No! in terms of our own 
psychological analysis at least, it is simply the name of the com- 
plex process which includes those three constituents. Their 
nature makes it plain enough why we regard attention a- sub; 
tive activity, without our needing to assume any special conscious- 
ness of activity independent of the other mental element.- 
concept of activity always presupposes two tilings, first, a change 
in the condition of something ; and secondly, a subj 
states vary with that change in such a way that the tv. 
exactly correlated. \Ye then regard the 
subject, and the changes set up as the eiiects of it- a I 
Now the sensations and feelings constituting ;i 
accidentally and equivocally ass ci.ited 
idea, but stand in very definite relation to it. 
strain-sensations and the preceding or .iccompai.yn 
entirely governed by the nature of tin: appi rc. 

250 Lectures on Humau and Annual Psychology 

changes, they change also. So that the phenomena which go to 
make up the apperceptive process possess all the characteristics 
required by the notion of an activity proceeding from an acting 
subject. This acting subject is given us simply and solely in the 
sensations and feelings accompanying the act of apperception. 
And since we find among these elements not only a constant 
alteration, but also a continuous connection of earlier processes 
with later, we come to regard the acting subject as persisting 
through all its changes. Language has given an expression 
to this view, which has been of determining influence upon the 
further conceptual development of the distinctions in question, in 
constantly rendering the notion of the permanent subject by 
the sign for the first person in simple verb judgments. 


It is in this way that the concept of the self (' I ') arises : a 
concept which, taken of itself, is completely contentless, but 
which, as a matter of fact, never comes into the field of intro- 
spection without the special determinations which give a content 
to it. Psychologically regarded, therefore, the self is not an idea 
among other ideas ; it is not even a secondary characteristic, 
common to all or to the great majority of ideas ; it is simply and 
solely the perception of the interconnection of internal experience 
which accompanies that experience itself. Now we have already 
seen that perceptions of this kind, perceptions which refer to 
the occurrence of a process, the manner in which it runs its course, 
and so on, are sometimes transposed back again into ideas. 
There is a deep-rooted tendency to hypostatise mental events, 
a tendency evinced by those theories which have regarded 
ideas themselves as permanent objects (pp. ?2i, 222). And 
there is a very special tendency to transpose the ' self into 
an idea of this character, though, as a matter of fact, it is 
nothing more than the way in which ideas and the other mental 
processes are connected together. Since, further, the manner 
of this connection at any particular moment is conditioned by 
preceding mental events, we tend to include under the term 
'self the whole circle of effects which have their causes in 
former experiences. The 'self is regarded as a total force, 

Self -COHSClOtt 'Slti'SS 2; I 

which determines particular events as they happ'.-n, mile- of 
course, they are occasioned by the action of tr.xu.-mai iinprcs 
sions or of those internal processes which we experience- iu>t .is 
passively as we do the external. And since the eiin.t 
of the preconditions of consciousness is the determination of tile- 
appearance and degree of clearness of idea*, we further bring the 
'self into the very closest connection with the proces* of ap- 
perception. The self is the subject which we supply for the 
apperceptive activity. It is plain enough that there is involved 
here a transference of the relations observed in external pcrccp 
tion to the sphere of internal experience. The self, you .see. i> 
regarded after the analogy of external objects, which we take t 
be the same in spite of variation in their properties, because the 
variation is always continuous both in time and space, lint 
without the continuity of our mental lite we should nt be abi- 
to cognise the continuity of objective things ; so that in : 
interplay of developments we have the self figuring both a> 
cause and effect. The perception of the interconnection < 
mental processes, which crystallises in the concept of the 
renders possible the distinction between objects and theirc 
ing properties ; and this distinction in its turn inclines 
ascribe an objective value to the concept 

There is another reason for this in the tact that the body, u 
which all the states of the self are connected, is itself an L 
object. In the first place, the self is a product of tw 
external perception and internal experience ; it is the 
gether with the mental processes connectet 
reflection destroys this unity ; but even then there rcmai 
faint trace of that object-idea which attached to the : 
And where the current view of lite is the practical one 
naive sense-reference, the body takes its place unqut 
as an inexpugnable constituent of the self. 



IMULTANEOUSLY with the development of self-con- 
consciousness, which we described in the previous lecture, 
proceeds the development of another complex process, that of 
attention. The two developments are in many respects similar. 
States of attention, like those of self-consciousness, present 
certain external differences which may be regarded as opposites ; 
though it is true that, to place the opposition in a clear light, 
we must more or less neglect intermediate processes which 
would enable us to pass from one to the other. For the 
extreme cases, however theoretically possible, never actually 
occur in the purity in which they can be obtained by analysis. 
However, if we disregard the concrete for a moment, we shall 
find evidence enough for the general possibility of the extreme 

We saw that in every act of apperception there are two 
principal conditions of the resulting effect, -first, the momentary 
condition of consciousness, itself determined partly by external 
influences, partly by those of its own earlier states which are 
directly related to these influences, and therefore with greater 
or less regularity associated with them ; and secondly, the 
entire previous history of consciousness, which may operate in 
the most various ways to alter the effect due to this momentary 
state. You must not, of course, suppose that these two con- 
ditions are at work in the individual case in the sense of two 
opposing forces. That would be impossible. For the earlier 

Development of .-itunnon 

states which are directly connected with any part, 
objective impressions, ideas, feelings, or whatever these st.ites 
may be, themselves form part of the previous mental rmtory. 
In other words, we have to do with a difference in decree, and 
not with a difference in kind. But this does not prevent the 
results in the two cases from appearing as opposite*. Suppose 
that the direction of the attention is determined merely by 
some chance stimulus, and by a 'state of mind 1 which is per- 
manently associated with that stimulus, or has been brought 
about by accidental circumstances. Then the immediate 
impression which we have of our internal experience i> th..t of 
a passive receptivity of what is going on in our minds. Sup- 
pose, on the other hand, that the direction of the attention i* 
determined by more remote conscious tendencies which have 
arisen from previous experience, and which are not directly 
related to the particular impressions of the moment. Then we 
have always the impression of a productive activity. Apper- 
ceptions of this kind we regard as actions of our ' self ; 'self 
just an expression for the total effect exerted by our previous 
mental experiences as a whole, without particular reference it. 
any special components of those experiences, upon the mei.t.i: 
processes which are running their course at any ^iven mom n:. 
To make this difference clear, we will call the first f>rm f 
attention ' passive,' and the second 'active,' att'-ntion. L_i me. 
however, warn you again, even at the risk- ol repetition, in 
calling attention 'passive' we by no means ieny to it any 
character of activity, that is, decline to see in it the operati >n of 
previous experiences. The contrary is true: such exj>eru 
are always operative ; it is only that the ext-nt and dire*, 
of their influence are limited and circumscribed. 
of course, to be thought that external influences ami t:u 
of mind that follow from them are wholly without e 
case of active attention. It is only true that they jetite in:, 
the background before the ir.tluence > of di-.po->itions cit.ibl 
during long periods of time and strengthened by mutual 
nection ; they are none the !es- 
these dispositions. To repeat once more. 
extreme cases, which can never 
the processes on which they dep"iul 

254 Lectures on Human and Annual Psychology 

continuous process-series. In both cases consciousness is 
functioning in the same way ; the difference is only a difference 
of more and less, of a greater and a narrower range. 

If we could only appeal to the ideational side of mind, then, 
we should not seldom find it difficult to decide in a particular 
case whether a particular apperception were active or passive. 
So that here again we find feeling playing a large part in our 
immediate apprehension of our own actions. You remember 
the general characteristic of feeling, that its peculiar quality 
gives expression to the total attitude of consciousness. In the 
present case, the presence of active apperception is invariably 
and unmistakably indicated by a feeling of activity. We can 
no more describe this than we can any other feeling ; we can 
only attempt to determine it by enumerating those of its con- 
ditions which belong to the ideational side of consciousness (cf. 
Lecture XIV.). The degrees of intensity of this feeling give us 
a direct measure of our own activity ; that is, of the preponde- 
rance of our whole mental nature over momentary and transitory 
excitations. There can be no doubt that we must regard it as 
a total feeling in our previous sense of the word (p. 219). It 
determines the attitude of consciousness at any moment. But 
its own peculiar and variable quality is itself determined by the 
special feelings dependent upon concurrent ideas and their 
mutual connections. Even passive apperception, therefore, has 
its attendant feelings ; only that these are associated to form 
a total feeling with a character of its own, either exclusively 
conditioned by the quality and intensity of the ideas that 
happen to be present in consciousness, or (and this is especially 
the case in apperception of very intensive external sense- 
impressions) consisting in a feeling of inhibition, which appears 
to arise from the sudden arrest of existing tendencies in the 
formation of ideas. In its latter form it may be intensified by 
sense-feelings of unpleasantness or pain, without, however, 
being dependent upon these. 

Considered from the point of view of these attendant feelings, 
the process of apperception and attention appears in a con- 
nection which points at once to those elementary mental 
processes which we have already discussed. Feelings we found 
to be invariable forerunners and concomitants of volition. 

Development of Attention 255 

They indicate the direction which an act of will will Ml,,\v 
before it has itself become conscious ; and when it ha-, attained 
its full force, they are still present to colour and explicate it^ 
effect. A second characteristic of volition beside feeling is an 
alteration in the state of consciousness on its ideational side, 
referable not to external influences, but to past mental disposi- 
tions. Both characters attach to the process of apperception. 
and, since the conditions of each form of this p:,ce over 
into one another without interruption, attach to active and 
passive apperception alike. For the raising of an idea t<> a 
higher level of distinctness can only come about in passive- 
apperception when there are present certain p. .<:tive nxntai 
dispositions to favour its preference. So that ideas and tin- 
feelings that are connected with them serve as motives to tin- 
act of apperception, while apperception itself shows all the 
characteristics of an act of will. More than this, its two funda- 
mental forms, the active and passive, obviously corre.-pond to 
the two fundamental forms of conative activity, the passive- 
form, the impulsive act, and the active, or act of choice. \\ e 
act impulsively when we apprehend an impression uniier the 
constraining influence of external stimuli, and of the ideas 
immediately and directly aroused by them ; we choose when 
out of a whole number of concurrent ideas we raise to a higher 
level of distinctness some particular one which long-established 
mental tendencies dispose us to regard as the fittest at the time. 
And the coincidence of the internal with external voluntary 
acts is proved by this, that not seldom our decision is piclaced 
by a clearly perceptible strife between different motives. 

Now it is plain that these internal acts of will .ire n<>t onl\ 
the analogues of the external, but at the same time their con- 
dition. There can never be an external act save a> the result 
of a previous inner selection, and this holds again both ol the 
impulse and of the act of choice. So that apperception is tin- 
one original act of will. It can exist without the 
which follow upon other acts of will, whereas these alwav> pre- 
suppose as their condition some internal act. 

256 Lectures on Human and Animal Psychology 

There is another property of apperception and attention that 
demands consideration under the head of its relation to will, 
and which plays an important part in the sequence of mental 
processes. We observe an alternation in the internal activity 
of attention, just as we find in external voluntary acts alternat- 
ing periods of rest and activity, recurring either at regular 
intervals, or, as the conditions chance to vary, after pauses of 
varying length. You know, for instance, how difficult it is to 
follow a lecture word for word with uniform attention. If it 
were really necessary for our understanding of the whole that 
we should apperceive each single word with equal clearness, 
it would be altogether impossible to follow what is said. But 
in most cases the context enables us to fill out passages to 
which we have not been especially attentive. And, to a 
certain extent, that holds also of the speaker. Language is 
fortunately of such a nature that a whole number of verbal 
ideas which are indispensable to the expression of thought 
associate by frequent repetition to a scheme which comes, so 
to speak, of itself; so that the attention may be rested while 
the speaking is following the lines of the customary associa- 
tion. We may assume that these fluctuations of attention 
are, as a rule, pretty irregular : they vary with external im- 
pressions and internal necessities. In other words, since its 
two conditions change, we shall not expect to find in attention 
as a whole any periodic function of consciousness. We are, 
however, able, by special experimental arrangements, to intro' 
ciuce regularity into the conditions, and to keep them practically 
uniform for a considerable length of time. If this is done, we 
still find that apperception is by no means constant at a certain 
intensity ; it still rises and falls, and its periods, owing to the 
uniformity of conditions, are fairly regular. 

For the purposes of such experimentation it is best to employ 
very weak sense-stimuli, such as can be easily perceived with 
some strain of the attention, but fall below the limen of conscious- 
ness with any relaxation. Under these conditions we find a 
reciprocal relation to exist between intensity and distinctness of 
ideas This is, of course, closely related to the law that intensity 

Attention and Will; Fluctuations of Attoiti-n 25 

favours distinctness in ideation. If we allow a very weak itnprc-.- 
sion which lies just above the limen of .stimulus to atiect a -ensc- 
organ, any momentary relaxation of attention will allow it t 
fall below the limen. In other words, the previously perceived 
impression becomes imperceptible. This phenomenon may ! 
regarded in two ways. It may be considered, first, as a decreav 
of sensation-intensity from the minimal perception-magnitude 
to zero ; or it may be looked upon, secondly, as the sinking of; 
previously relatively distinct idea below the iimcn of consci 
ness. There is no real contradiction between the two interpre- 
tations. The two sets of expressions can only be equivalents 
if the concepts "stimulus-limen " and " limen of con-cii.usne-* 
mean the same thing, regarded from different point-- < f view : 
an impression which passes the stimulus-limen cros-c> at the 
same time the limen of consciousness. That is, the equivalence 
of the two expressions is due to the fact that the stimulus-limen 
is a value depending upon stimulus-intensity on the one hand 
and upon the state of consciousness, i.e., of attention, on the 

Weak auditory stimuli furnish us \vith the simplest mean.- 
of observing the periodic fluctuations of attention under the 
influence of constant conditions. If you place a watch, say at 
night-time, when everything is quiet, at such a distanc 
your ear that its ticking can be just heard with strained atten- 
tion, you will find that at intervals of three to four seconds the 
regularly recurring impressions alternately appear and disappear. 
Very similar fluctuations of sensation maybe noticed if the 
is stimulated by a uniform induction current of very 
intensity ; only in this case the periods are somewhat 
Sight can be best investigated, not by means of appro.. 
liminal stimuli, but by the aid of stimulus- 
differences at the limit of notice-ability. 
The difference-limen, as we may call it, 
takes the place of the stimulus-iimen in the 
other two sense-departments; the diner-] 
ence is alternately noticed and not notice 
The phenomenon may be very conveni- 
ently studied on quickly rotating disc- 
small piece of a black sector measuring only 

258 Lectures on Human and Animal Psychology 

a few angular degrees is painted upon a white disc. In quick 
rotation we have a grey ring on a white ground. If the sector 
is made of the proper breadth, the ring will be just noticeably 
different from its background. If you fixate it continuously, 
you will find that it is alternately visible and invisible. 

It has often been conjectured that the phenomena which we 
have been describing are dependent upon purely physiological 
conditions, lying in the peripheral nerves and organs of sense ; 
e.g., upon a periodically restored exhaustion of the organs or 
upon an alternation of strained and relaxed movement. But, so 
far as the matter has admitted of experimental test, these hypo- 
theses have not been confirmed. Where peripheral changes were 
found, they proved to be either effects of the fluctuation of 
attention, or secondary conditions which, though they might 
affect the temporal course of the phenomena, were not the 
occasion of them. It has also been noticed, especially in obser- 
vations upon the concomitant feelings, that so soon as the 
impression falls below the limen there is a sudden and stronger 
strain of the attention, immediately attended by the reappear- 
ance of the sensation. All these facts make for the assumption 
that the phenomena in question belong directly to the sphere of 
the functions of the attention. But we must not suppose, of 
course, that these functions have no physical attendants, central 
or peripheral. So that the conditions which serve to vary these 
ivill also affect the time-relations of the fluctuations of attention. 

The instantaneous production of transitory sense-ideas, which 
served in the first place to facilitate the investigation of the vary- 
ing degree of ideational distinctness, has proved a method of 
widening our knowledge of a large number of important mental 
phenomena. It has, however, been found inapplicable to the 
problem of the range of consciousness for the solution of which 
it was originally employed, just by reason of this continuous 
gradation of the distinctness of ideas. At the same time, 


observation of the effect of sudden visual impressions pointed out 
the path which investigators of this problem must follow. 

Suppose that, at a given moment, a complex impression affects 

Range of Consciousness 259 

the eye in such a way that only a part of it can be clearly 
apperceived. It may be a large number of letters, or a compli- 
cated geometrical figure (Fig. 37, p 242). And suppose that the 
moment after, there is given either a similar or a somewhat 
different impression. The comparison of the two complex 
impressions is found to be based not only upon clear appercep- 
tion, but also upon those constituents of the idea which had been 
merely obscurely apperceived. It is not infrequently possible to 
say in such a case that the two impressions are " alike " or 
" unlike," without our having any account to give of the elements 
which condition the judgment of unlikeness in the second 
instance. It follows from this that the more obscure constitu- 
ents of an impression are taken up into, and are capable of 
modifying, its total idea. But if the experiment is varied in the 
way that a complex image is divided into halves and one of these 
presented at one moment, the other after a small but noticeable 
interval, it is found that the two successive impressions cannot 
be at all combined, as were the simultaneously given constituents. 
If the two half-images, a and b, are compared with the total 
image, a + i>, shown in a later experiment, one of two things 
may happen. Where the complexity of the impression is con- 
siderable, it may simply not be seen that a + b is identical with 
the sum of the successive ideas a and b. Or, if the identity is 
recognised, it is clearly perceived that reflection and thought 
have taken the place of direct observation. Suppos 
instance, that the first impression given is that of a uniform 
duodecagon. If the same object is shown again in a second 
experiment, it will be at once recognised and clearly 
guished from a decagon, although there can have been no count- 
ing of the angles, and although nothing is otherwise known as 
to" their number. Now suppose that in a second 
experiments there are presented first one half of the 
decagon, then its other half, and thirdly the complete 
No one will obtain from the perception of these 
the idea that the two first together made up the 1 
to say, our subjective perception marks off as totally di. 
processes the immediate and perceptual recognition, and 
mediate and logical recognition of a compoum 
former is an instantaneous process of perception, the latter a 

260 Lectures on Human and Animal Psy etiology 

serial process of comparative judgment. With the former, 
again, is bound up a characteristic feeling, an unfailing con- 
stituent of perceptual recognitive processes to which we shall 
recur later ; whereas the latter shows no trace of it. 

These invariable and obvious characteristics of immediate 
recognition, which are, of course, not confined exclusively to the 
ideas of sight, lurnish us with a means of answering the general 
question as to the range of consciousness. For immediate 
recognition it is necessary that the recognised idea has, at some 
time or another, been present in consciousness as a whole. The 
problem, then, will now be to determine how many separate 
ideas may be combined in ? total image without doing away 
with the possibility of an unfailing perceptual recognition when 
the same impression is repeated. The separate ideas which are 
thus connected into a complex need not necessarily come from 
impressions which are objectively simultaneous. Suppose, for 
instance, that a number of auditory impressions be given in 
fairly quick succession. The series forms a total impression, of 
whose constituents there are certainly more than one in con- 
sciousness at any moment. Thus we should obviously be quite 
unable to estimate the rapidity at which one hammer-stroke in 
a series followed another, if there were not present in conscious- 
ness at the instant when a new sound comes one or more of 
those preceding it ; if, that is, the time-intervals between every 
two strokes were not directly given in perception. You can 
see that the same conditions will hold for the perceptual com- 
parison of different series of this kind as hold for other 
complex impressions. It is only what has been present in con- 
sciousness at sometime or other as a whole that can be a unity 
for perception, and as such a unity compared with another 
similar whole. And there are special reasons for preferring 
auditory impressions to those of other sense-stimuli in our 
present investigation. In the first place, it is particularly easy to 
obtain a relative simplicity and uniformity of sound-impressions. 
Secondly, the sense of sight, the only possible rival of that of 
hearing, is liable to disturbance owing to the differences between 
direct and indirect vision. Lastly, we have had most practice 
in the apprehension of uniform series of auditory impressions ; 
so that it is easiest for us in their case to perform the act of 

Range of Consciousness 


recognition at once, and with the necessary certainty. You can 
see for yourselves how experiments can be carried out. The 
very simple technique of such experimentation is shown in Fi\'. 
39. It requires a metronome, ;, of the kind generally used for 
marking time in music. Metronome-strokes will serve as the 

FIG. 39. 

simple ideas, the maximal number of which in consciousness we 
have to determine. Affixed to the pendulum of the metronome 
is a small iron plate, projecting on either side. This is arrange 
between two electro-magnets, e v and e?, in such a way that the 
pendulum can be arrested or set swinging at any moment by 
the closing or opening of a current passing through them from tl 
battery k v . The current is made by simply closing the key s \\ 
the left hand. In order to mark off for perception the 
series of metronome-beats, we make use of a small electric 
supplied by a second current, /. This current is made tor 4 
moment, and then broken again by an instantaneous pre-nrc 
upon the button of the telegraph key /. The experiments are 
carried out as follows : a signal is given to the obscrvcj that 
all is ready ; and then the experimenter opens s, and lets the 

262 Lectures on Human and Animal Psychology 

pendulum swing. Simultaneously with the first stroke, he 
presses t, and the bell sounds. After the right number of strokes 
has been given, a second series is at once begun, its first stroke 
being again marked by a simultaneous bell-stroke. As soon as 
the second series is concluded, the experiment is brought to an 
end by closing s ; i.e., by the arrest of the pendulum by one of 
the two electro-magnets. If we denote the metronome-strokes by 
quavers, and the bell-signal by an accent placed above them, 
an experiment consisting of two successive series may be repre- 
sented in this way : 

In this instance, the two series are of equal length. In actual 
experimentation the second series is made to contain one stroke 
more or less than the first ; while the length of the series and 
the rapidity of the pendular oscillations will also be varied. 
The pendulum can, of course, be slowed or quickened within 
limits wide enough for the present purpose by moving the 
running weight up or down. The problem, then, is to ascertain 
how long a series may be, at a given rapidity of vibration, for 
the immediately following series to be cognised as equal when 
equal, and as unequal when unequal, without there being any 
counting of the strokes. And a further question is how the 
length of the just recognisable series varies with variation of 
the rapidity of vibration. 

A circumstance which facilitates these experiments, and tells 
at the same time very strongly for the correctness of the 
interpretation which we ascribe to them, is this : that the point 
at which immediate recognition ceases to be possible can be 
very accurately indicated by the observer. This point is really 
settled at the conclusion of the first series ; for the scries 
is either apprehended directly as a coherent whole, or, if the 
limit is past, appears as a discrete and indeterminate impression. 
Peculiar feelings are connected with both phenomena, feelings 
of distinctly opposite character, which make the observer fairly 
certain at the beginning of the second series whether he will 
be able to institute any comparison at all between the coming 
group and that which has just ended. The most favourable 

Range of Consciousness 263 

objective conditions for the apprehension of the largest pos- 
sible number of impressions are obtained when the interval 
between every two strokes is O'2"-O'3". The number decreases 
if the pendulum moves more quickly or more slowly ; and 
grouping ceases to be possible at a lower limit of O'l", and at 
an upper of about 4". Particularly interesting are the sub- 
jective impressions in the neighbourhood of the upper limit. 
You recall in this case the previous strokes as each new one 
comes, but your recollection is accompanied very plainly by the 
same feeling of cognition that you have when you recall pre- 
vious ideas which have disappeared from consciousness. Each 
single stroke, that is, stands to the foregoing one as (where 
grouping is possible) each whole series stands to the preceding 

Within the limits of the possibility of conscious grouping 
there comes to light a further phenomenon which is so variable 
that it exercises a decisive influence on the result. If we give 
ourselves quietly up to the apprehension of the impressions, 
we observe that the separate strokes are not all alike, even 
though they are really and objectively perfectly equal. We 
alternately accentuate and slur them, just as we do in marking 
time in speech, by a voluntary and regularly alternating in- 
tensity of accentuation. If we denote the accented impressions 
by points placed above them, we have the two series of our 
former figure as they usually occur in reality : 

! I I I I I ! I I I I I I | I 


The series of twelve strokes, that is, consists not of twelve 
equivalent ideas, but of six ideas, each of which has two parts. 
With this simplest form of accentuation we are able to group 
into a recognisable series at most sixteen single strokes ; i.e., 
eight bimembral ideas. 

But the same phenomenon may show itself in more com- 
plicated form. The series need not be divided on this simplest 
scheme in 2 : 8 time. There may be different degrees of 
accentuation regularly alternating with one another and with 
slurs ; and so there may arise a more complicated rhythm. 
There need not be any intention to form these secondary 

264 Lectures on HUJUCDL and Animal Psychology 

groups : a certain degree of complexity may result simply from 
the effort to hold as many impressions as possible together 
in consciousness. You may quite easily obtain the following 
system, e.g., in which the different degrees of accentuation are 
again denoted by points, the strongest by three, the next by 
two, and the weakest by one, 

r r 

and so on. By employing this graduated accentuation, we 
divide up the total idea of the current series into component 
ideas, each of eight single impressions. 

Detailed and varied observations of this nature have shown 
that we are able with strained attention to hold in mind, and 
to compare with an immediately following group of similai 
extent, a series consisting of five of these compound im- 
pressions ; that is, of forty separate impressions. If the ideas 
are made as little complex as possible, therefore, eight is the 
maximum of grouping; if they are made as complex as possible, 
five. But, on the other hand, the number of ideational elements 
simultaneously present in consciousness may be raised through 
progressive complexity from sixteen to forty. 

We never find more than three accents employed to divide a 
series, for the obvious reason that more than three cannot be 
distinguished with certainty. This reminds us of the fact that 
in other cases of purely quantitative discrimination in im- 
mediate sensation we cannot pass beyond the three-limit 
without imperilling the accuracy of recognition. We can easily 
interpolate a sound of mean intensity between a stronger and a 
weaker, but hardly more. Ordinary language designates grey 
as an intermediate quality between black and white ; and grey 
itself is further distinguished as dark grey, grey, and bright 
grey. This limitation of our capacity to graduate quantitatively 
may well be referred to that principle of relativity which 
underlies mental measurement in general (cf. pp. 62, 63). The 
principle tells us that any estimation of intensive magnitude 
must be made in terms of other magnitudes simultaneously 
ideated. So that we can easily apprehend a given sensation in 
its relation to a stronger and a weaker, but are hopelessly at 

Range of Consciousness 265 

sea when required to hold in mind any larger number of sensible 

If we look at the metrical forms employed in music and 
poetry, we find again that the limit of three degrees of accent 
is never exceeded. The absolute amount of accentuation may, 
of course, be very different in different cases. But in immediate 
perception these different degrees are always arranged in three 
principal classes, which alone are of any real importance in 
metrical division as a basis of classification for rhythmical 
forms. As a matter of fact, however, music and poetry never 
push their use of this aid in the formation of easily compre- 
hended ideational series to the extreme limit of conscious 
grouping. Each member in a rhythmical series must be 
referred to its predecessors ; and for this to be done with 
pleasure and without effort, it is necessary that the grasp of 
consciousness be not too heavily taxed. So that a time like 
the 6 : 4 is one of the most complex of the rhythms employed 
in music. Its scheme is the following : 

It contains, you see, only twelve simple impressions. We must, 
of course, remember that in this case there is present, beside 
the intensive, a qualitative tone-variation, capable of far wider 
variation and, therefore, setting all the narrower bounds to 
intensive change. 




WE have solved the problem of determining the ideational 
content of consciousness at any given moment. The 
next question that arises is that of ideational succession. 
This falls into two parts : we must investigate, first, the time- 
relations of ideational change, and, secondly, the qualitative 
relations obtaining between the changing ideas. The actual 
train of ideas must always be regarded under both its temporal 
and qualitative aspects. A quantitative consideration of its 
time-relations cannot, therefore, neglect the qualitative relations 
of the single ideas, just because mental time-relations in general 
must be essentially dependent upon the quality of conscious 
content. At the same time, we shall find it best to separate the 
two sides of the problem so far, at least, as to deal predomi- 
nantly with temporal properties, and to attend only to those 
more general and fundamental qualitative relations which 
exercise a decisive influence upon them. 

The coming and going, the rise and fall, of ideas have been 
often enough described, though a guarantee for the absolute 
correctness of the description cannot always be found. The 
alleged facts rest partly upon all manner of speculative assump- 
lions, partly on the uncertain ground of introspection. Intro- 
spection unsupported by experiment can just as little lead to 
any certain result here as it could in the inquiry regarding the 
range of consciousness. And, unfortunately, the phenomena of 
ideational succession with which introspection was specially 

concerned happened to be just those which are least of all 


The Personal Difference of Astronomers 267 

accessible to exact investigation, the internal train of fancy and 
memory-images which we find running its course in the absence 
of external sense-perceptions. There was total disregard of 
the ideas directly excited by sense-impressions, or directly and 
uniformly connected with sense-perceptions. In their case there 
seemed to be no need of question, inasmuch as in sense-percep- 
tion the course of objective impressions and the train of subjec- 
tive ideas were entirely congruent. 


The first indication of the wrongness of this opinion, and the 
first sign that the shortest and indeed the only road to the in- 
vestigation of the temporal course of conscious processes lay in 
observing the ideas directly aroused by external stimuli, came 
to psychology from outside, from a science in which obser- 
vational methods had come in the course of time to a high 
development, astronomy. Astronomers had noticed certain 
sources of error in the temporal determination of movements of 
the heavenly bodies which, while they tended to invalidate the 
objective value of an observation, cast at the same time a most 
instructive light upon the subjective peculiarities of the observer. 

Suppose that we have to determine the time of the passage 
of a star at some distance from the pole across the meridian. 
We may employ an old astronomical method, which is still 
sometimes used for temporal determinations, and which is called 
the 'eye and ear method.' A little before the time of the ex- 
pected passage, the astronomer sets his telescope, in the eye- 
piece of which there have been fixed a number of clearly visible 
vertical threads, in such a way that the middle thread exactly 
coincides with the meridian of the part of the sky under obser- 
vation. Before looking through the instrument, he notes the 
time by the astronomical clock at his side, and then goes on 
counting the pendulum-beats while he follows the movement of 
the star. Now the time-determination would be very simple, if 
a pendulum-beat came at the precise moment at which the star 
crosses the middle thread. But that, of course, happens only 
occasionally and by chance : as a rule, the passage occurs in the 
interval between two beats. To ascertain the exact time of the 
passage, therefore, it is necessary to determine how much time 

268 Lee lures on Human and Animal Psychology 

has elapsed between the last beat before the passage and the 
passage itself, and to add this time, some fraction of a second, 
to the time of the last beat. The observer notes, therefore, 
the position of the star at the beat directly before its passage 
across the middle thread, and also its position at the beat which 
f % comes immediately after the passage, and 
then divides the time according to the 
length of space traversed. If f (Fig. 40) 
is the middle thread of the telescope, a 
the position of the star at the first beat 
and b at the second, and if a f is, e.g., 
Fiu. 40. twice as long as f b, there must be added 

f" to the last counted second. 

When the errors dependent upon accidental circumstances 
have all been eliminated, these measurements still show differ- 
ences between different observers. They persist even when 
there is no external reason discoverable. The fact was first 
noticed in the annals of the Greenwich Observatory for 1795. 
The astronomer writes that he dismissed his assistant as un- 
reliable because he had acquired the habit of seeing all stellar 
transits half a second too late. Not till many decades later was 
the scientific honour of the assistant vindicated. It was the 
celebrated German astronomer Bessel who proved that this 
difference between two observers is only a special case of a 
phenomenon of universal occurrence. Bessel compared his own 
results with those of other astronomers, and came to the sur- 
prising conclusion that it is hardly possible to find two observers 
who put the passage of a star at precisely the same time, and 
that the personal differences may amount to a whole second. 
These observations were confirmed at all observatories, and in 
the course of the experiments many other interesting facts came 
to light. It was found, for instance, that the personal difference 
between two observers is a variable quantity, fluctuating, as a 
rule, but little in short periods of time, but showing larger vari- 
ations in the course of months and years. 

It is plain that these differences could not possibly occur if 
the idea of an impression and the impression itself came simul- 
taneously. It is true that differences might appear between the 
determinations of various observers owine: to uneliminaHe 

The Personal Difference of Astronomers 269 

errors of measurement, but these would disappear if a sufficient 
number of observations were taken. A constant and regular 
difference, such as this actually is, is only explicable on the 
assumption that the objective times of the auditory and visual 
impression and the times of their subjective perception are not 
identical, and that these times show a further difference from 
one another according to the individual observer Now attention 
will obviously exercise a decisive influence upon the direction 
and magnitude of such individual variations. Suppose that one 
observer is attending more closely to the visual impression of 
the star. A relatively longer time will a c f Id 

elapse before the apperception of the 
sound of the pendulum-beat. If, there- 
fore, the real position of the star is a 
at the first beat and b at the second 
(Fig. 41), the sound will possibly not 
be apperceived till c and d, so that FIG. 41. 

these appear to be the two positions of the star. If a c and b d 
are each of them ", the passage of the star is plainly put -V" 
later than it really should be. On the other hand, if the atten- 
tion is concentrated principally on the pendulum-beats, it will 
be fully ready and properly adjusted for these, coming as they 
do in regular succession, before they actually enter conscious- 
ness. Hence it may happen that the beat of the pendulum is 
associated with some point of time earlier than the exact 
moment of the star's passage across the meridian. In this case 
you hear too early, so to speak, just c a f d b 

as in the other case you heard too late. 
The positions c and d (Fig. 42) are now 
inversely related to a and b. If c a and 
d b ate again -", the passage is put 
^" earlier than it really occurs. If 
we imagine that one of two astronomers FIG. 42. 

observes on the scheme of Fig. 41, the other on that of Fig. 42, 
in other words, that the attention of the one is predominantly 
visual, that of the other predominantly auditory, there will be 
a constant personal difference between them of %" -- -^". You 
can also see that the smaller differences will appear where the 
manner of observing is the same in both cases, but there arc 

270 Lectures on Human and Animal Psychology 

differences in the degree of the strain of the attention ; while 
the larger differences must point to differences like those just 
described, in the direction of the attention. 

It is, unfortunately, not possible in these astronomical obser- 
vations to eliminate the errors introduced by the mental ten- 
dencies of the observer. We do not know the time of the actual 
passage of the star, and we can only infer from the personal 
differences that the observed time of passage is not the real 
time. But the exact deviation of the individual observer from 
the true time remains undetermined. Hence the explanation 
which we have offered for personal differences in general and 
the larger ones in particular is so far no more than a hypothesis. 
To prove that it is right, we should have to determine the actual 
position of the star at some point in its passage and compare 
this with the estimated position given by different observers. 
This is, of course, impossible ; the heavenly bodies are beyond 
our control. But there is nothing to prevent the repetition of 
the phenomenon by artificial means under circumstances which 
readily allow of a comparison of actual and estimated times. A 
very simple apparatus of this description is represented in Fig. 

43. It is the one by the aid 
of which I carried out my 

^ \ 

^""* <- -,.. 

! J 

s ,,-"-"": 

a 1 


\ """"*- j 










first experiments on the time- 
relations of mental processes 
in 1861. It consists of a large, 
heavy wooden pendulum. The 
bulb carries a pointer, which, 
as the pendulum swings, passes 
over a circular scale. Near 
the point of rotation m there 
is fixed to the stem a hori- 
zontal metal bar : s s. A 
movable, vertical standard, h, 
has attached to it a small metal spring, also in the horizontal 
line. The spring is fixed in such a way that the bar s s in 
passing by produces a single short click, the end of the bar and 
the point of the spring just touching each other, while the shock 
is so slight that the course of the heavy pendulum is not notice- 
ably affected. By watching the course of the pointer attached 

The Personal Difference of Astronomers 271 

to the bulb of the pendulum, while the upper part of the instru- 
ment remains concealed, we can determine at what point of its 
passage to or fro the click of the spring takes place. For 
example, if the pointer appears to be at e at the moment of 
the sound, the bar s s will be in the position a b, and this will 
mean that the passage is put too early. If the pointer seemed 
to be at e ', the bar would be in the position c d, and this would 
mean that the passage was put too late. If we know the dura- 
tion and amplitude of the pendular vibration, and measure the 
angular difference between e or e" and the actual point at 
which the bar s s comes in contact with the spring, we can 
easily calculate the interval between the giving of the sound 
and its apperception. To obviate the influence of preconceived 
judgments, the spring is given a slightly different position in 
each experiment, so that the observer never knows when the 
sound is really coming. It was found by this method of inves- 
tigation that a slow vibration-rate gave a time-displacement 
averaging ". The time of the click was put that much too 
early ; the sound-impression was associated with the position of 
the pointer, which in actual fact preceded it by ^". Later 
experiments with a more adequate technique * have shown that 
the magnitude and direction of this time-displacement are con- 
ditioned in the most various ways. Of especial importance is 
the rapidity at which the sound-impressions succeed one another. 
In a slow series we tend to antedate the passage, in a rapid 
series to postdate it. Moreover, the temporal localisation of the 
sound becomes later if other impressions, e.g., electrical cuta- 
neous stimuli, are given simultaneously with it. The nature of 
these influences confirms the explanation of the varying time- 
displacement in astronomical observations offered above. For 
all the conditions which delay our apprehension of the passage 
are conditions which either prevent or retard the completion of 
a preparatory adjustment of the attention. To the former class 
belongs a high rapidity of the succession of sound-impressions ; 
to the latter, the simultaneous excitation of other senses. 

1 1 have described and figured the apparatus, under the name of Pendel- 
apparat fur Komplikationsvcrsuchc, in my Physioi. Psych., 3rd ed., ii., p. 

272 Lectures on Human and Animal Psychology 


But however interesting- these astronomical observations and 
the psychological experiments by the eye and ear method 
modelled upon them may be for a theory of attention, they 
give us no direct information with regard to the duration of 
mental processes. It would be altogether wrong to take the 
absolute difference between the actual and estimated time of 
the passage for a time-period corresponding to the duration of 
any particular mental act. For we have seen that this difference 
simply depends on the interrelation of the objective change of 
impressions and the variation in attentional adjustment. It will 
be positive, negative, or zero, according to the experimental 
conditions ; the latter, of course, when the rate of succession is 
found at which the actual and apparent times of passage are 
approximately coincident. 

But there is another method which has brought us nearer the 
desired result. Like the first, it has come to psychology from 
astronomy. To avoid the considerable personal differences of 
the eye and ear method, and at the same time to obtain greater 
accuracy in the estimated times of stellar transits, the astrono- 
mers have recently been led to prefer a different mode of pro- 
cedure, which is called the registration-method. The passage 
of the star across the field of the telescope is observed in precisely 
the same way as before, except that there is no counting of pen- 
dulum-beats. At the moment when the star passes the meridian 
thread, a movement of the hand is made, which records the 
transit upon a chronometrical instrument. The apparatus em- 
ployed is usually as follows: an endless sheet of paper is 
transferred by clockwork from one cylinder to another, so as to 
move at a constant velocity before a twofold registration instru- 
ment. One half of this consists in a writing lever, which is 
moved by an electro-magnet every time that the pendulum of 
the clock swings through its position of rest. If the pendulum 
makes one complete to-and-fro movement in the second, the 
lever moves every half-second, making a momentary elevation 
in the line it is describing upon the moving sheet of paper (// //, 
Fig. 44). The other half of the registration instrument consists in 
a similar writing lever, which is connected with a key of the 

Metliod of Registration 273 

kind used in telegraphy. The observer keeps this key closed by 
pressure of the hand, and opens it at the moment when the star 
crosses the middle thread of the telescope. A movement of the 
lever follows, the beginning of which can be exactly determined 
from the simultaneously recorded half-seconds of the pendulum. 
'Ihus, if U U' is the half-second line upon the moving paper, and 
R R' the line recording the reaction-movement of the observer's 
hand, we can ascertain the time at which the second lever began 
to rise, r, by drawing a perpendicular, c b, to the line u //, and 
measuring the time a b which has elapsed between b and the 
beginning of the last half-second. This is done, again, by putting 
space for time: if a b -^ a a, -" must be added to the time- 
value of a. 

a I) a' 


FIG. 44. 

Astronomical observations of stellar transits by the registration- 
method showed, as had been expected, smaller personal dif- 
ferences than those of the eye and ear method. But the differences 
by no means disappeared. They may still amount to hun- 
dredths or even tenths of a second. And this is not difficult to 
understand. We cannot suppose that the reacting hand-move- 
ment takes place simultaneously with the actual passage of the 
star: a certain time will elapse between the transit and its 
perception, and again between the perception and the execution 
of the movement, which may possibly be different in different 
individuals, and so condition 'personal differences.' Indeed, the 
composition of these lesser time-values is plainly a matter of 
more complexity than that of the times found by the eye an i car 
method. In the first place, physiological processes occupying 
a certain period of time enter into the total movement-process 
under consideration. The impression of the star upon the 
meridian thread must be conducted to the brain, must arouse an 
excitation there ; and then, before the hand-movement can take 
place, the impulse of will must be conveyed to the muscles. rt nd 


274 Lectures on Human and Animal Psychology 

these stimulated to contract. To these two purely physiological 
must be added the psychological or psychophysical processes of 
apperception of the impression and impulse of will. Even 
though in actions like this, where the movement answering to 
the impression is so accurately anticipated, the two acts of appre- 
hension and corresponding movement may possibly be exactly 
coincident in time, still it seems necessary to suppose that this 
whole psychophysical process will occupy no inconsiderable, per- 
haps the major, part of the total time elapsing between sense- 
impression and reaction-movement. The supposition becomes 
probability when we take into account the magnitude of the 
personal differences which are still found to occur. Differences 
so large as these may be expected where the processes involved 
.are mental, but hardly where they are physiological and simi- 
larly conditioned. But neither does the registration-method tell 
us anything of the actual time-values of the various processes. 
We do not know the real time of the stellar transit, and so are 

FIG. 45. 

still restricted to the inference that, since the differences between 
separate observers are so considerable, the times whose dif- 
ferences they are must themselves be comparatively large. 

But here, again, it is not difficult to introduce artificial experi- 
mental conditions which allow of the measurement of the 
absolute times in question. We can use for this purpose the 
same astronomical method, with the single difference that, in 
addition to the instruments for registering time and movement, 
there is introduced a third lever to mark upon the moving paper 
the moment at which the external sense-stimulus is given. It is 
also advantageous, since the times and time-differences to be de- 
termined may possibly be very small, to substitute for the vibra- 
tions of a clock-pendulum in the record of the time-curve some 
other and more accurate chronometrical instrument. The best 
is a vibrating tuning-fork ; and the technique is so far simplified 
that a tuning-fork with a bristle attached to it can very well 

Method of Registration 275 

trace its own movements. For instance, if S S in Fig. 45 denotes 
the line written by the vibrating fork, and R R' that of the hand- 
reaction, a third line, E E', between the two, will represent the 
self-registration by the stimulus of the moment of its objective 
occurrence. From the beginning of the elevation indicating the 
giving of the stimulus and from the beginning of the reaction- 
movement answering to it, perpendiculars,^ b and c d, are drawn 
to the tuning-fork curve, and the time between <and d measured 
by reference to the known duration of its vibrations. If, e.g., the 
fork makes 100 vibrations in i", a hundred full vibrations, each 
comprising one hill and one valley of the record, -every tenth 
part of a to-and-fro movement corresponds to j^V'' a space- 
value which is not too small to admit of fairly accurate measure- 
ment. The distance b d would then correspond to some 
-f^ = 1 1 4 or OTO4". The time thus measured between im- 
pression and reaction-movement is called the reaction-time. It 
is made up, as we have seen, of purely physiological and of 
psychological processes ; and we cannot separate the two, or 
ascertain with even conjectural probability the time-value of the 
mental component. But although the mere determination of 
the reaction-time possesses scarcely any importance for psy- 
chology, it, nevertheless, is the necessary first step in all possible 
mental time-measurements. Recognising this, we are bound to 
consider it in some little detail. 

Since the reaction-time may in certain circumstances amount 
too 1 1 O'2",but the time-values of the mental processes connected 
with it and approached by means of it often be considerably 
larger, this method of counting the vibrations of the tuning-fork 
becomes too cumbrous and tedious for experimental use. 
There is generally substituted for it an apparatus of more recent 
construction, which works as follows : the vibrating body does 
not record its movement upon paper, but regulates the course of 
a very rapidly running clockwork. A vibrating spring, which 
takes the place of the less convenient tuning-fork, interlocks 
with a toothed clock-wheel in such a way that at each vibration 
the wheel can only turn by the width of one tooth. The same 
wheel is connected with a clock-face, on which the times that 
have elapsed can be directly read off. To allow of the measure- 
ment of longer periods, connections are introduced in the wheel- 

2 76 Lectures on Human and Animal Psychology 

work of the clock similar to those which in an ordinary seconds 
watch join the wheel that carries the seconds hand with that 
regulating the large hand. There is further introduced a system 
of electro-magnets, which enables us to arrest or start the move- 
ment of the hand at any moment by the make or break of a 
current. It is now easy to arrange the experiment in such a 
way that the giving of the external sense-impression sets the 
clock in movement, and the reaction of the observer stops it. 
The difference between the position of the hand before and after 

gives us the reaction-time. 


In making experiments of this kind with chronometrical in- 
struments, or drawing inferences from experiments, we must 
remember that the chronoscope, whose unit is a thousandth of 
a second, cannot be regarded as a simple watch. It would be 
quite wrong to read off the times from it without paying regard 
to the numerous sources of error which minute chronometry 
always involves. Unless the apparatus is continually and ac- 
curately tested, and the observer thoroughly practised in the 
technique of such experimentation, there can be no hope of 
obtaining reliable results. And you will find during the course 
of practice that there are individuals who are entirely incapable 
of any steady concentration of the attention, and who will, there- 
fore, never make trustworthy subjects. That should not be 
surprising. It is not every one who has the capacity for astro- 
nomical or physical observation ; and it is not to be expected 
either that every one is endowed with the gifts requisite for 
psychological experimentation. This is, unfortunately, not 
seldom forgotten. And the consequence is, that the literature 
of psychological chronometry, which has assumed such imposing 
dimensions in the course of the last few years, gives but ? scanty 
store of sifted grain to the inquirer who would turn it to psy- 
chological account. 

At the same time, the simple reactions to impressions of 
sound, light, and touch are happily possessed of certain charac- 
teristics which render it an easy matter to separate the useful 
from the useless, provided only that the experiments are de- 
scribed in sufficient detail to allow of their being definitely 

Reaction- time 277 

known. The first consists in the mean value of the reaction- 
time, the second in the relative constancy of this value. In 
opposition to all the earlier statements of large individual 
differences in reaction-time, it has been shown with ever-increas- 
ing certainty since the introduction of the more accurate obser- 
vational methods that, other conditions equal, there is great 
uniformity in reaction-time, a uniformity which is independent 
of all individual differences when once practice has been carried 
sufficiently far. Equality of conditions means, first, likeness of 
quality and intensity of the sense-impression, and, secondly, 
likeness in the condition of the sensory and motor apparatus 
concerned in the reaction-process. As regards the former point, 
it is noteworthy that the different sense-departments show cer- 
tain constant differences, and that very weak stimuli lengthen 
the times, though these are absolutely constant for impressions 
of moderate intensity ; as regards the second, that the condi- 
tion of the organs of sense and movement, however uniform 
their functioning, never fails to present one difference, which is 
determined by differences in the direction of the attention, and 
therefore so far psychophysical in nature. The attention may 
be principally directed upon the expected sense-impression. 
In this case the specific muscular apparatus of the sense-organ 
(e.g., the tensor tympani, or the muscles of accommodation) are 
strongly innervated, the muscles concerned in the reaction- 
movement only weakly. Or the attention may be mainly 
turned to the movement which is to be made. In this case the 
energy of innervation is distributed in the converse way. We 
may, therefore, designate the first kind of reaction, where the 
sense-organ is attended to, the sensorial form ; the second, where 
the attention is directed upon the organs of movement, the 
muscular. No one, then, can be regarded as capable of experi- 
menting upon the time-relations of mental processes until he is 
able to change at will from either of these forms of reaction to 
the other. The question as to which form we have in a parti- 
cular case can be easily answered by reference to the magnitude 
of the measured times and the degree of their average constancy. 
If we take as our time-unit, for simplicity's sake, not the second, 
but the thousandth part of a seccuid, and denote this unit by 
the Greek letter <r (sigma) written above the line, we may put 

278 Lectures on Human and Animal Psychology 

it that the sensorial reaction amounts to 210 290"", the muscular 
to no i So ". The smaller number in each case gives the time 
for touch and hearing, the larger for that of sight. The mean 
variation of the separate experiments in an experimental series 
of at least 25 observations amounts in the first case to 20 40"" 
(where again the larger number refers to the sense of sight), 
in the second to 10 20". Whenever, owing to insufficient 
practice or general inability to concentrate the attention, there 
is oscillation between the two kinds of reaction, or neither is 
attained in its extreme form, we find differences in the dura- 
tion of the average values and (what is a still less mistakable 
indication) in the magnitude of the mean variation. 

We may designate this kind of reaction to an expected im- 
pression of sound, touch, or light the simple reaction. In each 
of its forms it may be regarded, for the reasons given above, as 
a composite process, containing both physiological and psycho- 
logical constituents. And the association of these constituents 
is so difficult of analysis that no conclusion can be drawn re- 
garding the duration of the mental terms of the series. But in 
our consciousness these mental terms are separated off from the 
others : and we may evidently insert other mental acts in the 
same process, and so lengthen the total time of reaction by the 
precise interval which they require for their completion. Such 
reactions we may call compound. And we shall obtain the dura- 
tion of the interpolated mental process by subtracting the simple 
from the compound time. For we may surely assume that the 
purely physiological processes are the same in both cases ; and 
that the apprehension of the impression and the impulse of will, 
implied in the simple reaction, recur in similar form in the 
compound. There is always one necessary condition, it is true : 
that the sensorial form be taken as the basis of comparison, and 
not the muscular ; the muscular is so automatic in character 
that the interpolation of new mental processes is impossible. 
For instance, in a first experimental series the observer may 
react to some light-impression without regarding its qualitative 
character, the reaction-movement simply following the impres- 
sion upon the eye. In a second series qualitatively different 

Temporal Determination of iMental Processes 279 

light-impressions may be presented irregularly and at random, 
and the observer required only to react after he has distinguished 
the quality of the particular impression. By subtracting the 
previously determined simple time from this longer time we get 
a discrimination-time ; that is, the time required to complete an 
act of discrimination. 

But now we can easily go a step farther. We may require 
the observer after discrimination to choose between different 
reaction-movements, and not to react until he has chosen. Thus 
two light-impressions, say a red and a blue, may be given in 
irregular order, the rubric being that red shall be reacted to 
with the right hand and blue with the left. Here, you see, 
there are two mental acts over and beyond the simple reaction, 
first, the act of discrimination as before ; and secondly, a new 
process, that of choice between two actions. If we subtract the 
compound time of the first order, the discrimination-time, 
from the compound time of the second order, the time of dis- 
crimination with choice, we obtain a simple choice time. Besides 
this, we may obtain compound choice times ; i.e., the duration of 
the act of choice between more than two movements. Since 
we have ten fingers at our disposal for experimental purposes, 
we can prescribe choice between as many as ten movements. 
In this case, of course, the association between the ten-finger 
movements and ten qualitatively different associations must 
have been made stable by practice, if the conditions of the ex- 
periment are to resemble those of simple choice in everything 
except the number of impressions. Impressions which are to 
be discriminated only may, naturally, be varied to a very much 
greater degree. We can determine the visual discrimination- 
times, for instance, not only of colours and brightnesses, but also 
of letters, words, geometrical figures, and other more or less 
well-known objects. 

Yet another method of obtaining reaction-times of the se.cond 
order is to set out from the time of discrimination or cognition, 
and to require that the reaction- movement shall follow only 
when some idea has been associated to the idea aroused by the 
impression. By subtracting again the cognition reaction from 
the total association reaction, we obtain an association-time ; that 
is, the time required for the appearance in consciousness of an 

2 So Lectures on Human and Animal Psychology 

associated idea excited by a perception. You will see at once 
that here again the conditions can be varied at will, whether by 
limiting the association to definite groups of ideas, by requiring 
the completion of trains of logical thought suggested by the 
sense-perception, or what not. 

We cannot, of course, enter in this place into the details of 
these measurements. Here is a brief table of figures which 
give the average time-value of some of the above-mentioned 
mental processes in thousandths of a second l : 

Cognition of a colour .... 30 

a short word ... 50 

Choice between 2 movements . . 80 

10 ... 400 

Association ...... 300 800 

A simple geometrical figure (triangle, square, etc.), or any 
other equally simple visual object, seems to be cognised almost 
as quickly as a simple colour. A single letter requires about 
the same time as a short word. Both these facts show the im- 
mense influence of habitual practice. The total impression of a 
well-known object is so familiar to us, that the need of analysing 
it into its constituents in order to distinguish it from other ob- 
jects is as remote as is the possibility of such analysis in the 
case of a simple colour. In the same way, when once we have 
learned to read, we do not divide up a word into its component 
letters, but apprehend it as a single total impression. And there 
belongs here a further observation of interest. A letter printed 
in the ordinary German type requires 10 20" more for its cogni- 

1 The two first lines of this table are taken from the results of an investi- 
gation recently carried out in my Institute by E. B. Titchener (Phil. Studien, 
vol. viii., part :). They are very much smaller than the values ascribed to 
the same acts in earlier researches, and published in the jixl edition of my 
Pliysiol. Psychologie, The reason for the difference is the more careful 
observation of a uniformly sensorial form of reaction, both in the discrimina- 
tion-experiments and the simple times with which they were compared. 
The earlier numbers were gained in experiments made before the discovery 
and consequent utilising of the difference between the two simple forms, and 
by observers who inclined to react muscularly, a circumstance which in- 
creased their cognition-times by something like the difference between the 
sensorial and muscular forms ; i.e., by about 80 100". 

Temporal Determination of Mental Processes 201 

tion than a letter of the same size in the Roman character. But 
there is no .such difference between words printed in the two 
types : the German word can be read in just about the same 
time as the Roman. The single German letter is harder to 
cognise, because of all its fine strokes and flourishes. You can 
see this very easily if you take the capitals ; now and again 
there occurs a word printed throughout in large letters. It is 
true that cognition is also retarded in this case by the un- 
accustomed nature of the whole impression. And the same 
factor is operative to make us slower in reading substantives 
printed or written in Grimm's way with small first letters, a fact 
which tells against the advisability of riding this Germanistic 

We can easily understand why the times of the cognition of 
complex objects, of association and of the formation of judg- 
ments, should be, not only longer, but at the same time more 
variable, than the others. The more complicated the processes 
become, the more dependent is each particular result upon the 
individual conditions of observation, and especially upon the 
disposition of the observer, itself determined by numberless 
past experiences and incalculable chances. A further general 
conclusion to be drawn from the numbers in the table is, that 
the duration of mental processes is by no means so brief as has 
often been assumed. The phrase ' quick as thought' does not 
refer so much to the actual rapidity with which idea succeeds 
idea in consciousness, as to our undoubted ability to drop out 
the intermediate terms in a train of thought, and so pass at one 
bound from the first to the last link in the ideational chain. 
Apart from this, it is obvious that the absolute time-values of 
the various mental processes are of no importance whatsoever 
in themselves ; they only become important when they help to 
throw light upon the nature and interconnection of our ' states 
of mind.' And for this reason the quantitative examination of 
the temporal course of ideas must always go hand in hand with 
the qualitative investigation of their mutual relations. If it pays 
heed to these facts, the psychological chronometry of the 
future may be looked to for the solution of many an important 



O J[ 

IF we look for a moment at the coming and going of ideas in 
our minds, we cannot fail to see that the plot of the play 
is determined by two influences, accidental external sense- 
impressions and previous experiences. Which of the two pre- 
ponderates at a particular time depends upon circumstances. 
Cast your eye over a landscape, or follow attentively the render- 
ing of a musical composition, and you will find yourselves 
seemingly wholly given up to the external impressions ; subjec- 
tive tendencies coming in only secondarily, and rather as feelings 
than as ideas. Now try to recall the events of the past few days. 
External sense-impressions are hardly noticed ; and the train 
of ideas, so far as it is clear and distinct, consists solely of 
reproductions of previous mental experiences. These are both 
extreme cases ; ordinarily we find ourselves in some intermediate 
frame of mind. Memory-ideas are aroused by sense-perceptions, 
and again interrupted by new impressions. Wherever the in- 
fluence of past experience is traceable, we find the memory- 
ideas aroused evidencing a definite relation to the condition 
of consciousness at the time. Sense-perception varies with 
every variation of the environment ; but the memory-image is 
always suggested, whether by a sense-perception or a previous- 
memory-image. You will object that now and again a recollection 
crops up suddenly and for no apparent reason. But attentive 
introspection will in most of these cases enable you to discover 
the thread of connection with your present state of mind. 
However little obvious this connection, then, we may safely 

Association of Ideas 283. 

assume that it is there. If it escapes our observation altogether, 
as it may do, that is only because the conditions are not favour- 
able for its apprehension. 

The interconnections of memory-ideas and sense-perceptions, 
or of memory-ideas with other memory-ideas, are called associa- 
tions of ideas. The term belongs to the English 'association' 
psychology. It was first employed to cover the phenomena of 
memory only, but afterwards extended to all possible connec- 
tions of ideas originating in the preconditions of consciousness. 
As customarily used, it is at once too narrow and too wide : too 
narrow because it leaves out of account a whole number of 
connections for the sole reason that in them the ideas do not 
come to consciousness in succession as in ordinary recollection, 
but, owing to special conditions, appear simultaneously as a 
complex totality ; too wide because it embraces all successive 
ideational connections, the act of recollection aroused by a 
simple sense-impression and the most involved process of logical 
thought. Now, true as it may be that in both these cases the 
ideational connection is determined by mutual relations implied 
in past conscious experience, it is equally true that they are in 
all other respects so different that to treat of them without 
further discrimination cannot but obscure the analysis of their 
constituent processes and hinder the understanding of their 
inter-relation. We shall ourselves mean by associations simply 
those ideational connections which do not exhibit the charac- 
teristics of the activity of logical thought. What these are we 
shall discuss later. 


The starting-point of the doctrine of association, in the usual 
sense of the term, was observation of the reproduction of earlier 
ideas. It has hardly done more than put into modern form 
what had been taught as the psychology of memory from 
Aristotle down. But conscious recollection implies a distinc- 
tion between the inducing and the induced ideas ; if the two are 
not discriminated, the process cannot be that of conscious recol- 
lection. Now it is evident that this recognition of an idea as 
having been previously experienced is a character which may 

284 Lectures on Human and Animal Psychology 

possibly attach to a revived idea, but need not by any means 
necessarily do so. The simpler case of association we must 
rather admit to be that in which ideas are connected simply by 
reason of their mutual relations in consciousness, and without 
there being any direct apprehension of the connection as an act 
of recollection. Certainly memory presupposes association, on 
the assumption, that is, that no idea comes into our minds with- 
out cause, but not every association involves an act of memory. 
That is, we must obviously set out first of all from the phe- 
nomena of association proper, and then go on to determine 
what new conditions are necessary for the association to become 

In this wider sense association embraces a whole number of 
connective processes in which the associated ideas do not suc- 
ceed one another, but come to consciousness as a simultaneous 
ideational complex. There can be no question here of an act 
of recollection, for the simple reason that the induced idea 
associated with the inducing is not in any sense separate from 
it, in other words, cannot be independently compared either 
\vith it, or with any other idea. We may call connections of 
this kind, in which the primary inducing and the associatively 
induced idea form a simultaneous conscious complex, ' associa- 
tions in simultaneous form,' or, for brevity's sake, simultaneous 
associations. There belong here, in the first place, all those 
products of the fusion of simple sensations which compose 
our complex sense-perceptions. These latter always consist of 
a connection of several sensations forming a simultaneous com- 
plex idea, such as a compound clang or some spatial idea of 
sight or touch. One difference there is between these connec- 
tions and other associations, that the sense-impressions which 
arouse the constituent sensations are themselves, interconnected; 
so that the re-excitation of earlier ideas, though of course not 
entirely inhibited, is completely overshadowed by the con- 
nective tendencies obtaining among the sensations. The sensa- 
tions composing a clang or a visual perception depend upon a 
simultaneous activity of sensory stimuli. At the same time 
this difference does not imply any essential difference in the 
psychological character of the general process ; that is, if we 
regard as the chief characteristic of association this property 

Association of Ideas 285 

of mind to connect certain ideas or ideational elements auto- 
matically into a complex idea. And there is every reason for 
doing this, since, surely, certain types of clang, certain spatial 
arrangements of sensations, are every whit as familiar as, nay 
even more familiar than is, for instance, the connection of a 
perception with a similar memory-image. But this means that 
in these simultaneous associations of perception also there is 
nothing to prevent a .sensation-element which is not actually 
given in the sense-impression being supplied by immediate 
reproduction. We have seen, for instance, that eye-movcmcr.r 
influences the perception of visual space even when the organ : 
at rest ; thus we may be deceived as to the position or move- 
ment of external objects in consequence of having intended, but 
not actually executed, an ocular movement, and this just 
because of the intimate association of movement-sensation with 
impulse of will. 1 

These fusions of uniformly connected sensations, which con- 
stitute sense-perception in general, are not very obviously 
related to the ' associations ' of current psychology. More 
akin to these are the interconnections of the perceptions of 
different senses. We see a musical instrument, and hear a 
clang from it. Our eye apprehends the white, crystalline nature 
of a lump of sugar at the same time that our tongue is 
experiencing a taste-sensation of the quality sweet. There 
arises in this way a connection between sensations and percep- 
tions of different senses so intimate, that if but one sense 
impression chances to be actually presented, or the memory- 
image of an impression aroused in the domain of one sense 
only, the other sensation is at once mentally associated with it. 
We hear the piano, and no sooner do we hear it than a va^iK- 
visual image of the instrument crops up in our minds. Or wi- 
taste sugar in the dark, and there is at once associated with 
its taste a general notion of what it looks like. These connec- 
tions of ideas of disparate senses which are referred to the 
same objects, and so belong closely together, we may term with 
Herbart complications. There can be no doubt that they arc- 
simultaneous associations. One sense-impression is so inti- 

1 For similar instances cf. Lecture V. (Associations of Tone-sensations), 
and Lectures IX. and X. (Associations of Spatial Perception). 

.286 Lectures on Hitman and Animal Psychology 

mately associated with another, or, at least, the origin of the 
t\vo is so little distinguishable in time, that the disparate 
constituents show themselves in consciousness only as the inter- 
related parts of a single idea. 

Most frequent and most important among the complications 
are verbal ideas. They stand, as a rule, in a twofold connec- 
tion : the acoustical impression is associated, first, with a 
sensation of movement, and then, secondly, at least, in many 
cases, with the visual impression of the printed or written 
characters. Movement-sensations are evidently complicated 
with other ideas as well. They acquire an especial significance 
from the fact that the memory-image of a movement is apt 
at once to arouse the movement itself. The consequence of 
this is, that movement- sensations often act vicariously for the 
sensations of certain senses whose memory-images are so faint, 
that we either do not perceive them at all or only imperfectly 
and by the aid of the muscular sensations customarily asso- 
ciated with them. It is, for instance, in most cases illusion 
when you think that you can recall the scent of a rose. 
Observe the working of your minds in the act of recollection 
more carefully, and you will find something like this. First, 
you 'nave a more or less distinct visual picture of the flower, 
and then, secondly, a movement-sensation in the nose corre- 
sponding to the inhalation of air, and then again, thirdly, 
a sensation of touch and temperature, arising from the air 
actually inhaled, your movement-sensation having been at- 
tended by an actual movement. The sensation of smell proper 
is either entirely absent, or, at least, so faint that it is altogether 
overshadowed by the other components of the complication. 
In the same way the complications consisting in memory- 
images of impressions of taste contain hardly anything of the 
sensation of taste, which is, however, quite adequately repre- 
sented by the movement-sensation, which varies for different 
gustatory substances with variation of the accompanying 
mimetic expression. 


Apart from these cases of the confusion of definite sensation- 
qualities with muscular and tactual sensations, the separate 

Assimilation 287 

constituents of a complication are in general clearly dis- 
tinguishable, belonging as they do to disparate modalities and 
occurring in other connections under other conditions. The 
same cannot be said of a second important class of simultaneous 
associations, the connections of an externally excited sense- 
perception with its related memory-images. This type of 
association we will call assimilation, and speak of the memory- 
image as the assimilating element, the sensations following 
from the sense-impression as the assimilated. These expres- 
sions imply that the memorial constituents are the determining 
factors in the result, while the incoming sense-impressions are 
determined by them. This is so far true. An impression may 
be apprehended in the most different ways, according to the 
disposition in which the mind has been left by previous expe- 
riences. The resultant complex idea is, therefore, a mixed 
product of the impressions given in perception and of an 
unknown number of memory-images. But, just because the 
idea is a single complex, there can be no question of analysis 
into these two constituents. Hence the reproductive elements 
are invariably referred to the sense-perception, which now 
contains constituents not to be found in the impression which 
aroused it. On the other hand, real constituents of the sense- 
impression may be wanting in the resultant idea, owing to their 
conflict with reproductive elements of greater intensity. 

The process of assimilation then, unlike that of complica- 
tion, is not one to be discovered by casual introspection. To 
examine it, we must carefully compare the impression with the 
idea aroused by it. The comparison shows the incongruence 
of the two, and so leads us to look for the ground of their 
difference where alone it can be found, in the activity of 
previously experienced ideas. When once our attention has 
been called to their influence, we have the key to a whole 
number of phenomena of ordinary life and of experimental 
practice, which, though striking enough, are generally left 
unnoticed or unexplained. We ordinarily read over a printer's 
error without seeing it ; that is, we read the familiar word- 
picture into the impressions presented to us. Or we fill out 
a sentence in a lecture which we have heard indistinctly 
without remarking that we have not heard it clearly. On the 

288 Lectures on Human and Animal Psychology 

other hand, we are equally liable to hear wrongly by supple- 
menting the indistinct sounds by a wrong set of memory-ideas. 
The crude outlines which serve to represent a landscape in 
stage scenery look by artificial light and at the proper distance 
to be a perfect reproduction of a real scene. Here the ready 
assistance rendered by the appropriate memorial elements is 
made still more effective by the confused outlines of the retinal 
image. Outline drawings of tridimensional objects, if purely 
schematic and unshaded, can be seen at will as tridimensional 
or areal, and if the former, as extending in this direction or 
in that ; it simply depends upon which of our familiar space- 
ideas we employ. For instance, the outline drawing of the 
head on a coin can be seen pretty much at pleasure as cameo 
or intaglio. You are all familiar with the puzzle pictures 
which have, e.g., the head of some well-known man outlined 
in the foliage of a tree. At first it is quite difficult to find 
the head. But when once you have it, it stays ; and you are 
hardly able to get rid of it again, however much you try. The 
same tiling may often be noticed in stereoscopic observations. 
For a time it will be impossible to get the idea of depth, and 
then on a sudden it comes with even plastic clearness. What 
all these cases mean is, that the assimilating memorial elements 
have required some little time to be called into activity by 
the appropriate constituents of the external sense-impression. 

But, of course, it is not usually a single memory-image which 
unites with the given impression in the process of assimilation. 
A stereoscopic object which we may never have seen in the 
exact form in which it is presented will arouse the co-operation 
of a whole number of memorial elements, taken from a whole 
number of originally separate perceptions ; and may be able in 
this way to call up the idea of three dimensions. But, for this 
very reason, it would be wrong to suppose that sense-impressions 
are first of all present in assimilation as independently co- 
existent ideas, and then fuse to an ideational unity. The assumed 
stage of independent co-existence of the components is neither 
discoverable by introspection nor actually possible, since, as a 
general rule, the assimilating effect proceeds from a large number 
of ideational elements, originally distributed through quite dif- 
ferent ideational series. We can only imagine that every sense- 

Assimilation 269 

impression acts as an excitant to numerous tendencies remaining 
from previous impressions ; and that such of these as are appro- 
priate to the impression, and, at the same time, more easily 
excited than the others, help to form the resultant idea. Lastly, 
in all these processes of assimilation, which follow directly upon 
sense-impressions, the peripherally excited sensations are so far 
of influence upon the memorial elements that they increase the 
intensity of the reproduced sensations. That is the only possible 
explanation of the fact that even in the normal assimilation it 
is impossible to distinguish between the ideational elements 
aroused by external stimulus and those excited by association. 
The impossibility becomes still clearer when the elements of 
the latter kind obtain so exclusive a predominance that the 
resultant idea is wholly inadequate to the sense-perception. 
Assimilations of this class we term illusions. In the illusion, we 
imagine that we perceive something which is not there ; that is 
to say, we confuse memorial elements with sense-impressions. 
And that again is only possible when there is no noticeable dif- 
ference in the intensity of the two constituents. 

The occurrence of a process of assimilation can be proved with 
absolute certainty, as these instances show, when the resultant 
assimilation-product is a sense-perception, whether actual or 
more or less illusory. In either case the new idea is so different 
from the sense-impression that the activity of assimilation is a 
matter of direct inference. But you will see that it is at least 
extremely probable that assimilations take place in terms of 
memory-image pure and simple ; and we have an obvious in- 
dication of this in the fact that a particular perception is not as 
a rule assimilated by a particular memorial idea, but by an 
indefinite number of such ideas. Suppose, then, that there is no 
sense- perception present, but that some memory-image crops up 
of itself. It will undergo continual variation by assimilating 
other ideas which refer to similar objects. So that we cannot 
draw any hard and fast line between a memory-image proper 
and what is called a fancy-image. Psychologists are accustomed 
to define memory-images as ideas which exactly reproduce some 
previous perception, and fancy-images as ideas consisting of a 
combination of elements taken from a whole number of per- 
ceptions. Now memory-images, in the sense of this definition, 

290 Lectures on Human and Animal Psychology 

simply do not exist. The ordinary memorial representation is 
determined by several perceptions of the same object. Thus, if 
we try to recall a person whom we have often seen, \ve never 
represent him exactly as he was on any particular occasion ; our 
idea of him is the resultant of many perceptions, whose con- 
stituents, mutually supplementing or inhibiting one another, 
combine partly to deepen, partly to soften, the general outline. 
This explains the indefiniteness of most memory-images. Even 
when we are recalling an object only once seen, the idea does 
not coincide with the original perception ; some elements are 
wanting, others which do not belong to the object are wrongly 
transferred to it from similar ones. Try, for instance, to draw 
from memory some landscape picture which you have only once 
seen ; and then compare your copy with the original. You will 
expect to find plenty of mistakes and omissions ; but you will 
also invariably find that you have put a great deal in which \vas 
not in the original, but which comes from landscape pictures which 
you have seen somewhere else. So that, according to the ordinary 
definition, every memory-image would be a fancy image, and 
ideational reproduction a concept with no corresponding reality. 
For there is no memory- image that reproduces either the 
primary perception-image, or any other memory-image of that 
same perception. And that is not hard to understand when we 
remember that our ideas are not permanent objects, but pro- 
cesses which can never exactly recur, because the conditions of 
their occurrence are never twice alike. 


The process of assimilation is, therefore, always a compound 
process, set up in any particular case by an incalculable number 
of elementary connective processes. We may now ask as to the 
character of these last, the indecomposable and fundamental 
processes of connection. In answering that question, we must 
again set out from those cases of assimilation which begin with 
a sense-impression, since they furnish the best material for the 
determination of the conditions of the phenomena. There can 
be no doubt that there are always two connective processes 
running side by side in every case of assimilation, whether this 
be normal and initiated by a sense-perception, or illusory and 

Theory of Simultaneous Association 291 

implying a misinterpretation of sensory impressions. First, the 
sense-impression calls up previous similar sensations, and, 
secondly, by the mediation of these sensations arouses other 
ideational elements not contained in the given impression, but 
which were connected with it on other occasions. What first 
happens when you look at a stereoscopic object is, that certain 
of the outlines correspond to those of some material object known 
from previous perceptions. But these coincident elements would 
in themselves be wholly inadequate to suggest the actual picture 
of a tridimensional object. There must be further aroused 
elements not present in the actual figure, but associated in 
previous ideas with the coincident parts, and now necessary to 
complete the image as that of some definite thing. When we 
read a wrongly printed word correctly, the primary suggestion 
proceeds from the rightly printed letters in it : they arouse the 
corresponding memory-images of the same letters, and these, 
again, recall to mind the letters which were visually connected 
with them in earlier perceptions, and which, taken together with 
them, give the correct picture of the word. So the disturbing 
elements in the impression are overlaid by the reproduced ideas. 
The older doctrine of association, i.e., of successive associa- 
tion, distinguished associations as those of similarity and conti- 
guity. In the first form an idea is excited which in certain of its 
characteristics resembles the exciting idea ; in the second, an 
idea which at some time or other was in temporal or spatial 
connection with it. If we apply these terms to simultaneous 
associations, we may obviously call what was above denominated 
the second act in assimilation a contiguity-association. On the 
other hand, we cannot reduce the first act in the same way to a 
similarity-association. Two objects are similar when certain of 
their characteristics correspond, while others are different. Now 
it does not seem possible that an impression should directly call 
up the memory of another, if this differs from it more or less. 
It can, surely, only arouse a memory-image like itself. There 
may, of course, follow upon the excitation of these like elements 
the reproduction of others which are unlike, provided that these 
have been connected with the like in previous ideas. In other 
words, a similarity-association always points to the union of a 
likeness-association with a contiguity-association. The wrongly 

292 Lectures on Human and Animal Psychology 

printed word calls up the right image through the likeness-asso- 
ciation of the coincident letters, and a contiguity-association, 
which takes the right elements, not given in the actual impres- 
sion, from previously seen word-pictures. The result of this 
compound process is a so-called similarity-association, for the 
wrongly printed and rightly ideated words are similar, but not 
like. It is just the same with the stereoscopic idea. The out- 
lines which originate the suggestion arouse a likeness-association, 
which is at once supplemented by one of contiguity, which partly 
fills out the resultant idea, and partly corrects it by suppressing 
any disturbing elements in the impression. Since no two ideas 
are absolutely like, it might be objected that our likeness-asso- 
ciation is itself nothing better than a similarity-association. But, 
as a matter of fact, we are not concerned here with associations 
between complete ideas, but with connections between constitu- 
ents of ideas. Absolute likeness between two ideas is impos- 
sible, for the very reason that every likeness-association gets at 
once attached to it a number of contiguity-associations, .the 
final result being either a 'similarity' or 'contiguity' association, 
according as one or other of the elementary processes predomi- 
nates. We cannot discover any other elementary processes than 
these two, the connection of the like and that of the contigu- 
ously associated in time and space. Each of them must oi 
necessity be present in every concrete association. An idea can 
only call up an earlier idea, if it has some elements in common 
with that ; ar.d since the reproduced idea contains unlike as well 
as like constituents, the likeness-connection implies the formation 
of a contiguity-connection. Likeness of the elements, you see. 
is directly effective. If a new impression contains elements like 
those of an earlier impression, these will separate from the rest, 
having become so much more familiar by repetition, and will 
preponderate in consciousness. Contiguity is only mediately 
effective. It works by way of reviving other elements ex- 
ternally connected with these like constituents of previously 
excited ideas. 

In view of this essential difference between the two processes, 
is it right to speak of a likeness-connection in the same sense as 
a contiguity-connection ? When a given impression arouses an 
idea consisting in part of elements belonging to the impression, 

Theory of Simultaneous Association 293 

in part of elements not actually present in it, but connected with 
it in previous perceptions, there can be no doubt that we are 
justified in speaking of a connective process in referring to these 
unlike constituents ; the corresponding excitations must be set 
up by an impulse proceeding from the impression. But the 
elements which pass directly from the impression into the assimi- 
lation-product hardly seem to need anything which might be 
reckoned among the association-processes: they are directly given 
by the external sensory stimuli, and would appear rather to be 
the condition of the origin of a connective process than its result. 
In other words, is not the assimilation-process entirely and 
exclusively referable to association by contiguity ? 

However tempted we may be at first sight to answer this 
question in the affirmative, a little more consideration will con- 
vince us that such an answer would be incorrect. As a matter 
of fact, the elements which pass from impression into assimila- 
tion-product are not in their second connection what they were 
in the first ; so that the word 'pass' is only in place if we are 
comparing the result with its constituents, and not as referred 
to the actual process. The ' passage ' involves the action of 
intermediary processes in two different directions. First, the 
passage of elements of an impression into the resultant idea is 
favoured, as we know from experience, by the frequency with 
\vhich they have presented themselves in previous impressions. 
The only explanation of that is, that the corresponding excita- 
tions are intensified by the dispositions left by the action of 
previous impressions. This intensification will, of course, be 
directly connected with the present impression ; it will not do 
to assume, as those psychologists do who make ideas into per- 
manent entities, that constituents of the new idea connect with 
constituents of some previous one. It will simply happen that 
the greater frequency of certain kinds of excitation implies the 
ascription of a higher intensity-value to anyone of that kind 
which affects us. And it follows from this that the final result 
is due not simply to the impression, but, and this holds of 
all association-processes, to the connection of the impression 
with the after-effects of previous excitations. Secondly, this 
passage of elements of an impression into the assimilation-pro- 
duct implies the presence of another process, the direction of 

294 Lectures on H^lman and Animal Psychology 

which is just the opposite of that of the former, a process of 
disappearance of elements which are contained in the impres- 
sion, but supplanted in the idea by other new and incompatible 
elements called up by contiguity. That is to say, the like ele- 
ments are not by any means after the ' passage ' what they were 
before it. They are partly intensified by previous practice, 
partly weakened or, at least, severed from their original connec- 
tions by inhibitory influences. All of which goes to show that 
the likeness-connection is as truly the result of many and differ- 
ent processes as is the connection by contiguity. At the same 
time, we must not forget that these determining processes are 
essentially different in the two cases. The best terms to indicate 
this difference in brief are perhaps those employed above : the 
likeness- connection is immediate, the contiguity - connection 




IN close connection with assimilation stands, as we saw above, 
the successive association of ideas. This is the process to 
which the general name of 'association ' was originally confined ; 
and it is still customary, even at the present day, to speak of 
laivs of association, and to distinguish in that way connection by 
similarity from that by co-existence in space or succession in 
time, or sometimes from that by contrast. It is scarcely neces- 
sary to say that these are really simply forms, and not laws, of 
association ; they are not universally valid conditions of its 
origin. They merely serve to furnish classificatory concepts 
under which the ready-made products of association can be sub- 
sumed. But, curiously enough, the authority of Aristotle and 
the constant inclination of the human mind towards logical 
schematisation have worked no less harm in this department of 
psychology than they worked in the sciences of nature. Aris- 
totle had distinguished four kinds of memory, in terms of the 
logical opposites 'similarity ' and 'contrast,' ' simultaneity ' and 
' succession ' : just as he had arranged the fundamental quali- 
ties of all natural bodies under the rubrics of the contraries 
' hot ' and ' cold,' ' moist ' and ' dry.' And these four forms 
have held the field, despite the evidence of observation, down to 
our own day. It is now pretty generally agreed that 'contrast ' 
may be omitted, or, where anything corresponding to it occurs, 
referred to 'similarity'; while spatial co-existence and tem- 
poral succession are brought under the general head of external 

296 Lectures on Hitman and Animal Psychology 

contiguity. This means a reduction of the four forms to two, 
association by similarity and association by contiguity. And 
the reduction is so far good that the different cases of successive 
connection may, as a rule, be arranged in one or other of the two 
classes. At the same time, the terms still tend to suggest the 
wrong idea: that they are the distinguishing.marks of elementary 
processes, instead of the classihcatory headings of association- 
products, each of which is constituted by a whole number of 
simple processes. In the matter of constituents there is, of 
course, no essential distinction to be drawn between the two 
forms. For it is obvious that just the same processes must be 
operative in successive association as in assimilation, the only 
difference between them being that the successively associated 
ideas are not combined into one simultaneous idea, but remain 
temporally separate, in obedience to conditions which we have 
still to discuss. 

Apart from this, however, we shall expect to find every suc- 
cessive association composed of two processes : a direct connec- 
tion of like elements of different ideas, and a connection, 
attaching itself immediately to this, of such elements of previous 
ideas as have been externally contiguous to those like constitu- 
ents. If, as we look at the total result, the connections of the 
like elements are predominant, we speak of a similarity-associa- 
tion ; if the external connections are the stronger, of a contigu- 
ity-association. Thus it is an association by similarity when the 
picture of a landscape reminds us of the reality as we have actu- 
ally looked upon it. Different as the picture and the retinal 
image may actually be, there are certain outlines that corre- 
spond. These call up the memory-images of earlier perceptions, 
and cause us to transfer to the picture many of the elements of 
the original which it does not really possess. Now this revival 
of elements which are not given in the picture is plainly an 
association by contiguity. Certain of these contiguity-elements 
work by way of assimilation ; they make the similarity of the 
picture to the original appear greater than it is. Others make 
against assimilation ; it is through these that we are able to dis- 
tinguish picture and reality at all, that the result of the whole 
process is not a simultaneous assimilation, but a successive simi- 
larity-association. On the other hand, if we have read the 

Association by Similarity and Contiguity 297 

letters abed, we are inclined to continue e f g h. This is 
a case of association by contiguity. But here, too, the original 
process is the direct connection of like with like. The letters 
when read call up the like letters previously read or heard. 
And it is at this stage that contiguity-effects must intervene, if 
the visual idea is to be apprehended as agreeing with previous 
perceptions. Then, by a further operation of contiguity, the 
absent letters are supplied to complete the usual series. 

Association by similarity and association by contiguity, then, 
differ in two points as regards the nature of their constituent 
processes. First, there is a general predominance of the ele- 
mentary connection by likeness in the former, of that by con- 
tiguity in the latter, form ; secondly, in the similarity-associa- 
tion our attention is directed upon the common properties of the 
ideas, in the contiguity-association upon their divergences. The 
association of the picture with the landscape is an association by 
similarity, because the resemblance of the two ideas makes us 
neglect not only their differences, but also contiguity-connec- 
tions which are essential for the institution of a comparison. 
The association of the letters of the alphabet is an association 
by contiguity, because we attend only to the added letters, and 
not to the cognition of the first letters and the likeness-connec- 
tions which it almost invariably involves. 


The result of all this is, that there are two fundamental forms 
of connection between ideational elements : connection by like- 
ness and connection by contiguity ; and that both of them are 
concerned in every case of actual association. For the proof of 
this fact our best recourse will be to the simplest cases of asso- 
ciation. They possess the further advantage of exhibiting with 
especial clearness the conditions which differentiate successive 
from simultaneous association, and particularly from assimila- 
tion. The constituent elementary processes are, of course, the 
same in both forms. 

The simplest case of assimilation is the cognition of an object ; 
the simplest case of successive association, its recognition. We 
cognise a picture as a picture even when we are perfectly sure 
that we have never seen it before ; we recognise it in remember- 

298 Lectures on Human and Animal Psychology 

ing that we have seen it, just this particular picture, on some 
previous occasion. The simple act of cognition is a process of 
assimilation. The present impression calls up earlier ideas : 
there are set up connections by likeness and contiguity, but there 
is no analysis into a succession of ideas ; the presented and the 
memorial elements combine at once to a single idea, referred to 
the actual impression. The fact that the resultant idea is, how- 
ever, no new one, but one which, upon the whole, is familiar, 
expresses itself in the character of the accompanying feeling. 
We will call this the feeling of cognition. Since feelings always 
have some ideational basis, we may suppose that the indefinite 
memory-images in the background of consciousness, whose func- 
tion is the assimilating of the given impression, serve as the 
intellectual substrate of this particular feeling. 

From this process of cognition is developed that of recogni- 
tion. The steps are three in number. 

Most closely related to the act of cognition is the process of 
immediate recognition. In this we are either unconscious or but 
obscurely conscious of the connecting links by whose aid recog- 
nition is effected. And here again two alternatives are possible : 
first, the idea is merely accompanied by the consciousness that 
it has been before our minds before, at some time or other, once 
or oftener, that is, the recognition takes place without there 
being any recollection of attendant circumstances. Secondly, 
though the recognition is immediate, it involves the recollection 
of attendant circumstances. We recall the temporal relations 
and spatial surroundings in which we previously made the 
acquaintance of the recognised object. In both cases the act of 
recognition is accompanied by a feeling. Where the first form 
of the process occurs, this is generally indefinite, and may be 
connected with the emotion of doubt. But it becomes distinct 
and vivid as soon as there is conscious localisation of the recog- 
nised idea in time and space. We may call this feeling the feel- 
ing of recognition. Now the recollection of attendant circum- 
stances consists merely in the excitation of secondary ideas 
externally contiguous with the recognised object in previous 
experiences. In other words, the act of recognition requires 
these contiguity-connections for its completion. 

The second form of immediate recognition furnishes the tran- 

Association by Similarity and Contiguity 299. 

sition to a third form of the general process, to mediate recog- 
nition. In this we are clearly conscious from the outset that 
recognition is brought about by the mediation of secondary 
ideas. Think how often you meet a person whom at first sight 
you take to be an absolute stranger. But he tells you his name, 
and on a sudden the face that was so unfamiliar shows you the 
features of an old acquaintance. Or there may be other mediat- 
ing circumstances. You see a third person whom you have 
often noticed in his company, and your eyes chance to fall on a 
coat or a travelling-bag that awaken your memory. Here again 
there is a special feeling regularly associated with the act of 
recognition. This feeling comes later and arises more gradually 
than the immediate recognition-feeling. At the same time, you 
will find that it may be very vivid, even when the apprehension 
of the agreement between the present idea and a previous one is 
still quite indefinite. 

There can be no doubt that instances of mediate recognition, 
(in this sense) occur which are wrongly taken for cases of imme- 
diate recognition, the reason being that we are not clearly 
enough conscious of the auxiliary ideas which mediate the 
recognition. Thus it has been shown by experiment that it is 
quite easy to hold in mind three shades of grey between the 
extremes of black and white, and to recognise each of them cor- 
rectly, and apparently immediately, after the lapse of some time ; 
while, if one more shade is interposed, their recognition is un- 
certain, and mistakes are many. Now there are in common use 
in language just three designations of shades of grey, dark grey,, 
grey, bright grey. So that it is not hard to see a reason for 
the definite limitation of recognition. We have only to assume 
that one of these three verbal ideas was involuntarily associated 
with each of the three impressions, and that it mediated the 
recognition. A musical ear can recognise a particular musical 
clang after a long time, if this possesses a definite tonal quality, 
and has its definite place in the musical scale marked by some 
note-name. But recognition becomes impossible very soon after 
the giving of the impression, if some other tone is taken which 
cannot be definitely associated with a name like c, c, d t etc. 

We have seen that these different forms of recognition pass 
over into one another by degrees. It might appear doubtful 

3OO Lectures on Human and Animal Psychology 

whether they should be regarded as different processes at all, 
and not rather as modifications of one and the same process, 
differing only as regards their secondary conditions, i.e., in the 
clearness or temporal course of the various conscious elements. 
Thus mediate recognition, and immediate recognition with attend- 
ant circumstances obviously differ only in this, that while in 
the former the secondary ideas are first apperceived, and there 
then arises the consciousness of the agreement of the principal 
idea with that experienced before, in the latter these secondary 
ideas are only clearly apprehended at the same time with the 
agreement of the two principals, and maybe even later. Now, 
the apperception of an idea is not the same thing as its appear- 
ance in consciousness. Our discussion of the phenomena of time- 
displacement (pp. 270, 271) showed us that when two ideas, a and 
b, follow each other in rapid succession, the second, b, may be 
apperceived before the first, a, which certainly anticipated it in 
consciousness. It is possible, i.e., that an apparently immediate 
recognition with secondary ideas is in reality a mediate recogni- 
tion as well. The secondary ideas might exert just the same 
influence in this latter case, although they were longer in coming 
to clear consciousness. The difference between the two forms 
will then depend essentially upon the rapidity with which the 
feeling of recognition arises. If it is excited by the bare 
entrance of the secondary ideas into consciousness, we call the 
recognition immediate. If there are required a longer operation 
and a greater degree of clearness of the secondary ideas, the act 
of recognition becomes mediate. 

Now, if the difference between these two forms reduces itself 
to a difference of degree in the efficacy of the secondary ideas, it 
is plain that there can be no sufficient reason for regarding simple 
recognition without attendant circumstances as a process sid 
generis. If the secondary ideas, the whole scope of whose acti- 
vity is that of auxiliary forces, may in cases be raised to clear 
consciousness only after the recognition has taken place, it will 
surely be possible that they may entirely disappear from con- 
sciousness as soon as the result of that recognition begins to 

o o 

take effect. As a matter of fact, a closer glance at the condi- 
tions serves to make the possibility a probability. Immediate 
recognition occurs, first, where the objects are completely fami- 

Association by Similarity and Contiguity 301 

liar to us from repeated experiences, and, secondly, where we 
have come to know them but a short while before, or under 
circumstances in which they made an especially deep impression 
upon our feelings. Now these are conditions which render in- 
telligible a quick apperception of the object, with an accompany- 
ing feeling of recognition, but which by no means indicate as 
probable the absence of the otherwise universally present second- 
ary ideas. When we see a person with whom we are daily 
associating, there are so many reproductions of the most various 
situations in which we were in his company, that it is very diffi- 
cult for any particular one among them to come to clear con- 
sciousness. But at the same time there may always be operative 
a certain number of these obscure secondary ideas, which will 
explain the presence and vividness of the feeling of recognition. 
It is a little different when we see some one for the second time 
whom we met but a short time before. The recognition-feeling 
in this case is certainly not without its foundation in attendant 
secondary ideas. But these are fewer in number, and not in 
opposition to one another. They therefore possess a more defi- 
nite character, and so are, as a rule, easily perceived, if the atten- 
tion is directed upon them. In other words, the process in this 
instance appears to assume the character of an immediate recog- 
nition, for the sole reason that the connection of the secondary 
ideas with the object is still so close, that no perceptible time is 
required for the excitation of the recognition-feeling. 

It would seem, then, as though the feeling of recognition, 
which, as introspection shows, takes on very different shades in 
the different cases we have been discussing, depended in every 
instance upon the excitation of auxiliary ideas. But the time 
which these secondary ideas take to appear is not always the 
same ; and that points to another difference between the process 
of assimilation and the act of recognition. If a given impression 
calls up an earlier idea without exciting secondary ideas, 
whether clear or obscure, and without reviving the feeling which 
is dependent upon them, the result is an assimilation. Impres- 
sion and idea are combined to form a simultaneous whole ; the 
conditions of recognition are wanting. We perceive the object 
as one of a class with which we are familiar without referring it 
to anything definite of which we have had previous experience. 

.302 Lectures on Human and Animal Psychology 

And it is for this reason that we speak of an act of cognition 
rather than of recognition. In other words, we may oppose the 
act of recognition in general, as a simple case of successive 
association, to the act of cognition, which is a simultaneous 

We have already seen that contiguity-connections are involved 
even in the act of cognition. We should not be able to bring 
our visual idea of an object under the head of a familiar class, if 
the likeness-connections which arise first did not immediately 
call up contiguity-connections from earlier perceptions. But 
since these last remain wholly indefinite, they may possibly be- 
long to quite different and unrelated ideas, the result is only a 
feeling of cognition : the object is regarded as a new presenta- 
tion, but one belonging to a class of known ideas. So that, 
-although the recognition-feeling is certainly akin to the cognition- 
feeling, there is a greater qualitative difference between these 
two than exists between the different shades of the recognition- 
feeling mentioned above. And the feelings are not only differ- 
ent in quality, as is to be expected when we consider the 
different conditions under which they appear, but also in inten- 
sity ; the feeling of recognition is generally far more intensive. 
Parallel with these differences, again, run differences in time- 
relations : the recognition-feeling comes later, and its gradual 
intensification can usually be clearly followed in introspection ; 
while the cognition-feeling generally seems to appear simul- 
taneously with the impression. These are differences which are 
at once explicable when we remember the different nature of the 
underlying association-processes. 

The two feelings correspond most nearly in all their attributes 
in the case of simple recognition of persons or objects familiar 
to our every-day experience. Here the total process of recog- 
nition is very closely related to that of assimilation. On the 
other hand, the feeling of recognition is most characteristically 
itself in the case of mediate recognition. 

There can be no doubt whatever, in the case of mediate re- 
cognition, that the secondary ideas upon which the recognition- 
feeling depends are actually present in consciousness. Indeed, 
we are in this process not only conscious of the presence of the 
secondary ideas : we see quite clearly that the attendant feeling 

Association by Similarity and Contiguity 303 

is bound up with them. But what shall we say of immediate 
recognition, when these auxiliary ideas are either not noticeable 
at all, or are only found by introspection after the act has taken 
place ? We may assume either that they are below the limen 
of consciousness, and rise above it, if ;\t all, only at a later stage ; 
or that they are in consciousness throughout, but in so obscure 
a fashion as not to be perceived at first. The experiments 
made upon the different degrees of clearness of ideas with 
momentary transitory impressions, which we have already re- 
ferred to (pp. 241 ff.), declare for the latter hypothesis ; it can 
hardly be doubted that the auxiliary ideas are in consciousness, 
however dimly and obscurely. The different ways in which 
those experiments showed us that obscure ideas could make 
their presence known in consciousness correspond exactly to 
the various phenomena of immediate recognition. Sometimes it 
is possible, after the completion of the act, to represent its 
circumstances in detail ; sometimes there is left only the in- 
definite feeling that the object was seen. So that the not 
infrequent impossibility of localising the recognised object in 
space and time is no proof at all against the presence of obscure 
auxiliary ideas. On the other hand, it is at least a very im- 
probable supposition that ideas which have disappeared from 
consciousness can nevertheless exert an influence upon it in the 
form of a definite feeling. For if an idea that has disappeared 
can still excite a feeling in consciousness, it must possess posi- 
tive attributes in its unconscious condition which completely 
resemble those attaching to it as a conscious process. The 
recognition-feeling, you see, is essentially the same whether the 
recognition is immediate or mediate, i.e., mediated by clearly 
conscious secondary ideas. That supposition, in other words, 
would plainly commit us to the adoption of the untenable 
position that ideas which have disappeared from consciousness 
still persist in an unconscious condition, possessed of the same 
attributes as attached to them in consciousness : or, to put it a 
little differently, the vanished ideas would be indestructible 
objects, and not those dispositions towards the repetition of 
previous processes which the facts tell us they really are. 

304 Lectures on Human and Animal Psychology 


The process of recognition was described above in general 
terms as a simple form of successive association. To this must 
now be added that its various forms present a continuous 
serial transition from simultaneous to successive association. 
Immediate recognition, which approximates most nearly to the 
simple act of cognition, comes in all respects very near that of 
assimilation. The only indication of an ideational basis lying 
outside the recognised object is the characteristic feeling of 
recognition, which even here usually requires a certain time to 
arise. If these extraneous ideas come subsequently to clear 
consciousness, the simultaneous association passes over of its 
own accord into one of succession. And just the same may 
take place with the act of cognition. When an object has been 
assimilated by previous ideas of the same kind, one of t\vo 
things may happen : certain particular secondary ideas, con- 
tiguous with the assimilating ideas, may enter into consciousness, 
or from the whole number of assimilating ideas there may be 
singled out particular ones, which subsequently attain to a 
greater clearness. 

If in a series like this an associatively excited idea is appre- 
hended as having been previously experienced in its own special 
quality, the process becomes a successive act of memory. Such 
a memorial act results directly from the different forms of cog- 
nition and recognition when the ideational acts which are in 
them given simultaneously or almost simultaneously are divided 
up into a clearly conscious temporal series. And just in these 
transition-cases we may perceive with especial clearness the 
condition of such a temporal analysis. This condition is given 
in the fact that the separate constituents of a total associative 
complex require periods of various duration to attain to clear 
consciousness. In immediate recognition and in the simple act 
of cognition there is no clearly noticeable succession, because no 
sooner is the impression given than the elements of the assimi- 
lating ideas which cohere with it are also apprehended. But 
even in mediate recognition there is not only a distinction 
between the secondary ideas and the principal impression, but a 
temporal dissociation of them in introspection : the secondary 

Theory of Successive Associations 305 

ideas come first. This time-relation may vary further in the 
most diverse ways. The principal idea may be assimilated first, 
the secondary ideas coming later as revivals of earlier experiences ; 
that is a case of ' association by contiguity.' Or an assimilation- 
process of the usual form, involving an indefinite number of 
assimilating ideas, may run its course, and then certain of 
these ideas be held in consciousness by themselves ; that is an 
ordinary ' association by similarity.' If this consists in an act 
of recognition of the kind described above, and if there are 
further associated with that other secondary ideas previously 
contiguous with the cognised object in space and time, the pro- 
cess is one of recollection. 

The analysis of associations into temporal series depends in 
all these instances upon two conditions. First, one of two re- 
vived ideas may enter consciousness later than the other. This 
is realised in the case of mediate recognition, and in the simple 
memorial processes developed from it. Secondly, while several 
revived ideas may appear simultaneously in consciousness, and 
perhaps exert each its own influence upon the state of feeling, they 
may nevertheless be successively apperceived, coming one after 
the other to the conscious fixation-point. This is, of course, the 
case in all acts of recognition with temporal and spatial locali- 
sation. At the same time, this condition is probably quite often 
crossed by the other ; it is in consequence of this temporal and 
spatial localisation that individual ideas become conscious at 

The result of these paragraphs is to show that the connections 
operative in successive associations are the same as those which 
constitute simultaneous associations. The first thing always is, 
that certain elements of our ideas call up the like elements of 
other ideas. To these attach others which at some time or 
another were connected with them. And the whole process is 
perpetually shaped and modified by two influences, the mutual 
intensification of the like and the mutual repression of tne 
opposing elements. So that all our mental experiences are con- 
tinuous and interconnected ; the sum of ideational elements 
which consciousness has at its disposal forms an unbroken, 
interlaced and intertwined whole, within which each separate 
point can be excited from any other point by the mediation ot 


306 Lectures on Human and Animal Psychology 

those lying between them. Every idea which comes into 
consciousness, so far as it is not the direct product of sense- 
impressions, is the result of associative operations initiated in 
this continuous interlacement of ideational dispositions. And 
the sense-excitations themselves, as the phenomena of assimila- 
tion and complication show us, everywhere connect with 
elements which belong to this network of association. So that 
the ideas of memory and imagination differ from the direct 
perceptions of sense only in the degree, not in the manner, in 
which they arouse the associative activity. 


But may there not appear in consciousness ideas due neither 
to sense-impressions nor to association ? Is it not a matter of 
frequent occurrence enough that some fancy-image conies up in 
our minds for no known reason, and with no discoverable refer- 
ence to any directly preceding experience ? 

Cases like that have led to the assumption of a spontaneous 
origin of ideas. The apparently unmediated idea, it is said, had 
been inhibited by others ; and as soon as these were in their 
turn inhibited by yet other ideas, it rose of its own accord into 
consciousness. These inhibitory processes are, of course, entirely 
hypothetical. No one has perceived them, or even any definite 
facts which would allow us to infer their existence. Moreover, it 
is clear that this explanation of the ' spontaneous origin ' assumes 
once more the imperishability of ideas, or at least their persist- 
ence in an unconscious state, in which they possess exactly 
the same attributes as attach to them in consciousness, except- 
ing the property of being conscious. Of this they are tempo- 
rarily deprived by their ideational enemies. Such an assumption 
is untenable ; the ideas, as we have urged so often, must not be 
regarded as unalterable objects, but only as varying processes. 
Experiment gives us the key to the difficulty. The phenomena 
of indirect association, which are very easily demonstrated, 
explain at once the apparently ' spontaneous ' origin of ideas. 
We can reduce the whole matter to terms of association. 

An observer is stationed in a dark room, or looks into a dark 
chamber. Momentary light-stimuli are presented to him in 
regular succession. First of all, the series consists, say, of the 

Indirect Association 307 

objects a, b, c, d, e, f. . . . In a second and immediately 
following experimental series, each of these objects is connected 
with another, and in such a way that certain members of the 
series have attached to them the same secondary ideas ; e.g.,aa, 
b @,cy,d&, e a, f<y, g ft. . . . After some time the principal 
terms of the series, a, b, c, . . . are presented again in a 
different order, and without the secondary objects a, /3, 7. . . : 
e -S'-> f*b, a, g,e. . . . If sufficient time is allowed after each 
impression for the formation of an association, it is found that in 
a relatively large number of cases there are associated ideas 
from the same series, which were connected with identical 
secondary ideas ; e.g., e will be associated to a, g to b, etc. This 
result is most striking when the principal ideas, a, b y c, d. . . . are 
familiar objects (a house, a tree, etc.), and the secondary ideas, 
a, /3, 7, 8. ... arbitrary signs (e.g., letters from a language with 
which the observer is unfamiliar). In this case the secondary 
ideas which are such effective aids to association are but seldom 
clearly remembered. And so the observer, when he is asked 
why he associates the definite idea e to another, a, replies that 
he cannot tell. If you recall what we have said above as to the 
efficacy of such secondary ideas in the acts of cognition and 
recognition, you will see that we must assume in this present case 
that the secondary idea a was obscurely present in conscious- 
ness and excited e, with which it had previously been connected, 
whereupon e, which had frequency and familiarity in its favour, 
appeared by itself in the foreground of consciousness. That is, 
the only difference between this process and an ordinary associa- 
tion is, that here certain members of the associative series remain 
unknown, with the result that the connection appears to be 
broken at the places where it is mediated by them. Occasions 
to this indirect form of association cannot be rare. We shall, 
therefore, be justified in always referring to it purporting in- 
stances of the spontaneous origin of ideas in consciousness, 
although, from the nature of the case, it is only exceptionally 
that we can prove the efficacy of the unapperceived middle 




A SSOCIATIONS are due to the interconnections obtaining 
/~\ within the whole circle of our ideational consciousness. 
And it is a necessary corollary from this that all the 
relations into which ideas can enter with one another take their 
origin from those connections by likeness and contiguity which 
lie at the root of the association-process in general. But it is 
equally plain that the inference so often drawn, ' all ideational 
connections are associations,' is wholly unjustifiable. This infer- 
ence has its source in an error with which we are already familiar, 
that which transformed the forms of association into ' laws of 
association.' It rests upon the supposition that these forms are 
themselves elementary processes, whereas they are really, as 
we have seen, complex products resulting from the elementary 
connections by likeness and contiguity. But while we grant 
that all the possible interrelations of ideas are reducible to 
these two elementary types, we do not mean to assert that the 
association-products can be exhaustively and without exception 
classified under the heads of simultaneous and successive associa- 
tion. There is one limitation which must not be disregarded. 
We never speak of association except where the elements 
which mediate the connection belong to a restricted circle of 
ideas. Thus assimilation is confined to perceptions of so homo- 
geneous a character that they can be connected to form one 
single idea, complication to disparate impressions, which are 
inseparable concomitants in perception. The same thing holds 
of successive associations by similarity and contiguity, which, 

you remember, only differ from simultaneous associations in the 



(specially conditioned) temporal separation of the individual acts 
of ideation. 

Now there can be no question that we find processes in con- 
sciousness which are inexplicable in terms of these associations 
between similar or frequently connected perceptions, although, 
certainly presupposing the existence of the association-products 
Let us consider for a moment one particular class of ideas, 
. which is of all the processes of the same generic kind that which 
most resembles association as regards conditions of origin, but 
which nevertheless is quite characteristically different from it. I 
mean the ideas which we call concepts. If our eye lights sud- 
denly upon the picture of a man, the first thing that occurs is an 
effect of assimilation : we cognise the picture as that of a man 
in virtue of its likeness- and contiguity-relations to previous 
perceptions. If these relations are of so individual a nature, 
that they suggest a similarity-association with some definite 
person, the originally indefinite act of cognition passes over into 
an act of recognition. There may then further attach to this 
a number of successive contiguity-associations ; we remember, 
it may be, the circumstances in which we saw the recognised face 
for the last time, or upon some special occasion, and so on. All 
these processes arise under obvious associational conditions ; 
but not one of them gives us the concept of man. It is, indeed, 
perfectly true that if the concept happens to be a very familiar 
one it may be more or less clearly present in the association. 
In the nature of the case, however, that is not necessary : the 
simple cognition of an object as known does not in any way 
imply a concept, although simple acts of cognition and recog- 
nition must inevitably precede the formation of concepts in 
general. How, then, do we distinguish a concept from an 
ordinary idea, which is cognised as agreeing with some other or 
with several others ? It is obvious that the conceptual quality 
cannot be a specific attribute of the idea which stands for the 
concept in consciousness. There is nothing in this, considered 
apart from its relations, to distinguish it in any way from any 
other particular idea. No ! the only distinguishing mark of the 
conceptual idea consists in the accompanying consciousness that 
the particular individual idea has only a vicarious value, and 
that therefore any other particular idea which belonged under 

310 Lectures on Human and Animal Psychology 

the same concept, or was in any way to be thought of as an 
arbitrary sign of it, might be just as well put in its place. This 
accompanying consciousness is also attended by a characteristic 
feeling, a conceptual feeling, which is wholly different from the 
feelings of cognition and recognition, and points to a divergent 
ideational substrate. And this feeling, again, can only consist 
in concomitant ideational processes, running their course very 
likely in the more obscure regions of consciousness. In the 
present case, these processes must evidently be those which give 
to the concept-idea the peculiarities distinguishing it from other 
ideas. Processes of this kind are processes of judgment, as 
may be seen from the fact that concepts do not exist from the 
first in isolation, but obtain their conceptual significance only as 
elements in judgment. Whenever, therefore, we think of the 
concept in isolation, we are thinking of it as a constituent of an 
indefinite number of judgments ; no other case is possible. In 
this instance, then, the secondary ideas will be obscurely con- 
scious judgments, in which the concept occurs ; and they will 
tend in particular to be judgments which somehow contribute 
to a definition. If, e.g., we think of the concept ' man ' in isola- 
tion, we have before us at the fixation-point of consciousness 
either the image of some individual man, or the word ' man ' (as 
vicarious sign), or perhaps a complication of the optic and 
acoustic images. In its more obscure and outlying regions, 
and probably moving restlessly from this part to that, are a 
number of judgments in which the concept is involved, and of 
which only an occasional one may rise here and there to clearer 
ideation. But, obscure as they may be, they serve to invest the 
concept-idea with the consciousness of its vicarious significance, 
and with the resultant concept-feeling. You see, this signifi- 
cance and the feeling which accompanies it attach to the 
immediate introspective perception that in all these judgments 
the idea might have been a different one. 

Now we have already seen that acts of cognition and recogni- 
tion are also attended by secondary ideas, which on the one 
hand give rise to the peculiar feelings accompanying these 
processes, and on the other, if they happen to come clearly 
before consciousness, arrange themselves to a temporal series of 
interconnected ideas. But if cognition, recognition, and 

Concepts and Judgments 3 1 1 

concept are so far alike, it is just at this point that we are able 
also to lay our finger upon the essential difference between 
the first two processes and the third. This temporal series is 
always a similarity- or contiguity-association, in which (as the 
name implies) each idea persists as an independent unity. For 
objects which resemble one another, or which are contiguous in 
space and time, may certainly be combined to form more com- 
plex ideas ; but every part of the resulting compound process 
is still independent, so that if it becomes dissociated from its 
companions, it continues unimpaired in consciousness. But with 
concept-ideas, as with all conscious content which belongs to 
a logically coherent thought-process, the case is very different. 
The significance of the individual is now entirely dependent 
upon the whole of which it forms a part. Dissociated from 
this, it no longer possesses any significance of its own. Or if 
it seems to, the explanation is the same as for the concepts 
which we think of in isolation ; we can confer a significance 
upon it in so far as, for this special purpose, we leave the 
logical connection in which it belongs undetermined. Thus the 
concept ' man,' when we think of it without reference to any 
context of judgment, can only have this significance : that it 
may be subject or predicate of a large number of judgments. 
Only as such an indefinite element of a logical thought is it a 
' concept ' at all. In all other cases the corresponding idea 
would be simply and solely a concrete particular idea. 

An objection might be raised to this train of argument. ' It 
may be quite true,' you will say, 'that concepts and their con- 
nections are in many respects different from ordinary associa- 
tions arising between particular ideas. But that is no reason 
for refusing to reduce them, and all logical thought-processes 
with them, to associations in the wider sense, perhaps associa- 
tions of a peculiar and more complex nature.' The objection 
does not hold. The differences between the two kinds of 
conscious process are characteristic and fundamental, as evident 
to subjective perception as they are manifest in the uniformities 
objectively discoverable. To employ one name for both, and 
so to suggest the view that the processes are of the same kind, 
would serve not to clear, but seriously to embarrass, the path of 

3 i 2 Lectures on Human and Animal Psychology 


The most obvious subjective mark of the intellectual conscious 
process, as distinguished from association, is the accompanying 
feeling of activity. The very best means of arousing a purely 
associative, i.e., not logical, train of ideas is to assume the 
most passive attitude possible, to repress that activity of 
thought which requires volition to initiate it, and is attended by 
the activity-feeling. The question what this activity is, and in 
what the feeling of activity consists, has been already answered 
in our discussion of voluntary action (Lecture XV.). In virtue 
of its subjective characteristics, intellectual activity falls at once 
under the concept of internal voluntary action, or active appercep- 
tion. In this sense, then, we may distinguish intellectual pro- 
cesses from associations, on the purely psychological basis, as 
apperceptive connections of ideas. By a 'free' or 'voluntary' 
action, we do not, of course, here, any more than in our previous 
discussion, understand an unconditioned action. The phrase 
simply means that changes in consciousness are involved 
which are not explicable by reference to individual ideational 
connections, but only as resulting from the general tendency 
of all our conscious content at any given time, in the last 
resort, that is, from the whole of the previous development of 
consciousness. If we term the result of this total development 
our ' self,' we must regard this self as the cause of all in- 
tellectual processes. 

These considerations make it self-evident that the range of 
the subjective relations to which these internal processes owe 
their origin may vary enormously. No excitation, of course, 
can affect at once the whole number of our acquired disposi- 
tions. For the action to take rank as a voluntary intellectual 
activity, it is only necessary that groups of ideas be impli- 
cated which do not stand in any obvious associational relation 
to those directly preceding them. And, in the same way, it is 
inevitable that associations intrude upon the intellectual 
functions. And with reference to this it is especially signifi- 
cant that the intellectual ideational connections, once formed, 
themselves enter into contiguity-relations, and can therefore be 
revived in the form of external association. In this case, of 

Distinguishing Marks of Intellectual Processes 313 

course, there is no trace of the feeling of activity which else- 
where accompanies the intellectual processes. Such a transition 
from apperceptive trains of thought to association is of the very 
greatest importance : it facilitates constructive mental work in 
a high degree. In this sense it forms one of the principal 
constituents of those various practice-processes which gradually 
enable us to perform voluntary actions, at first matters of 
intention and reflection, as appropriate mechanical reactions to 
definite external stimuli. So universal is it, that for the 
accomplishment of external acts of will the interposition of a 
voluntary decision is only necessary at certain critical moments ; 
their detailed execution is relegated to the mechanism which 
practice has perfected. And, in the same way, the active work 
of thought in the intellectual processes becomes more and 
more confined to the essential moments in the flow of thought, 
while our thinking on all subordinate points goes on with no 
other aid than that of logical associations. The more practised 
in thinking we are, the more numerous become the middle 
terms which suggest themselves, and the more real force and 
energy has thought to bestow upon decisive issues. 

With these subjective characteristics are united not less im- 
portant objective peculiarities, distinguishing the intellectual 
processes from associations. They are given in the totally 
different character of the temporal succession in the two cases. 
In successive association one idea follows on the other as the 
various likeness- and contiguity-connections operative in the 
case determine. Each particular idea retains its independence. 
And since in a long associational series a new idea is regularly 
associated with only one of its predecessors, generally the most 
immediate, the series is subject to abrupt changes of the most 
varied character. Beginning and end in particular may be 
wholly unrelated, however complete the chain that connects 
term with following term. In contradistinction to this, the 
intellectual processes always begin with aggregate ideas. These 
differ from the complex ideas resulting from simultaneous 
association in that they do not consist of connections which 
(like position in time and space) appear as direct objective 
attributes of the idea ; but the relations existing among their 
constituents are regarded as conceptual determinations, into 

314 Lectures on Human and Animal Psychology 

which the complex object is analysed by the activity of 
thought. The basis of such an aggregate idea, however, is 
always a complex idea associatively produced. Thus the im- 
pression of a red house gives rise, by means of associative fusions 
and assimilations, to a complex visual idea. This only becomes 
an aggregate idea when the red colour is separated from the idea 
of the house as such. For then attribute and object are con- 
ceptually thought, and brought into mutual relation in the. 
aggregate idea. 


The first products of intellectual activity, then, are simul- 
taneous aggregate ideas. Their only difference from ordinary, 
associatively formed ideas consists in this, that the ideated 
object is regarded as analysable in terms of arbitrarily selected 
thought-relations. As soon as one or more such analyses have 
been carried out, their simultaneous connection gives rise to a 
train of thought. This process is best seen in the acts of logical 
thought in the narrower sense ; i.e., in the processes of judgment 
which are expressed in language. These are completely 
different, even in outward form, from the associative series. 
In the latter one idea joins on to another indeterminately ; but 
logical thought is governed by a dichotomic law, admitting of 
no exception, except when associations intrude upon the apper- 
ceptive train of ideas in the manner specified above. The 
clearest expression of this law is to be found in the grammatical 
distinction of the parts of the sentence. The division is either 
simple, as in the simple sentence, where subject and predicate 
each consist of one single idea; or it is multiple, as in all kinds 
of compound sentences, where each of the principal constituents 
can again be subdivided on the same plan, the subject into 
substantive and adjective, the predicate into verb and object, 
verb and adverb, etc. 

But this external law or uniformity is the result of internal 
conditions. Thought is always a discriminating and relating 
activity. That its analysis follows the above rule is due to the 
fact that it separates the constituents of an aggregate idea only 
to bring them at once into some mutual relation, a relation 
determined after the comparison of numerous ideas, partly 

Development of the Intellectual Functions 3 1 5 

alike and partly different The aggregate idea of the red house 
and the resultant judgment, ' the house is red,' can plainly not 
arise until many ideas of houses have been formed, with van-- 
ing colour-attributes. Then, and then only, will it be possible 
in a particular case either to abstract the attribute from the 
object of experience, or to refer it to an object thought as exist- 
ing independently of it. 

It might perhaps be objected to this discrimination of the 
aggregate idea from the judgment that when once we have 
separated the former from the ordinary complex idea we have 
ipso facto constituted it a judgment. It is impossible, e.g., to 
think of an object as logically connected with any of its 
attributes without this connection finding immediate expression 
in a judgment. But certain as it is that the simplest judg- 
ments (like ' the house is red ') can hardly be distinguished 
in fact from their corresponding aggregate ideas ('red house'), 
the possible confusion ceases with them. For consider those acts 
of thought which presuppose a continued process of subdivision. 
When we are about to express a complicated thought, what is- 
first of all in consciousness is the total thought in the form of 
an aggregate idea. But it is quite impossible to say of this that 
it is identical with the judgments into which it is analysable. 
No ! we can perceive well enough that while the whole thought 
is already there as an aggregate idea before its expression, its 
separate constituents are only raised to clear consciousness in 
proportion as their analysis is actually carried out. Aggregate 
ideas are, therefore, the more indefinite the more comprehensive 
they are, the more numerous the acts of judgment they require 
for their complete determination. 

We must, however, remember that the logical judgment is 
not the only, more than that, it is not the original, form 
which apperceptive ideational processes take in consciousness. 
The commonest case is, that aggregate ideas of a more or less 
comprehensive character are consciously analysed in the form 
of sense-perceptions. The. unitary character of the complex 
process means even here that each act of division within an 
aggregate idea (provided that this has been held in mind for any 
considerable time) attaches closely to its neighbours. But the 
purely perceptual nature of the contents and the absence of any 

3 1 6 Lectures on Human and Animal Psychology 

formulation of it in logical terms lead to the replacing of the 
dichotomic law by the general impression of an analytic process, 
which runs its course in orderly sequence, and sets out from a 
single aggregate idea. To this must be added that it is a more 
obvious fact in such instances than it is in conceptual logical 
thought that ideational analysis implies an ideational explication. 
The products of analysis, at first but obscurely apprehended, 
oftentimes fail to obtain a more clear and distinct content until 
they have entered into new associations. This perceptual form 
of intellectual elaboration is the activity of imagination. In other 
words, imagination is in reality a thinking in particular sense- 
ideas. As such it is the source of all logical or conceptual 
thought. But it continues to exist alongside and independently 
of the latter both in the unguided play of fancy of our every-day 
life, and in the finished creations of the artistic imagination. 


To give a complete account of the intellectual functions 
would extend beyond the limits of our present undertaking. 
Part of that account, the description of the conceptual or 
logical forms of thought, lies within the province of logic ; 
part of it, the consideration of imagination as a form of 
intellectual activity, within that of aesthetics. But it is of 
psychological interest, and desirable for the right understanding 
of the relation of association to intelligence, that we should 
pause here to cast a glance at the alterations produced in the 
train of ideas and the mental processes to which it gives ex- 
pression by the different forms of mental derangement. 

The most clearly marked and most permanent of these 
derangements are the various kinds of insanity. The particular 
forms of insanity, as you know, are so many and so different, 
that pathological psychology has as good a claim to rank as 
an independent discipline, beside normal psychology, as has 
the pathology of the body to be separated from its physiology. 
This latter separation was long ago effected ; the two disciplines 
are independent sciences, quite apart from the fact that patho- 
logy has also practical applications. Add to this that 
every mental derangement involves, besides alterations of the 
intellectual processes and associations, other and equally 

Mental Derangement 


fundamental modifications of the mental life: especially de- 
rangements of sense-perception and of the emotions, as 
compared with which the ideational disturbances often appear 
as merely secondary consequences, though it is true, and you 
will understand it when you consider the inextricable inter- 
dependencies of all these mental processes, that these altera- 
tions of ideational content will react again upon the affective 
and conative side of consciousness. Here, however, we shall 
leave all this out of account, and simply look at mental de- 
rangement from the single point of view of alteration in the train 
of ideas. And in this respect we shall find it in its funda- 
mental character, despite differences of detail, uniform and 
homogeneous throughout. 

This much may, however, be said with regard to the altera- 
tions produced by mental derangement in the spheres of sensation 
and emotion : that, regarded purely psychologically, they 
embrace the most various divergences from the norm in all 
directions, from the apathy of idiocy, which is only to be 
moved by the most intensive sense-impressions, to the enormous 
excitability of delirium, when the slightest external or internal 
stimulation suffices to call up hallucinations and misleading 
illusions ; or from the deep depression of melancholy, which 
clouds the present and the past alike, to the bursts of maniacal 
passion and the immovable cheerfulness of paralysis. The 
deviations from the normal train of ideas will, of course, be 
correspondingly different ; its course will be too quick or too 
slow, too crowded or too sparse, as the case may be. And 
these deviations stand in an intimate relation to emotional 
changes ; they are not really disturbances of which these latter 
are independent, but the two derangements are both implied in 
any mental disturbance. In melancholy and paralysis alike, 
the train of ideas moves haltingly, arrested by definite impres- 
sions and memories, only that the affective colouring is wholly 
different ; while in mental exaltation and mania it is accelerated, 
springing from topic to topic without order or control. 

All the more noteworthy is it, then, that, despite all these 
differences, the departure from the norm is constant in one 
particular respect, in regard to the relation of associations to 
the intellectual processes. If there is any single criterion of 

3 1 S Lectures on Human and Animal Psychology 

mental derangement, it is this, that logical thought and the 
voluntary activity of the constructive imagination give way to 
.an incoherent play of multifarious associations. If the de- 
rangement has not gone very far, e.g., in the first period of a 
slowly developing disease, this change in conscious content 
may be hardly noticeable, either because long intervals of 
sanity interrupt the progress of the derangement, or because the 
latter appears to confine itself to some particular ideational and 
affective connection. But even in these cases, borderland- 
cases, midway between the normal and the abnormal, where a 
good natural constitution may more than hold in check the 
encroachments of a disease, even in such cases there can be no 
doubt that from the moment at which consciousness is over- 
come by the derangement the normal equilibrium of association 
and active apperception is once for all destroyed. 

The most general way in which this disturbance of equilibrium 
manifests itself is by a defective concentration of tJte attention. 
It arises from the liability of the intellectual processes to be 
continually interrupted by sudden associations. And the states 
of mind in which the patient is always concerned about 
particular impressions or feelings are only an apparent ex- 
ception to this rule. When the melancholiac broods incessantly 
over the crushing sorrow which he supposes himself to be ex- 
periencing, it is not that he voluntarily directs his attention upon 
it, and so controls the direction of his thoughts ; his mind is 
dominated by an ever-present group of intensively toned ideas, 
against which the will struggles often enough, but struggles in 
vain. In the condition of mental exaltation we have an un- 
naturally strong excitation of the sense-centres, giving to the 
associational contents the character of external sense-impres- 
sions ; so that these associations, which normally make against 
the influence of the active attention, are rendered unusually 
powerful. Ideational assimilation plays an especially large part 
in this case. In normal mental life the assimilating: elements 


are just strong enough to render acts of cognition and recogni- 
tion possible ; in hallucination they become so potent as to 
throw the sense-impression into the position of a mere external 
accident, which sets in motion ideational tendencies that have 
not the remotest resemblance to itself. 

Mental Derangement 319 

Notice the way in which the insane express their thoughts. 
Their language is abrupt ; it will vacillate in a purposeless way 
between the most heterogeneous subjects, or it will come back 
-again and again to the same topic, for no assignable reason. 
All this is not hard of explanation, if we suppose a lack of 
voluntary control over the unruly crowd of associations. Of 
course, this lack of control may exist in very different degrees, 
from the eccentricity of thought which just oversteps the norm 
to the wildest flights of fancy, in which thought follows thought 
without the possibility of the mind's dwelling for even a short 
time upon any. The last stage is a total incapacity to frame 
any judgment that is at all complex. The patient begins to 
utter some sentence. But his attention is taken captive by new 
sense-perceptions, or by some extraneous circumstance, perhaps 
by associations aroused by the sound of his own words. So 
another and heterogeneous train of thought is set going, only 
to be interrupted in its turn by still other associations. And 
so the mad hurry goes on, until mental exhaustion puts a 
temporary end to the tangled drift of ideas. 

When we consider the vast importance of language for the 
development of thought, we can understand that associations of 
words and sounds play the leading part among the thousand 
forms of ideational involvement in the fancy-flights of the 
insane. Words of like sound are heaped together in meaning- 
less confusion in the middle of a sentence ; or a word will 
suggest some totally heterogeneous thought, in which it also 

O O i/O O 

happens to occur. So that the speech of the insane is our best 
opportunity to observe ' similarity '- and ' contiguity-associa- 
tions ' in simultaneous and successive form, and with Che utmost 
variety of content. But it not seldom happens that when the 
.flight of ideas is covering a large range of topics some particular 
word will call up not another word, but perhaps simply a series 
of articulate sounds, belonging, it may be, to like-sounding 
words, or contained in others which have accidentally become 
connected with the first. When this is the case, the language 
of the insane becomes an unintelligible jargon, composed of 
-articulate sounds that occur in real speech, but putting them 
into quite new connections. There may then arise in the 
patient's mind the delusion that he is speaking a previously 

320 Lectures on Human and Animal Psychology 

unfamiliar language, an idea which in its turn may give rise to 
other delusions. But if we consider with some attention the 
bewildering confusion of sound-conglomerations whose genesis 
we have traced, we shall see that even here the influence of 
practice, so important for association in general, is distinctly 
traceable. The more frequently a particular sound-complex 
has been repeated, the greater is the inclination to utter it 
again. More especially does it tend to enter into fresh associa- 
tions, whether with other sounds or with external objects ; and 
so there may arise a dialect of insanity which in certain of its 
constituents possesses all the characteristics of a new-formed 
language : certain sounds or sound-complexes become de- 
terminate symbols of definite concepts. At the same time we 
can hardly say that this language is the invention of the insane. 
It owes its origin to the blind chance of associational activity ; 
and this continues to modify it in the most haphazard way. 

There is scarcely anything more interesting to the psycholo- 
gist than the observation of the gradual decay of the intellectual 
functions as manifested in the language of the insane. Written 
records are even better for this purpose than oral speech, for 
the torrent of words, which must flow with incredible rapidity 
to keep pace with the flights of fancy, cannot readily be 
followed. There are printed works extant in all literatures 
that declare themselves as products of a deranged mind. In 
them we have most beautifully demonstrated the separate 
trains of association, their intrusion upon the course of logical 
thought, the gradual disintegration of this latter, and not 
infrequently the influence of hallucinations and fantastic mis- 
interpretations of sense-impressions. The final stages of this 
mental decay are, of course, usually lacking ; but all the rest 
are there. But I chanced once to pick up a book, ' privately 
printed for the author,' as you can imagine, in which nearly 
every stage of the process of disintegration, from one end to 
the other, could be clearly traced. Its opening sentences are 
correct in form and expression, although their content shows 
from the first the beginnings of abnormal thought. Then 
follow, with increasing frequency, descriptions of unmistakable 
hallucinations and linguistic solecisms, while the intrusion of 
disconnected associations makes itself more and more evident ; 

Mental Derangement 321 

O J 

till finally, on the last few pages, there is not a single sentence- 
that is brought to its correct grammatical conclusion. 

When we subject these phenomena to a careful analysis, 
especially as they are exhibited in such permanent form in 
productions of the diseased mind which are more accessible 
to investigation than the spoken word, we see at once how 
inexact and superficial it is to speak of them as due to 'a lower- 
ing of the mental functions.' As regards the flights of fancy 
for instance, tne mentai functions are rather raised in a pa:- 
ticular direction than lowered. The normal mind has not at 
its disposal anything like such an abundant supply of associa- 
tions as is not seldom met with in mental derangement. At 
the same time, it is in this very mobility of association that 
the germ of decay is to be looked for. It is the unfailing 
symptom of the weakening of voluntary control over all those 
connections of the ideational elements which the manifold 
ramifications of the associational network render possible in 
consciousness. We may safely say that no intellectual function 
is possible until these relations and connections have been 
collected from previous impressions. But, nevertheless, mental 
activity only becomes intellectual when the total torce resulting 
from the whole sum of these previous experiences, the will, 
controls and gives definite shape to the associative material 
lying to its hand. In relation to these associations, the will 
is at once an active and an inhibitory force, it furthers the 
connections concerned in the predominant interest of the 
moment, while it inhibits all that might draw the attention 
some other way. You see, then, that a sane man may volun- 
tarily call up experimentally, as it were, a train of ideas which 
very nearly corresponds to that of the insane. He has simply 
to repress the regulative and inhibitor}- function of the will with 
regard to the associations which crowd into consciousness. Put 
yourselves in this condition, and write out the thoughts and 
ideas which come to you ' of themselves ' ; you will have an 
inextricable tangle of fragments of half-completed thoughts, 
of chance impressions, with here and there a new-formed 
association, a picture that you might easily take to be that 
of a deranged mind. 

It is true that this luxuriant growth of associations, which 


322 Lectures on Human and Animal Psychology 

dwarfs the intellectual functions as creepers cramp and st ,nt 
the trees that they entwine, is not a permanent condition. If 
the process of derangement continues, the associations becone 
increasingly restricted to fewer and more stable ideas, which 
repeat themselves over and over again. These ' fixed ' ideas, 
called up in the first place by the particular tendency of the 
diseased mind, become more and more insistent as the process 
of associative practice keeps even pace with the general dis- 
integration. When these fixed associations become exclusively 
dominant in consciousness, the influence of the will is destroyed 
once and for all. At the same time, the patient becomes less 
responsive to external stimuli. The emotional depression con- 
sequent on disquieting hallucinations, the affective disturbances 
caused by painful impressions, disappear as the sensibility 
becomes less, and the general mental dulness greater ; and give 
way to a mood of uniform cheerfulness or indifference. And 
with that is reached the final stage of mental derangement, 
the saddest for the observer, the most happy for the patient 




WE have seen that a person of sound mind is able of his 
own will to give himself up to the play of association, 
and so induce a state of mind which more or less resembles 
the ideational condition of the insane. That is not all, however. 
We are all of us normally subject to experiences which bring 
us still closer to a realisation of mental disturbance. One such 
condition of what we may call normal temporary ipsanity **- 
that of dreaming. 

In every relation of life we find the omne igiiotum pro mag- 
nifico borne witness to. Mankind tends always to regard the 
unaccustomed as more wonderful than the usual and normal 
The glamour of mystery surrounds the unfamiliar, just because 
it is unfamiliar ; while the commonest phenomena, which so 
often present the really most difficult problems, are looked 
upon as matters of course. Former ages regarded the insane 
as favoured of Heaven and illuminated above their fellows, or 
as possessed of devils, according as the pendulum of circum- 
stance swung. And even to-day the subjective ideas of these 
unfortunates are at times affected by such thoughts : thoughts 
which arose in the first place from the contemplation of mental 
derangement in its various forms. Even after this view had 
died out as regards insanity, dreams were still invested with 
something of the miraculous. The popular belief in premoni- 
tion by dreams we need not notice. But there are still philo- 
sophers who incline to think that when we dream the mind 

324 Lectures on Minium and Animal Psychology 

has burst the fetters of the body ; and that dream-fancies 
transcend the activity of the waking consciousness, with its 
close confinement to the limits of space and time. 

An unprejudiced observation of the phenomena of dreaming 
must convince us that, beautiful as these theories are, they are 
pretty nearly at the opposite pole from truth. When we are 
awake, we are generally able by an effort of will to overcome 
petty bodily distractions without much difficulty. The dreamer 
is absolutely at their mercy ; the train of his ideas is diverted 
by every chance impression that affects his senses, by every 
accidental association. The commonest causes of the most 
vivid dreams are indigestion, palpitation of the heart, dimculty 
of breathing, and troubles of that sort. It is a matter of 
dispute whether there is any such thing as dreamless sleep ; 
and it will probably always remain so, seeing how easily we 
forget what we have dreamed. But it is certain that if such 
a state does occur, it will be most easily found in cases where 
all bodily stimuli are wanting, or where they are at leas: 
too weak to call up ideas. 

The physiological nature of sleep we need not stop to discuss. 
But little is known about it, if we except the general fact that 
sleep is one of those periodic vital phenomena which originate 
without exception in the central nervous system. There is also 
one fact that is psychologically important in a teleological 
consideration of life : that during sleep there takes place a 
reparation of the forces expended in the waking state by the 
functioning of the various organs. Sleep is not seldom hin- 
dered in the performance of this important duty by its attend- 
ant, the dream. Vivid and unrest ful dreams will detract 
from the refreshing effect of sleep. The actual phenomena of 
dreaming, however, make it probable that its greater or less 
interference with sleep is due to a greater or less degree of 
abnormal irritability in the sense-centres of the brain or in 
particular parts of them ; this in its turn being caused, perhaps, 
by a disturbance of the intracranial circulation. A confirma- 
tion of this view is found in the fact that pathological altera- 
tions in blood and blood-supply (such as occur, e.g., in fever) 
may considerably intensify dream-phenomena, and even occa- 
sion similar mental conditions, those of febrile delirium, in the 
waking state. 

Dreams 325 

We have already, then, in outline the essential characters 
of the dream-idea. It is a hallucination ; its intensity i= as 
great as that; of a sensation given in direct perception, and it 
is as such that it is regarded by the dreamer. The principal 
d ream-constituents are memorial images, but memories which 
are interwoven altogether at random, whether they refer to the 
immediate past or to some more remote experience, whether 
they belong together or are wholly unrelated, by the unregu- 
lated play of association. Dreaming has, therefore, at first 
sight, some resemblance to the normal activity of imagination ; 
it tends to combine memory-ideas in new and unaccustomed 
ways. But it entirely lacks that purposive arrangement and 
grouping of ideas which is the one criterion for the discrimina- 
tion of imagination from the activity of memory. 

The world of memory and the world of dreams are alike 
dominated by ideas of sight. Auditory ideas are also found 
to occur. No other sense appears to furnish dream-material 
to any considerable extent except when directly stimulated 
from without. Of course, there may be direct external excita- 
tion in the cases of sight and hearing also ; and indeed it is 
probable that dream-ideas are aroused in this way far oftener 
than is generally assumed. It may even be that the predomi- 
nance of the visual idea in the world of dreams is to be 
accounted for not solely by the very great importance of sight 
for memory, but also by the peculiar nature of the eye, which 
is more exposed dian any other sense-organ to the continual 
operation of weak external stimuli. If we look attentively 
at the darkened field of vision of the closed eye, we notice 
an unceasing appearance and disappearance of light-pheno- 
mena : now single points of light will shoot like meteors from 
side to side ; now a veil of twilight is drawn across the blacker 
background ; now again parts of this appear in the most 
brilliant colours, There can be no doubt that these pheno- 
mena persist during sleep to call up the memory-images that 
resemble them, and of which consciousness possesses so large 
a store. 

Dreaming, then, is related to the train of ideas in the waking 
consciousness in that its proximate cause is usually some 
external sensory excitation, to which memorial images readily 

326 Lectures on Human and Animal Psychology 

attach themselves. But in two respects the processes are very 
different. The ideas called up by the sense-impressions are 
more or less fantastic illusions ; and the consequent successive 
associations do not possess the character of ordinary memory- 
images, but of hallucinations : and, like these, they are taken for 
actual experiences. It is, therefore, very seldom, if at all. 
only at the moment when sleep is passing over into waking, 
that \ve dream of remembering anything. Dreaming is as 
immediate as any experience can be. It shows no trace of 
the usual discriminatory marks of imagination and reality. 

If dreaming reminds us, from this point of view, of mental 
disturbance, it has one characteristic which does not appear 
in the same degree in any of the manifold forms of insanity, 
its complete restriction to the ideas of the immediate present. 
The illusions or hypostatised memorial images of the deranged 
mind can never do more than partially prejudice the normal 
apprehension of an object ; while there is a clear distinction 
drawn between them and the customary activity of imagination 
and memory. And dreaming occupies a peculiar position in 
yet another respect. If we look at the part played in it by 
hallucination, we shall be inclined to parallel it with the initial 
stages of certain mental disturbances, which bring with them 
an abnormal excitability. But in the incoherence of dream- 
ideas, in the clouding of judgment and the lapse of self- 
consciousness, we have a series of phenomena which only find 
a parallel in the most extreme forms of mental disorder. 
Probably the greater number of dreams come and go without 
involving any really intellectual process at all. The dreamer 
acts, or looks on at action, without ever making his experience 
the subject of reflection. It is generally when dreaming is, as 
it were, dovetailed in with waking, just before we fall asleep 
or shortly before waking, that a real activity of intelligence 
is noticeable : we make dream-speeches, or carry on dream- 
conversations. But the language used is of a curiously mixed 
kind. Sometimes there is an almost normal capacity of con- 
nected expression, though when we analyse we find that the 
dream-speech consisted entirely of familiar phrases and current 
turns of language. Sometimes there is no normal connection 
in the thoughts uttered ; the whole is a strange medley, the 

Sleep-walking 327 

judgments meaningless, the conclusions wrong. It may be that 
the confusion of thought extends even to the sound of the 
words employed ; so that we have new formations of articulate 
syllables, like those which occur in the talk of the insane. And 
these are connected, by the way, with the same set of subjective 
ideas ; the dreamer, like the maniac, thinks that he is speak- 
ing fluently a new and unknown language, or perhaps some 
real language, which he has actually studied, but only very 
imperfectly mastered. 

All these phenomena tend to show that the relation of the 
intellectual functions to associations in dreaming is altered very 
much as it is in the more advanced stages of mental derange- 
ment. The control of the will over the mob of ideas and 
feelings has been abrogated. The dreamer is completely at 
the mercy of associations instituted by accidental external 
impressions. And, in addition to that, the hallucinatory 
character of dream-ideas gives them their peculiar ability to 
pass for real experiential events. 

A special kind of dreaming, which is generally a symptom 
of an abnormal excitability of the nervous system, is sleep- 
walking. It is a dream carried one step beyond the halluci- 
natory conversations mentioned above. For sleep-walking con- 
sists simply in this : that the connections between conscious 
states and external voluntary acts which hold in our waking 
life are realised to the same extent during sleep. And since, 
of all these connections, that of idea with the muscles sub- 
serving language has become the most customary and auto- 
matic, we can understand that these will be most often and 
most readily exercised. Sleep-walking, then, like dreaming, 
has nothing mysterious about it ; it is simply an event of 
infrequent occurrence, which for that very reason has sometimes 
been looked upon as inexplicable. Indeed, when we consider 
the character of the connections obtaining between sensations 
and the movements which they stimulate, it becomes rather 
a matter for surprise that dream-walking is not a far commoner 
phenomenon than it is. We may explain the facts in one of 

328 Lectures on Human and Animal Psychology 

two ways. We can assume either that the sensory centres ar^ 
more exposed during sleep to the operation of the various 
causes of excitation than the motor centres are, or that the 
latter are in general subject to certain inhibitory influences. 
However this may be, there can be no doubt of the very great 
utility of this separation of the world of dream-ideas from the 
sphere of external action. Think what would happen if we 
actually did everything that in our dreams we imagine we 
are doing ! 

But sleep-walking presents other differences from ordinary 
sleep. Especially worthy of note is an increased excitability of 
the sense-organs in presence of external stimuli. The sleep- 
walker sees and cognises external objects up to a certain 
point. But his dream-perceptions are of an illusory character, 
and so he misinterprets them : he may take the window for 
the door, or the ridge of the roof for a boulevard. While, 
therefore, he can perform simple acts, and especially such as 
have become more or less automatic by practice, he will hardly 
go beyond these. The tales that are told of wonderful dream- 
performances, the sleep-walking mathematician who solves a 
difficult problem or the schoolboy who regularly does his 
work in this very convenient way, we may consign without 
more ado to the limbo of the fabulous. No reliable observer 
has ever confirmed reports like these ; and they conflict with 
all that we know of the nature of dream-ideas in general. 


It is but a short step from sleep-walking to phenomena 
which have lately formed the topic of much discussion, the 
phenomena of hypnotism. The principal condition for the 
induction of the hypnotic state is a vivid idea of a passive 
surrender of the will to that of some other person, who is 
able to influence his subject by words, acts, or gestures. An 
abnormal excitability of the nervous system favours this influ- 
ence. But the experiment will, as a rule, succeed in the long 
run even with persons who at the outset were proof against 
it ; or, in other words, the frequent repetition of the state 
facilitates its induction, and furthers the passage of the initial 

Hypnotism and Suggestion 329 

stages into the higher. Other conditions which have often 
been regarded as auxiliary or as the sole causes of hypnosis, 
especially weak and uniform stimulation : the steady gazing 
at an object, the 'magnetic' stroking of the skin, and so 
on, are, obviously, only secondary and indirect means to the 
end. They serve in part to weaken the active attention, in 
part to arouse the idea of an influence militating against the 
independence of the subject's volition. It was proved by the 
cures worked by Mesmer and his successors in terms of 'animal 
magnetism' (which is in essentials just the same thing as 
hypnotism) that it was necessary for the success of the experi- 
ment that the patient should believe in the efficacy of the 
strokings and other manipulations, but that this belief was 
sufficient ; so that the passes and all the rest of it might be 
left out, if only the subject thought that they were there. 

The symptoms of hypnosis vary according to the stage which 
the condition has reached, and the susceptibility of the subject. 
We can distinguish three degrees of it, which, from their resem- 
blance to the corresponding stages of normal sleep, have been 
termed drowsiness, light sleep, and deep sleep. The similarity is, 
however, confined to merely external characteristics, and in 
particular obtains only for the appearance and behaviour of 
the subject before definite iniluences are allowed to play upon 
him and assume control of his perception and volition. It is 
this possession, as we may term it, which really differentiates 
the two states. Even in the light hypnotic sleep 'suggestion,' 
the influencing of ideation and volition from without, begins 
to play a part. The hypnotised subject cannot open his eyes 
of his own accord, cannot perform any voluntary movement 
whatever, though he recovers this power the instant that an 
action is suggested to him by a word of command from the 
hypnotiser. The skin is anaesthetic, which is never the case 
in sleep ; so that needle-pricks are often not sensed at all, or, 
if they are, only as pressures from blunt points. Conjoined 
with anaesthesia we find all the various phenomena of ' auto- 
matic reaction to command.' The subject executes movements 
that are suggested to him, puts his limbs in the most uncom- 
fortable positions, and keeps them there until another command 
suggests relaxation. In many persons a rigid or tetanic state 

330 Lectures on Human and Animal Psychology 

of muscles appears even in the absence of suggestion. On. 
waking, the hypnosis can be dispelled instantaneously by a 
word from the operator, the subject generally retains some 
sort of cloudy recollection of what occurred to him during his 

This power of recollection serves to distinguish the lighter 
trom the deeper hypnotic sleep, somnambulism, as it is also 
called, after which the memory is a simple blank. All the 
other symptoms are also much exaggerated. The automatic 
reaction in particular extends beyond movement to sense-per- 
ception. The somnambulist will objectify any ideas that are 
suggested to him. There are two proofs of the intensification 
of ordinary fancy-images, arbitrarily aroused by the suggestion 
of the hypnotiser, to hallucinations : first, this confusion be- 
tween imagination and reality, and secondly, the production of 
complementary after-images of the suggested perceptions. Here 
is a particular observation to illustrate the second fact. The. 
operator calls out to the subject, ' Look at that red cross on 
the wall ! ' When the latter has found it, he is told to look 
on the ground, and asked, ' What do you see there ? ' The 
answer is, 'A green cross.' That is, the after-effect of the 
hallucination is just what that of the actual impression of a red 
cross would have been (cf. p. 109). Illusions of taste are also 
very easily induced. The subject will take a glass of water for 
champagne, and drink it with every sign of satisfaction ; while, 
if he is told a moment after that he has been drinking ink, he 
will spit it out with equally evident marks of disgust. At the 
same time I am not sure that all these are cases of real hallu- 
cination. When we consider how very rarely hallucinations of 
smell and taste occur in dreaming, we shall be inclined to think 
that in these instances of suggestion too the sensations com- 
posing the ideas may be confined to those of mimetic movement* 


There are many other phenomena, manifested especial!}- in 
the somnambulistic stage, which have often been employed to 
shroud the hypnotic sleep in the veil of mystery and wonder. 
Thus a suggestion may very readily lead to the formation of the 
idea that the subject is to obey the suggestions and commands 
of one person only, generally of the operator, while remaining 

Autosuggestion and Post-hypnotic 1 mine nee 331 

indifferent to any attempt at influence on the part of others. 
There then arises what the animal-magnetism school term the 
rapport of the medium with the magnetiser. This is really, as 
we have indicated above, only the result of a secondary sug- 
gestion, which is favoured by the special circumstances of the 
induction. The exclusive direction of the attention of the sub- 
ject upon the operator is itself at times sufficient to produce 
this rapport, even without any express command, especially if 
he is always put to sleep by the same person, as is of course 
the case in the instances of ' magnetic ' cure. There is no 
intrinsic reason, that is, why the hypnotic subject should not be 
accessible to other influences. Without them we should be unable 
to explain the fact of antosu^cstion. Autosuggestion__irmolies 
an_ abnormally strong tendency towards hypnosis. Frequent 
hypnotising may lead in the long run to an irresistible passion 
for the hypnotic sleep, in which case the impulse to obtain it 
acts like the morphine-habit or habituation to any particular 
stimulant or sedative. The confirmed hypnotic will try in every 
possible way to procure the enjoyment which he craves. And 
he has in autosuggestion, when he has once discovered how to 
use it, a means lying always ready to his hand. By voluntarily 
arousing and fixing the idea that he will fall into the hypnotic 
sleep, he can induce all the phenomena that ordinarily accom- 
pany it. It appears, moreover, that in the condition as brought 
on both by autosuggestion and by suggestion from without 
there may be a continuous memory from sleep to sleep, such 
as is sometimes observed in dreaming and in certain forms of 
periodical mental derangement. The psychology of the act 
of recollection gives us the key to the explanation of this pheno- 
menon. It is wholly unnecessary to assume the existence of a 
mysterious mental double, the 'other self or second personality, 
or to set up any other of the fanciful hypotheses so plentiful in 
this field. There is, as you know, one invariable condition of 
the occurrence of an act of recollection : if we wish to bring 
about a complete renewal of a past experience, we must repro- 
duce the whole of the ideational and affective content of con- 
sciousness which characterised that experience. Now, since there 
is a great difference between the waking consciousness and that 
of the hypnotic sleep and since the difference increases with 

33- Lectures on Human and Animal Psychology 

the progress of that sleep, we can readily understand that its 
suggestions will be forgotten on waking ; while we also see how 
the recollection of those suggestions will be possible when the 
hypnotic state is renewed. 

Many cases of the post-hypnotic effects of suggestion appear 
explicable on these two principles of the conditions of memorial 
functioning and of autosuggestion. When, e.g., it is suggested 
to a somnambulist that he will perform a definite action at 
some particular time of another day, take a certain walk, exe- 
cute a given order, mix a special draught and offer it to some 
third person who is described in detail, the idea recedes into the 
background of consciousness at the moment of waking ; but, 
as the appointed time draws near, comes up again in obscure 
form, as the 'feeling' that something particular is going to 
happen. This idea of the time of the performance of the action, 
usually furthered by special insistence upon it in the primary 
-uggestion, is still more intensified, in accordance with the 
^jneral laws of association, when the time actually arrives ; and 
from this moment the indefinite idea of a commission under- 
taken,- ivJiat commission is still wholly unknown, exerts an 
autohypnotic effect, There follows a partial revival of the 
somnambulistic state, sufficient to re-excite the memory of the 
suggested ideas, and at the same time to exclude any considera- 
tiun of the motive or purpose of the action. This is then 
performed in an automatic, lethargic condition, which, however, 
as an imperfect repetition of the previous complete somnam- 
bulism, does not preclude the memory of it in the normal 
\\aking state. Asked why he did so-and-so, the awakened 
subject is unable to give any explanation ; or simply says that 
he could not help it, or that it was suggested to him in sleep, 
the latter answer giving clear evidence of a persistence of the 
suggested ideas in an obscure form into the waking state. 
Where the post-hypnotic effect takes place immediately after the 
awaking from somnambulism, it is probably the direct result of 
this persistent operation on the part of the suggested ideas. It 
looks sometimes, under these circumstances, as though the sub- 
ject were not really fully awake ; his behaviour is more like that 
of a person just aroused from ordinary sleep. Indeed, those of 
you who are subject to very vivid dreams may now and again 

Autosuggestion aiul Post- hypnotic Influence 


observe quite analogous phenomena \vhen you are suddenly 
awaked from one. You think and act for a short while in terms 
of the preceding dream-ideas ; but waking impressions keep 
continually mixing with them, until at last they gain the upper 
fland, and you are fully awake. 

But it is plain enough that autosuggestion cannot be the 
exclusive cause of post-hypnotic effects. When it is a question 
of executing some simple order directly after waking, or after 
che lapse of a definite time, the actions very frequently follow 
A'ithout any symptom of a partial relapse into hypnosis. We 
must therefore suppose that the suggested idea, with its corre- 
sponding motor impulse, may he latently operative in conscious- 
ness ; so that it will come to the conscious fixation-point either 
immediately after waking or at the time suggested. It will 
then, like every other impulse, continue to exert its influence 
until the action results, or it is inhibited by opposing forces oi 
the waking consciousness, themselves impulses, sensory or in- 
tellectual. This view is confirmed by the frequent occurrence 
of phenomena of resistance, often successful, to the performance 
of the suggested action. At the same time, there is positive 
proof that even here consciousness docs not at once return to 
its normal condition, proof furnished by post-hypnotic hallu- 
cination. The awakened somnambulist, in obedience to sug- 
gestion, may see the operator in fanciful costume, perhaps 
with a red cloak over his shoulders and horns upon his head ; 
he will find a flower in his buttonhole which is not there ; or he 
will overlook a doorway, and declare that the room has no 
outlet. It is obvious that such hallucinations and illusions 
require us to assume an extreme excitability of the sensory 
centres, such as certainly does not exist after waking from 
normal sleep. 

You are familiar with the very astonishing cures now and 
again worked by physicians who employ hypnotism therapeu- 
tically. These must be referred to the post-hypnotic effects of 
suggestion. It cannot be disputed that a cautious and intel- 
ligent use of suggestion may be of avail for the temporary, 
perhaps even for the permanent, removal of diseases due to 
functional derangement of the nervous system, or to harmful 
practices, like alcoholism or the morphine-habit. But it is an 

354 Lectures on Human and Animal Psychology 

equally undeniable fact that suggestion is in the long run just 
as ineffective for the cure of diseases arising from some palpable 
pathological cause as would be any other form of command to 
the patient to grow well again. 

There is, indeed, one exception to this rule, an exception 
which is explicable from well-known physiological facts. Mental 
influences may, of course, affect the functioning of the bodily 
organs, and especially the excitations of the vasomotor and 
secretory nerves. And suggestion takes rank with other mental 
influences in this regard, becoming increasingly efficacious as 
the subject surrenders himself more and more permanently to 
the power of the suggested ideas. Thus an arrest of any par- 
ticular secretion, provided always, again, that the derangement 
is not due to pathological conditions seated in the organ itself, 
may be overcome under the influence of suggestion. Dilata- 
tion of the blood-vessels, with all its consequences, may be 
suggestively induced, particularly if actual external stimuli are 
present to help produce the effect. An innocent scrap of 
postage-stamp paper, stuck upon the skin, has been known to 
answer all the purposes of a blister, if the idea were suggested 
that it was really a blister which was being applied. It is true 
that these phenomena cannot be obtained in the case of every 
hypnotised subject, or even of every somnambulist: the right 
disposition is requisite for the manifestation of such intensive 
organic effects. As for the physiological results of suggestion in 
themselves, they are simply more intensive and permanent forms 
of familiar and universal relations existing between mental states 
and bodily processes. If the transient emotion of shame can 
normally bring about a temporary distension of the blood-vessels 
of the face, it is aiter all not surprising that an abnormal excita- 
bility of the vasomotor and secretory nerve-system, combined 
with a domination of consciousness by ideas and feelings tending 
definitely to oppose the accompanying mental disturbance, 
should condition a more intensive and permanent physiological 
reaction to mental stimulus. And in saying this we are stating 
that hypnotism as a therapeutic agency is a two-edged instru- 
ment. If its effects are strongest when the patient is predisposed 
to it in body and mind, or when suggestion has become a settled 
mode of treatment, it may obviously be employed to intensify 

Errors of Ike 'Hypnotism-psychology ' 335 

or actually induce a pathological disposition. It must be looked 
upon, not as a remedy of universal serviceability, but as a poison 
whose effect may be beneficial under certain circumstances. We 
find, of course, not only the dabbler in hypnotism, who has no 
claim to a judgment on the question, and in whose hands the 
practice of suggestion becomes a public nuisance, but also the 
physician, to whom thinking men will no more deny the right 
to employ this dangerous remedy in certain circumstances than 
that of using any other, asserting that the hypnotic sleep is not 
injurious, because it is not in itself a pathological condition. But 
surely the facts of post-hypnotic hallucination and the diminu- 
tion of the power of resistance to suggestive influences furnish 
a refutation of this statement which no counter-arguments can 
shake. It is a phenomenon of common observation that fre- 
quently hypnotised individuals can when fully awake be per- 
suaded of the wildest fables, and thenceforth regard them as 
passages from their own experience. 

But we are not concerned in this place with the physiological 
effects of suggestion, and its consequent significance for thera- 
peutics. For psychology the questions of special interest ar-j 
that of the alterations of consciousness occurring during the 
hypnotic sleep, and, connected directly with this, that of the 
nature of suggestion. The difficulty of the former lies in the 
impossibility of actual introspection on the part of the hypnotic 
subject. When aroused from the lighter form of hypnotic sleep, 
he has no clear recollection of what has taken place ; while after 
the somnambulistic state the memorial activity is in complete 
abeyance. Observation is therefore more difficult here, if that 
be possible, than it is in the case of dreaming. So there is all 
the more opportunity for fanciful hypotheses, to which the lay 
mind is tempted by the unusual and apparently mysterious 
character of the phenomena. And the lay mind is unfortunately 
an all too frequent possession of those who have desired to 
make hypnosis the object of psychological observation. Most 
hypnotic investigators are either physicians, who employ sug- 
gestion for therapeutic purposes, or philosophers, who think that 

336 Lectures on Hinnan a/ui Animal Psychology 

they have discovered in hypnotism a basis for new metaph\\>ical 
systems, and who, instead of examining the phenomena in the 
light of well-established psychological laws, reverse the matter 
and erect their psychological superstructure upon hypnotic 
foundations. So that it is hardly to be wondered at that the 
modern hypnotism-psychology has time and again manifested 
its descent from spiritism. Clairvoyance and the magic of tele- 
path}* play a suspiciously important part in it ; and though 
there are found observers who have remained sane enough to 
hold aloof from all these absurdities, many even of them evince 
the fatal effect of the influence under which they have fallen 
by declaring all these superstitions to be after all "'open ques- 
tions,' which deserve, if they do not demand, a closer examination. 
Like the scientific superstition of all times, this modern one 
arrays itself in garments borrowed from real science. It deter- 
mines the credibility of clairvoyant somnambulists, or the 
occurrence of a telepathic miracle, by the rules of mathematical 
probability. It terms this whole field of hypnotic mysticism, 
and here again it follows in the footsteps of the spiritism that 
preceded it, 'experimental psychology.' It organises 'societies 
for psychical research,' which are devoted to the cult of hypnotic 
experimentation. The chief danger of all this, it seems to me, 
does not lie in the abuse of post-hypnotic suggestion for criminal 
purposes which may happen once in a while. Crimes have 
hardly as yet been committed by 'mediums' as a result of 
suggestion. Xo ! the great danger is, that persons of insufficient 
medical training, working not for therapeutic ends, but ' in the 
interests of science,' though there is absolutely no guarantee 
of the real existence of their scientific devotion, may exert an 
influence upon the mental and bodily life of their fellow-men 
such as, if continued for any length of time together, cannot 
fail to be injurious. 

It must, moreover, be plain to you all that there can be no 
question of an experimental psychological method, in the exact 
sense of those words, in this matter of hypnotising. The con- 
dition of hypnosis is such as absolutely to preclude the possi- 
bility of a psychological experiment in the real sense. The 
psychological experiment demands from its subject concentra- 
tion of the attention, practice, skilled introspection, in short, 

Theory of Hypnotism and Suggestion 3?" 

the fulfilment of all manner of conditions, which, if not alto- 
gether and normally out of the reach of the hypnotic subject, 
is at least wholly impossible during the course of the induced 
sleep. If we compose ourselves to sleep, with the intention of 
observing our dreams so far as that is possible, or even if we 
take morphine for the same purpose, we are not making an 
experiment, not doing anything that in execution or result is 
essentially different from simple observation. The conditions 
of dream-observation are not altered in the slightest degree by 
the fact that we have brought on sleep intentionally The cha- 
racteristics of the experimental method are variation and grada- 
tion of the phenomena, and elimination of certain conditions. 
Such a mode of procedure can be followed out in artificially 
induced sleep as little or, let us say, as imperfectly as in 
natural sleep : we shall gain no more by investigating the former 
than by collecting casual observations of normal dreams. And 
all this holds in still greater measure of hypnotism, since just 
in the cases which present the most interesting phenomena 
there is a total absence of any subsequent recollection. We 
can only infer what goes on in the mind of the somnambulist 
from his words and actions ; if we wish to subject him to special 
influences, we are hampered by the same conditions as hinder 
the investigation of sleep and dreams. 


Any explanation of the phenomena of hypnosis must, ob- 
viously, begin with the cognate facts of our normal mental life. 
We have not to explain the usual functions of consciousness in 
terms of hypnotism, but conversely. The established facts of 
normal life, and especially those which best admit of introspective 
control, must be employed to throw light on these phenomena, 
which, if not pathological, are at least due to unusual conditions 
Now there is one state which you will see at once evinces a very 
great similarity to the hypnotic sleep, heaviness after waking. 
In this condition we may perform actions, obey orders, answer 
questions ; but when we have fully awakened we realise that 
all this was done half automatically, without any assistance from 
the will. In other words, there may be developed in ' automatic 

33$ Lectures on Human ana Animal Psychology 

reaction to command/ very like that of the hypnotic subject. 
The sleepy soldier acts automatically at the word of command ; 
the half-roused servant executes a commission ; but it is only 
when fully awake that either remarks what he has done: indeed, 
if the sleepiness continues, he may entirely forget it. Years 
ago, when I myself went through several such experiences, I 
noticed particularly that I felt myself entirely at the mercy of 
external impressions, and acted, under their influence and that 
of the first dreamlike associations that they aroused, without 
any consideration, and so without any consciousness of the 
occasional wrongness of what I was doing. This self-surrender 
to external impressions approximates the dream-state ; while 
the capacity of action and the general absence of hallucinations 
and illusions suggest, on the contrary, the waking condition of 
consciousness. But this exemption is not altogether constant. 
Illusions in particular frequently occur in the state of drowsiness. 

Suppose now that the condition just described is brought 
one stage nearer that of sleep, suppose that the self-surrender 
to external impressions results in exclusive conscious control by 
definite ideas and feelings suggested b}^ the operator, and you 
have the hypnotic sleep. One of the chief effects of suggestion 
is to increase the duration of this state. The subject, who fell 
asleep at the word of command or under the influence of ideas 
equivalent to it, remains dominated by the idea that he cannot 
wake except by a new command. So that the most marked 
characteristics of hypnosis are loss of volitional initiative, re- 
striction of the receptivity of consciousness to external impressions 
and the first associations that they call up. and usuaHv^ a deter- 
minate direction_of_the_attention induced .by_the_ influence of 
theoerator'-s.- suggestion,. Its effects are intensified by the 
tendency, to haJlucinationj which, combined with the absolute 
surrender to external irn.pres.sions, results in a transformation of 
the sugStkias.-Qf. the hyfinQtic'<on^iQUgnejsjnlQ_real objects. 

You see, then, that the hypnotic sleep is akin tonormal_slee_ 
and dreaming, occupying an^mtermediate position between these 
and drowsiness, but characterised by the surrender of our will 
to that of another person, and by the consequent efficacy of 
suggestion. Especially to be noticed is the inhibition of volun- 
tary activity. This is not only one of the chief diagnostic 

Theory of Hypnotism and Suggestion 339 

symptoms, but an important condition of the origin of the other 
phenomena. You must not, however, imagine that will-power 
is altogether in abeyance Hypnotic actions are always volun- 
tary actions in the wider sense of the word. But they are not 
self-initiated, resulting from a consideration of motives and a 
decision of the agent's own mind ; they are impulsive, univocally 
determined by a suggested idea and by the associations which 
it directly excites. 

Regarded from the point of view of will and voluntary action, 
then, the hypnotic sleep is an abnormal condition. But it is a 
condition which follows all the psychological laws of our waking 
life. And the same is true of suggestion, that other so significant 
factor in the origin and progress of hypnosis. Ideas are con- 
tinually aroused in our minds by the words which we hear and 
the actions which we see. Word and act are intimately associ- 
ated with ideas, and affect our mind and will uith all the greater 
force the smaller the number of conflicting associations or 
inhibitory intellectual motives that opposes them. Looked at in 
this way, suggestion reduces itself simply to an external impres- 
sion, followed by associations characterised less by definite and 
positive attributes than by the negative quality of the absence 
of inhibitory influences. This impression and the consequent 
ideas will, therefore, continue to be the exclusive determinants 
of volition until other suggestions (i.e., other similarly excited 
associations) give a contrary direction to the hypnotic conscious- 
ness. How these facts enable us to explain certain special 
forms of suggestion, autosuggestion, post-hypnotic suggestive 
influences, etc., I attempted to show in describing those phe- 
nomena themselves. 1 

1 For the physiological substrate of the hypnotic condition, and for the 
discussion of many other points which can only be briefly touched on here. 
cf. my detailed account of hypnotism in the Philosophised Studicn, vol. 
viii., pp. I ff. 





IN the preceding lectures we have considered the associative and 
intellectual processes of consciousness, first in their genera! 
and normal features and then under the various aspects which 
they present in mental disturbance, dreaming, and certain con- 
ditions related to that of sleep and dreaming. There now 
remains one last question, the answer to which is important if 
we are to understand the nature of these processes and their 
relation to the other functions of the mind, the question of 
animal intelligence, or, to express it more exactly, of the nature 
and significance of those animal actions the conditions of whose 
origin lead us to refer them to mental processes similar to our 
own associations, and possibly even to our own processes of 
judgment and inference. 

The study of animal psychology may be approached from 
two different points of view. We may set out from the notion 
of a kind of comparative physiology of mind, a universal 
history of the development of mental life in the organic world. 
Then the observation of animals is the more important matter ; 
man is only considered as one, though, of course, the highest, of 
the developmental stages to be examined. Or we may make 
human psychology the principal object of investigation. Then 
the expressions of mental life in animals will be taken into 
account only so far as they throw light upon the evolution 
of consciousness in man. You will remember that we decided 
at the outset of these lectures to deal with animal psychology in 
this second sense, and for the more limited purpose. 

Problems of Animal Psychology 341 

If we compare these two ways of treating psychology with 
comparative and human physiology, we cannot fail to see that 
the two spheres of investigation are very different as regards 
methods and appliances. The bodily organs and functions of 
animals are just as accessible to objective examination as are 
those of man . indeed, in the living condition for obvious reasons 
far more completely so. So that there is no need to remind 
human physiology that it must never omit the comparative 
reference to animals. It follows this rule of its own accord, 
oftentimes more than it cares to, or than might be wished in the 
interests of physiological theory, because it must necessarily 
have recourse to animals where observation on man is impos- 
sible. Human psychology, on the contrary, may confine itself 
altogether to man, and generally has done so to far too great an 
extent. There are plenty of psychological text-books from 
which you would hardly gather that there was any other con- 
scious life than the human. It is entirely different with com- 
parative physiology and comparative psychology. It would be 
possible, if needs were, to write a monograph upon the physiology 
of an order or a species, say the infusoria or the frog, without 
paying any regard to the analogous functions in man. But not 
the least advance can be made, either in the psychology of a 
particular animal or in that of the animal kingdom, without 
starting out from the facts of the human consciousness. And, 
here again it is psychology that has been at fault. Following 
the analogy of comparative anatomy or physiology, psychologists 
have attempted without more ado to schematise the evolution 
ot mental life in animals, and then to apply their results directly 
to that of man. The outcome could be nothing else than that 
which always follows from the premature use in one connection 
of concepts found in another. So that Bacon's comparison of the 
insufficient observation of nature by the Aristotelians of his day 
to the report of an ambassador who based his knowledge of the 
measures of a Government upon town gossip, and not upon 
accurate examination, applies fairly enough to the animal 
psychology of our own time. It is permeated through and 
through by the concepts of the every-day psychology, which is 
thought to suffice for the requirements of ordinary life, and too 
often also for the sciences which cannot do without psychological 

34 2 Lectures on Hitman and, Annual Psychology 

reference. The one great defect of this popular psychology is, 
that it does not take mental processes for what they show them- 
selves to be to a direct and unprejudiced view, but imports into 
them the reflections of the observer about them. The necessary 
consequence for animal psychology is, that the mental actions of 
animals, from the lowest to the highest, are interpreted as acts 
of the understanding. If any vital manifestation of the organism 
is capable of possible derivation from a series of reflections and 
inferences, that is taken as sufficient proof that these reflections 
and inferences actually led up to it. And, indeed, in the 
absence of a careful analysis of our subjective perceptions, we 
can hardly avoid this conclusion. Logical reflection is the 
mental process most familiar to us, because we discover its 
presence whenever we think about any object whatsoever. So that 
for popular psychology mental life in general is dissolved in the 
medium of logical reflection. The question whether there are 
not perhaps other mental processes of a simpler nature is not 
asked at all, for the one reason that, whenever self-observation is 
required, it discovers this reflective process in the human con- 
sciousness. The same idea is applied to feelings, impulses, and 
voluntary actions, which are regarded, if not as acts of intelli- 
gence, still as affective states which belong to the intellectual 

This mistake, then, springs from ignorance of exact psycho- 
logical methods. It is, unfortunately, often rendered worse by 
the inclination of animal psychologists to see the intellectual 
achievements of animals in the most brilliant light. This, of 
course, is due to the natural pleasure which the objects of our 
observation always give us, and which is the most effective spur 
to continuous devotion to a particular subject. In the present 
case it is transformed into the unintentional endeavour to 
observe things which shall be as interesting as possible. Un- 
bridled by scientific criticism, the imagination of the observer 
ascribes the phenomena in perfectly good faith to motives which 
are entirely of its own invention. The facts reported may be 
wholly true ; the interpretation of the psychologist, innocently 
woven in with his account of them, puts them from first to last 
in a totally wrong light. You will find a proof of this on nearly 
every page of the works on animal psychology. Take a few 

Problems oj Animal Psychology 343 

instances, chosen at random, from Romanes' industriously com- 
piled volume on Animal Intelligence. ' While we admire the 
diligence with which the author has observed and collected the 
observations of others, we cannot but notice the unfortunate 
absence of the critical attitude in a field where it is especially 
desirable. Turn to the chapter on ants. An English clergy- 
man writes apropos of the 'funereal habits' of these animals: 
' I have noticed in one of my formicaria a subterranean 
cemetery, where I have seen some ants burying their dead by 
placing earth above them. One ant was evidently much 
affected, and tried to exhume the bodies ; but the united exer- 
tions of the yellow sextons were more than sufficient to 
neutralise the effort of the disconsolate mourner ' (p. 92). I low 
much is fact, and how much imagination? It is a fact that the 
ants carry out of the nest, deposit near by, and cover up dead 
bodies, just as they do anything else that is in their way. They 
can then pass to and fro over them without hindrance. In the 
observed case they were evidently interrupted in this occupation 
by another ant, and resisted its interference. The cemetery, the 
sextons, the feelings of the disconsolate mourner, which impelled 
her to exhume the body of the departed, all this is a fiction of 
the sympathetic imagination of the observer. Another friend 
of the ants gives this account : ' .At one formicary half a dozen 
or more young queens were out at the same time. They would 
climb up a large pebble near the gate, face the wind, and assume 
a rampant posture. Several having ascended the stone at one 
time, there ensued a little playful passage-at-arms as to position. 
They nipped each other gently with the mandibles, and chased 
one another from favourite spots. They, however, never nipped 
the workers. These latter evidently kept a watch upon the 
sportive princesses, occasionally saluted them with their antrnn;c 
in the usual way, or touched them at the abdomen, but 
apparently allowed them full liberty of action ' (p. 88). The 
correctness of this observation need not be questioned. \\ hy 
should not a number of young queens have been crowded 
together upon a pebble, and some workers have been with them, 

1 Animal Intelligence, by G. J. Romanes, Int. Sci. Series, vol. xli., |th ed. 
'vS6). Cf. the same author's M,-i'.tiil E\','!:;t;on in Animals i8X;1. 

344 Lectures on Human and Animal Psychology 

and occasionally touched them with their antennae, as ants do 
everywhere ? But that they ' sported ' and played, that the 
others ' kept a watch upon them ' like chaperones, and now and 
again did homage to them by ' saluting,' that, again, is all due 
to the imagination of the observer. He would hardly have told 
the story as it stands had not zoology introduced the mislead- 
ingly suggestive term ' queens ' for the mature female insects. 
If the adults are ' queens,' the young females must, of course, be 
' princesses.' And since no princess ever went out without an 
attendant or a chaperone, the rest of the narrative follows as a 
matter of course. Written in jusi the same style is the follow- 
ing account of the education of ants, taken from the same work. 
It may serve at the same time as an instance of the more 
general remarks current in the literature of animal psychology. 
' The young ant,' the author tells us, ' does not appear to come 
into the world with a full instinctive knowledge of all its duties 
as a member of a social community. It is led about the nest, 
and " trained to a knowledge of domestic duties, especially in 
the case of the larvae." Later on the young ants are taught to 
distinguish between friends and foes ' (p. 59). These illustra- 
tions will enable you to judge how much of similar descriptions 
is actual fact, and how much is due to the imagination of the 

How easy it is to misinterpret an observation if the very 
greatest care is not taken in recording it. and if it is impossible 
to vary the circumstances by experiment, and so obtain accurate 
knowledge of its conditions, is well shown by the following facts. 
Pierre Huber, one of the most reliable students oi the habits of 
ants, stated that he had assured himself that an ant, if taken 
from the nest and returned after an interval of four months, was 
recognised by its former companions ; for they received it 
friendlily, while members of a different nest, even though they 
belonged to the same species, were driven away. Huber regards 
this as evidence of the extraordinary accuracy of memory in these 
insects. Now the correctness of his observation cannot be 
doubted ; and, besides, it has been confirmed by another 
experienced investigator, Sir John Lubbock At first sight, 
therefore, the conclusion seems perfectly justifiable. But if a 
single individual were really recognised alter so long an interval, 

Methodological Rules 345 

think what the ger.eral mental capacity of the ants must be ! 
Fortunately, Lubbock made the matter a subject of experiment. 
He took ant larvae from the nest, and did not put them back 
again till they v/ere fully developed. The result was, that they 
too were quite friendlily received. Plainly, then, there can be no 
question of ?.n act of individual recognition. There must be 
some characteristic peculiar to all the members of a particular 
nest, possibly a specific odour, which determines the in- 
stinctive expression of ' friendship.' 

It is the same with those facts upon which the conclusion has 
been based that ants and other insects that live together in 
' states/ as they are called, possess a fairly well-developed lan- 
guage. The animals are undoubtedly able to call in the assist- 
ance of others for the accomplishing of tasks too great for their 
ow/. strength. But this purpose can be served by quite simple 
rr/jvements, which are common to very many species of social 
/ jimals. Since these movements are manifestations of impulse, 
hey exert a reflex influence upon the corresponding impulses 
of all individuals of the same kind. It is true that action must 
in every case be preceded by certain ideational connections. 
The ant that finds a load too heavy will connect this impression 
of weight with the often repeated perception of its mastery by 
united effort, and that again with the idea of assistance by other 
ants. But all this is a matter of very obvious association. To 
assume a supra-associational intellectual activity for processes 
which consist in manifestations of animal instinct of every-day 
occurrence, and repeated through countless generations, seems 
altogether unjustifiable. It would be entirely contrary to every- 
thing that introspection tells us of our own impulsive and 
automatic actions. 


These considerations lead up to a question which it is important 
to raise with regard to the study of animal psychology in 
general. We have no other means of estimating the mental 
processes of animals than in the light of those of our own con- 
sciousness. We must employ these in such a way as to gain 
the best and surest knowledge possible of the animal mind. 

346 Lectures o)i Human and Animal Psychology 

How is this 'to be done? The current animal psychology does 
not trouble itself to give any very careful answer ; indeed, a 
general answer is not given by it at all. In any particular case, 
however, as we have seen, it selects from the totality of mental 
processes the intellectual functions, and explains the mental life 
of animals in terms of them. Its implicit principle, that is, is 
precisely the opposite of the approved maxim of the exact 
natural sciences that we should always have recourse to the 
simplest explanation possible. It attempts to derive all the facts 
of its objective observation from the most complicated mental 
processes of human subjective experience. Where there are two 
alternatives, derivation from logical reflection and explanation 
in terms of simple association, it invariably chooses the former. 
And the fact that logical reflection is everywhere a possible 
explanation is taken as sufficient proof that it is the right one. 
But there are two reasons why this particular inference from 
effect to cause is not permissible. In the first place, mental 
activities are so complex and multifarious, that practically every 
objective action is capable of more than one interpretation ; 
and it needs very careful consideration of all the secondary cir- 
cumstances to decide the actual psychological conditions of a 
given result Secondly, logical reflection, just because every- 
thing can be subsumed to it subjectively, may itself be trans- 
lated into the objective condition of anything. So that the 
psychologist who interprets a fact of consciousness as a logical 
reflection, when it is not so given, is acting in principle as a 
student of natural science would who ascribed the properties 
of certain natural objects to the usefulness which they happened 
to possess for ourselves or for other organisms. But since 
logical reflection is itself a mental process, and may therefore 
take its place with other processes as a possible cause of some 
psychical effect, the danger of the error becomes much 
greater, and the proof of it much more difficult. All the more 
reason is there to emphasise the necessity, when we are attempt- 
ing a psychological explanation of phenomena which can only be 
inferred from their objective results, of always inquiring for 
those special characteristics which are of determining value for 
ne mode of explanation as compared with another. 

Acts oj Cognition and Recognition auiang .-Luuuals 347 


If, with this in mind, we ask : what are the most simple mani- 
festations of animal life which cannot be explained without the 
introduction of the mental factor ? we come first of all upon 
voluntary actions, referable to acts of cognition and recognition. 
You see at once that the very names employed to denote theso 
simplest ideational connections show traces of the fatal 
tendency to dissolve all the facts of consciousness in the medium 
of logical reflection. That the processes in question consist of 
nothing more than simple associations was shown when we 
\vere analysing the corresponding phenomena of the human 
consciousness. We used the word 'cognition ' because there was 
no other term current to denote the process ; but it is sufficiently 
plain from the language of our description that any idea of an 
act of logical cognition was very far from our minds. Now 
the analogous phenomena of the animal consciousness must, of 
cor-se, be treated from the same psychological standpoint. The 
acts of cognition and recognition are processes which represent 
the simplest modes of connection between present impressions 
and past ideas ; in other words, they are mental phenomena, 
which in a large number of cases furnish us with the only 
reliable evidence of the existence of a mental life. For to prove 
mentality we must be able to prove a persistence of the sense- 
impression, some form of memory, however elementary. 
Memory of some kind is presupposed in the cognition of an 
impression ; memory of a definite kind, in its recognition. In 
the former the impression is assimilated by numerous previous 
impressions of the same character ; in the latter it is associated 
with a single or with a strictly limited number of past im- 
pressions, the terms of the association frequently arranging 
themselves in a temporal series. 

In the very lowest animals it is only the simple association 
process that we term ' cognition ' of an object which plays any 
important part ; ' recognitions ' are hardly demonstrable. That 
the creature knows its proper food, and may be determined in 
its knowledge of it by previous impressions, is regarded as the 
first and primal indication of the presence of animal, i.e., mental 
life. But that an animal recognises a certain object, e.g., the 

348 Lectures on Human and Animal Psychology 

place where it finds its food, presupposes a separation and 
differentiation of ideas which in all probability is not realised at 
the lowest stages of organic existence. The simple acts of cog- 
nition develope from the primitive animal impulses which are 
at the root of the earliest manifestations of life that can be called 
psychophysically definite. The origin of the selection of food, 
for instance, can only be explained on the assumption that 
inherited organisation determines the correlation of pleasurably 
toned sensations with certain sense-impressions, and that these 
sensations are connected with the movements subserving nutri- 
tion. The fact, often confirmed by experiment, that even the 
protozoa congregate in light of one quality and avoid spots 
illuminated by that of another, must depend upon some such 
original sensation-character. The discrimination is made in 
terms of sensation and the affective tone attaching to it, as in all 
the higher animals. Ants try to escape from a violet light, 
but crowd together on a blue surface. Lizards and blind-worms 
avoid blue and all the more refrangible colours, but are fond of 
red. Again, there may be connected with these instinctive 
manifestations of mental life others which indicate an intensifi- 
cation of the original affective distinctions by previous impres- 
sions. The more frequently experiments are repeated with the 
different Kiumination of different portions of space, the more 
quickly and certainly does the animal decide for its favourite 
colour. And the fact that all animals prefer colours and bright- 
nesses corresponding to those of the medium in which they live, 
the winged insects, e.g., blue or white, may be referred to the 
influence of previous impressions. At the same time it is not 
improbable that these psychophysical influences have in the 
course of generations modified the sensitive elements of the 
sense-organs ; so that the affective quality which determines the 
preference or avoidance of certain stimuli is so far connate that 
it connects at once with the sense-impressions. Thus the exci- 
tability of the eye of the owl and eagle is entirely different from 
the very first. Nevertheless, there is every reason to suppose 
that this difference has been developed in the course of generic 
evolution, simultaneously with the differentiation in mode of life 
and habit. And the relation of animals to their proper food- 
supply is an altogether analogous case. The actions prompted 

Acts of Cognition and Recognition among Animals 349 

by the nutritive impulse will be first excited by the affective 
colouring of certain definite sensations. But when the enjoy- 
ment of food has been once experienced, the new and the old 
impression become associated. We therefore find that in all 
animals the discovery and right discrimination of food-stuffs 
are perfected in the experience of the individual life This 
'experience,' psychologically analysed, consists wholly in simple 
acts of ' cognition,' i.e., in the assimilative effect which previous 
impressions have upon new ones of the same character. 

It is generally more difficult to say whether an act of 
recognition has taken place in a given instance. In recognition 
a perception is apprehended as agreeing either with a previous 
perception referring to some individual object, or with a strictly 
limited group of such previous perceptions. The process ob- 
viously presupposes a more highly developed discrimination of 
ideas, i.e., a greater store of determinate associations. Often 
enough actions of the lower animals which are interpreted as 
recognitions belong to the sphere of indefinite cognitions. A 
good instance of this is the ant experiment quoted above. The 
insect returned to its nest by Pierre Huber, after a separation 
of four months, was really known not as an individual, but in 
virtue of some generic or family characteristic. This does not, 
of course, mean that individual recognition is not possible 
among the same insects, though it would probably only extend 
over a much shorter time. The ant which brings up its com- 
panions to assist it in carrying a load must, naturally, recognise 
the object to be carried and the road it has passed over, 
possibly also some of the helping ants. It is well known that 
in the higher animals the memory is able to retain particular 
facts and objects for a comparatively long period. The dog 
recognises its master after an absence of months ; the domestic 
cat is thoroughly familiar with the rooms of the house in which 
it is kept, and at once makes itself at home after quite a long 
absence. The dog, the elephant, and many other animals have 
an accurate memory for any previously unknown individual 
who teases or strikes them, and show it by very evident tokens 
of displeasure. When it is assisted by instinct and specific 
sense-development, this faculty of recognition among the 
animals may far exceed that of man in its own particular 

350 Lectures on Hum.m and Animal Psychology 

sphere. The dog can find its way over a long and complicated 
road by means of smell ; the carrier pigeon recognises the 
direction in which it flew some long time previously, thanks to 
its trained .space-memory and far-reaching vision. All these 
phenomena, which have sometimes been referred to a mysterious 
' sense of locality,' are explicable as dependent upon recollec- 
tion, highly developed in one particular direction and assisted 
by sense-peculiarities. The carrier pigeon taken from its home 
in a closed cage to some unknown and remote spot will either 
not return at all, or only by a long and roundabout way. 


Wherever recognition is possible, other associations may 
also be observed. The animal, like ourselves, naturally 
associates the similar and the contiguous in time and space. 
And we frequently find in the lower animals what we have 
observed in man, that associations give rise to actions whose 
result is equivalent to that due to the operation of the in- 
tellectual functions. It is here, therefore, in the various forms 
of successive association proper, that the act of interpretation 
which resolves the mental life of animals into concepts, judg- 
ments, and inferences, according to all the rules of logic, finds 
freest play. But if the whole body of reliable observation is 
carefully tested, and due regard paid to the lex parsimonies, 
which only allows recourse to be had to complex principles of 
explanation when the simpler ones have proved inadequate, it 
seems that the entire intellectual life of animals can be ac- 
counted for on the simple laws of association. Nowhere do we 
find the characteristic marks of a true reflection, of any active 
functioning of imagination or understanding. In saying this, 
we are, of course, regarding only well-authenticated facts, and 
not those ' travellers' tales ' of which animal psychology has as 
many as it has wrong explanations of actual observation. You 
may remember the story in Pliny's Natural History of the ele- 
phant who was punished during a performance for his bad 
dancing, and who secretly set to work to practise in the 
night, so as to do better the next time. We may be pretty 
confident in supposing that this tale and many others like it, 

Association among the Lower Annuals 351 

which are still current in the animal psychologies, are either 
pleasant inventions or, if they contain a grain of real observa- 
tion, have received so much imaginative embellishment, that its 
discovery is practically impossible. 

As a matter of fact, the mental life of animals is rich enough 
to be able to do without all this. The life of the more highly- 
organised invertebrates, even apart from the phenomena con- 
nected with the social instincts, to which we shall return later, 
has many features which would astonish an observer untrained 
in psychological analysis. Instead of giving you a whole 
number of illustrations which are all variations of the same 
theme, I will narrate an observation of my own. I remember 
it very distinctly, because it was the first time that I had 
brought home to me the mental life of a lower animal. I had 
made myself, as a boy, a fly-trap, like a pigeon-cote. The flies 
were attracted by scattered sugar, and caught as soon as they 
had entered the cage. Behind the trap was a second box, 
separated from it by a sliding-door, which could be opened or 
shut at pleasure. In this I had put a large garden spider. Cage 
and box were provided with glass windows on the top, so that 
I could quite well observe anything that was going on inside. 
At first nothing particular happened. When some flies had 
been caught, and the slide was drawn out, the spider, of course, 
rushed upon her prey and devoured them, leaving only the legs, 
head, and wings. That went on for some time. The spider 
was sometimes let into the cage, sometimes confined to her own 
box. But one day I made a notable discovery. During an 
absence the slide had been accidentally left open for some 
little while. When I came to shut it, I found that there was 
an unusual resistance. As I looked more closely, I saw that 
the spider had drawn a large number of thick threads directly 
under the lifted door, and that these were preventing my closing 
it, as though they had been so many cords tied across it. 

What was going on in the spider's mind oefore she took this 
step towards self-preservation a step, mark you, which but for 
the vis major of the boy-master would have been perfectly 
adequate to effect the desired result ? The animal psychologist 
will possibly say : ' the spider must first of all have come to 
understand the mechanism of the sliding-door, and must have 

352 Lectures on Human and Animal Psychology 

said to herself that a force operating in a definite direction cculd 
be compensated by another in the opposite direction. Then 
she 'set to work, relying upon the perfectly correct inference 
that if she could only make movement of the door impossible s 
she would always have access to the victims of her murderous 
desires. There you have a consideration of general issues, an 
accurate prevision, and a cautious balancing of cause and effect, 
end and means.' Well, I am rather inclined to explain the 
matter otherwise. I imagine that as the days went by there 
had been formed in the mind of the spider a determinate asso- 
ciation on the one hand between free entry into the cage and 
the pleasurable feeling attending satisfaction of the nutritive 
instinct, and on the other between the closed slide and the 
unpleasant feeling of hunger and inhibited impulse. Now in 
her free life the spider had always employed her web in the 
service of the nutritive impulse. Associations had therefore 
grown up between the definite positions of her web and definite 
peculiarities of the objects to which it was attached- is well as 
changes which it produced in the positions of certain of these 
objects, leaves, small twigs, etc. The impression of the falling 
slide, that is, called up by association the idea of other objects 
similarly moved which had been held in their places by threads 
properly spun ; and finally there were connected with this 
association the other two of pleasure and raising, unpleasantness 
and closing, of the door. That was surely enough to rouse the 
prisoner to action. Any other intellectual or inventive activity 
is entirely unnecessary. If she had not had these associations 
at her disposal, she would certainly never have hit upon the 
plan she did. 




THERE seems at first sight to be a very considerable differ- 
ence between the expressions of mentality in the higher 
animals, more especially the more highly developed do- 
mestic animals, and the simple or complex associations which 
we ascribed in the previous lecture to certain of the invertebrates, 
such as spiders, ants, and other insects with very definite instincts. 
The perpetual intercourse of the domestic animals with man 
seems to bring them quite near to him on the mental side ; he 
exerts a determining influence upon the contents of their ideas, 
the direction of their associations, and their whole affective life. 
The dog shares the joys and sorrows of its master ; it reads 
anger, happiness, or despondency in his face. The trained 
poodle is made as happy as can be by its master's praises, and 
shows every sign of pride when entrusted with certain duties, as 
the carrying of a basket or a walking-stick. Now it is true 
enough that all this points to a great diversity of feeling and a 
considerable adaptability to the feelings of others. But the 
emotions expressed never belong to the sphere of intellectual 
feeling, logical, aesthetical, etc. So that the only certain 
inference from the actions in which the animal appears to 
resemble man so closely is that it is endowed with a very active 
associational mechanism. 

There can be no doubt that the behaviour of our more highly 
developed domestic animals indicates this activity of association. 
As soon as you have taken your hat and stick, your dog knows 
that you are going out, and shows by tokens of pleasure and 

353 A A 

354 Lectures on Hitman and Animal Psychology 

other unmistakable gestures that it wishes to go with you. A 
poodle of my own used to be subjected to a thorough washing 
every Saturday, and disliked it very much. Various prepara- 
tions that \vere going on in the house would remind him that 
Saturday had come ; and he not infrequently disappeared early 
in the morning, and only returned late in the evening, when all 
fear of the cold water was over. On these occasions he usually 
spent the day on the square in front of the house, looking 
longingly up at the windows and obviously divided between the 
wish to return, and aversion to the fate that awaited him. He 
was all the happier when Sunday morning arrived. On that 
day my brother, who was living in a neighbouring town, was 
accustomed to come over and see me. The dog was more 
attached to him than to any of the inmates of the house. He 
never failed to keep watch behind the front door from the very 
first thing in the morning, welcoming approaching footsteps 
with a hopeful wag of his tail, and dropping his head de- 
spondently if they disappointed him. At last, when my brother 
really appeared, he was greeted with every manifestation of the 
most extreme joy. Experiences of this kind not only show that 
the mechanical operation of association may extend over a 
considerable time, as is proved by the recognition of an indi- 
vidual' after a number of days, but also, and it is this which 
distinguishes the present instance from a case of simple recog- 
nition, that an animal is itself able to associate events which 
coincide in space or time, and to extend these associations over 
a relatively long period. The poodle knew, of course, that it 
was Saturday simply by the special preparations for cleaning 
the house. With that idea was inseparably associated the very 
unpleasant idea of his own washing. This association was not 
only strong enough to keep him away from the house for the 
whole day, but was further connected with the other and far 
more complicated association of the following day and the 
arrival of my brother. Of course, the regularity of the visits 
furthered the formation of the association. At the same time, 
we have here a development of temporal ideation reaching 
far beyond the connection of directly simultaneous or successive 
events It would be utterly wrong, though quite in the manner 
of current animal psychology, to make the dog reason in this 

Mentality of the Higher Animals 355 

way : ' Yesterday was house-cleaning and washing-day ; my 
friend usually arrives on the day after that ; therefore he will 
come to-day.' The simpler, and therefore the only justifiable, 
explanation is, that the experience of previous weeks had made 
the succession of these events a stable contiguity-association in 
the animal's mind ; and that its expectation of my brother's 
arrival after the preceding occurrences was of the same kind as 
its expectation that it would get something to eat after the 
filling of its platter. The only difference between the two cases 
is that the former association extended over a longer time and 
embraced a greater number of events than the latter. 

A whole host of observations, which are usually interpreted in 
terms of intelligent action in the strict sense of the word, are 
more or less like these examples of my own. I will only cite 
the following, which I take from Romanes' book on Animal 
Intelligence (p. 418). The servants of a house had been accus- 
tomed during a frost to throw the crumbs remaining from the 
breakfast-table to the birds. A cat took advantage of this, and 
now and again obtained a hearty meal from one or two of 
the assembled guests. The practice of feeding the birds was 
therefore discontinued. The cat, however, scattered crumbs on 
the grass itself with the obvious intention of enticing them. 
Romanes supplements this by another story, in which crumbs 
were sprinkled on the garden-walk, and the cat lay in ambush 
to catch the sparrows. In this case, however, the cat used to 
conceal itself from the walk in a neighbouring shrubbery, and 
there await the coming of the birds. But the sparrows showed 
themselves more wide-awake than the cat : they waited on a 
wall, never venturing to fly down till their enemy wearied and 
went away. Romanes credits the two cats with the same form 
of inference. In the latter case the animal reasoned : 'crumbs 
attract birds ; therefore I will wait for birds when crumbs are 
scattered.' The first cat proceeded a stage further : ' therefore 
I will scatter crumbs to attract birds.' There can be no doubt 
that the two cases are analogous, and that the simpler is alto- 
gether a matter of ordinary association of the sort which is 
constantly directing animal action. But there can be no doubt 
either that the cat did not make the author's inference. When 
once the association between scattering crumbs and attracting 

356 Lectures on Human and Animal Psychology 

birds had been formed, the feeling of expectation was sufficient 
to cause it to lie in wait for prey as it was accustomed to do in 
other cases. This association was precisely like that which the 
sparrows had formed, even in its effect upon the will ; except 
that the cat was led to undertake a particular action, and the 
sparrows to refrain from it. We must admit that the case is 
somewhat different when the cat itself scatters crumbs. The 
tale is so unlikely, in view of what we know of the general 
behaviour of the animal, that we may possibly refer it to self- 
deception on the part of the observer, or conjecture that some 
chance incident has been wrongly interpreted. The same story 
told of monkeys would sound more probable, on account of the 
high development of the imitative impulse in those animals. 
There is nothing to take the place of this in the cat except the 
carnivorous instinct to secure prey. But even if we grant the 
correctness of the observation, the action might be explained 
by associative processes. The stable association of scattering 
crumbs and attracting birds, taken together with the impulse to 
catch them, might have enabled the cat to supply one term of 
the association-series for itself. But if we leave out this element 
of spontaneous activity connected with the specific carnivorous 
instinct, we see that the range of association in the cases cited 
is not nearly so great as in the above-quoted recollection and 
recognition by my dog. 

That the action of animals may be determined by memorial 
ideas, as well as by the corresponding sense-impressions, can be 
very readily shown. I often made the following amusing ex- 
periment with my own poodle. I had taught him to spring over 
a stick which I held out at the word ' Jump!' One day I called 
the word out to him without presenting the stick. At first he 
looked at me in surprise, and then, as I repeated the command, 
barked impatiently. At last, after I had given the order several 
times with a stern face, he decided to make a spring into the 
air, but barked loudly at me afterwards, as though to complain 
of the absolute absurdity of my command. When I had re- 
peated the experiment a number of times, the animal came to 
respond at once by springing into the air, but never failed to 
protest by growling and barking. The word of command 
aroused the memorial idea, and this was sufficient to arouse the 

Animal Play 357 

action produced by the actual presentation of the stick ; while 
the feeling of contrast between idea and object, and of the pur- 
poselessness of the act gave rise to unpleasurable emotions 
conflicting with the dog's habitual obedience. 

The criterion of 'intelligent' associative action and of in- 
telligent action proper can only be this, that the effect of 
association does not go beyond the connection of particular 
ideas, whether directly excited by sense-impressions or only 
reproduced by them ; while intellectual activity in the narrower 
sense of the word presupposes a demonstrable formation of 
concepts, judgments, and inferences, or an activity of the 
constructive imagination. If Pliny's story of the elephant 
practising dancing by moonlight were true, it would undoubtedly 
be a case of imaginative activity in this sense. On the other 
hand, the jumping of the dog over an imaginary stick at the 
word of command implies memory, but not imagination; i.e., it 
depends not upon the spontaneous arousal of associations, but 
upon their discharge by external impressions. 

The play of man and the animals differs in the same way as 
their ' intelligence.' We regard certain actions of the higher 
animals as playful when they take the form of imitations of 
purposive voluntary actions. We know that they are imitations 
because the end pursued is only a fictitious end, the real end 
being excitation of joyous emotions similar to those which 
follow as secondary effects from genuine purposive action. 
That means, you see, that the play of animals is for all practical 
purposes identical with play among mankind. Our own play, 
at least in its simpler forms, e.g., in the play of children, is 
merely an imitation of the actions of every-day life stripped of 
its original purpose, and resulting in pleasurable emotion. The 
play of animals bears the same relation to the play of man as 
animal life does to human life. The impossibility of transcend- 
ing a certain circle of ready-made associations characterises 
the play of animals (even of the most highly developed), as it 
characterises their mental life in general. Over against the 
countless varieties of the play of children, reflecting all con- 

358 Lectures on Human and Animal Psychology 

ceivable relations of life, stands the single form of mock fighting 
among the animals. (Trained animals do not, of course, concern 
us ; their performances are not real play.) Dogs, cats, and 
monkeys, even when they are playing with their young, show 
their affection by pretending to fight with them. And though 
it is true that play is an indication of high mental development, 
and brings the animal nearer to ourselves than any other activity, 
it is rather the fact that it plays than the nature of the play 
itself which is the important point. Only those animals can 
play which reproduce in memory pleasurable experiences, and 
are able to modify them so that only their pleasant side comes 
to consciousness, and their unpleasant aspect disappears. At 
the same time, any comparatively complex associative and 
affective activity in the mind of an animal is a sufficient reason 
for the origination of playful actions. Animal play never shows 
any inventiveness, any regular and orderly working out of some 
general idea. And only where this is found can play be re- 
garded as the expression of really imaginative activity. The 
simple and original play of the animals is, if the expression may 
be allowed, a purely associational play. When a dog sees 
another dog, it does not necessarily feel any hostility towards 
it, but still has an inclination to exercise its strength in a mock 
fight, and so to gain the pleasure which it has experienced in 
real ones. If it obeys this impulse while its mood is friendly, 
or at least not hostile, the result is a mutual trial of strength in 
a playful contest ; ending often enough, as you have probably 
observed in dogs and monkeys, in the arousal of the real joy 
of battle and a fight in bitter earnest. In many animals, how- 
ever, and especially in those which, like the dog, have been 
domesticated for a long time, the inherited instinct appears in 
a moderated form from the first ; the connate fighting impulse 
seems to become a connate impulse to play. 


We must conclude, then, that those animal actions which 
remind us most forcibly of imaginative activity do not show 
any of the specific characteristics which distinguish imagination 
from memory. There is no purposive and comprehensive con- 

Alleged Formations of Judgments and Concepts 359 

nection of ideas, none of the inventiveness which that implies. 
In the same way the animal actions which border most closely 
on the realm of human understanding give us no warrant for 
inferring the existence of true concepts, judgments, and in- 
ferences. That writers upon animal psychology have so often 
asserted the contrary is due to the interpretation of compara- 
tively simple associative processes as apperceptive intellectual 

Romanes gives an account of a series of tests of the intelli- 
gence of an elephant sent him by one of his correspondents (p. 
401). The story goes to show that 'elephants possess abstract 
ideas.' Even if we do not give the phrase ' abstract ideas ' its 
philosophical meaning, but simply understand by it general 
experiential concepts, we must still admit that the facts recorded 
do not prove their existence, but merely indicate a fairly 
highly developed associational activity. An elephant was 
made to lift various objects with its trunk, bundles of clothes ; 
tree-trunks ; heavy pieces of metal. It was noticed that the 
animal gradually ' took in a knowledge of the nature of the 
things it was required to lift ' ; light objects were thrown up 
easily and quickly, heavy ones slowly and with obvious muscular 
preparation, cutting instruments with a certain degree of caution. 
The observer concludes ' that elephants recognise such qualities 
as hardness, sharpness, and weight.' You will probably agree 
that nothing more is necessary for such actions than the for- 
mation of definite associations between the visual impressions 
of an object and its tactual properties. Such associations 
would in any case have been necessary to produce the observed 
result, even though the elephant really possessed the general 
concepts of light, heavy, cutting, etc. But if once the associa- 
tions were formed, they would be altogether sufficient to 
determine the ' intelligent ' lifting of such objects, and there 
would be no need at all of the further formation of general 

I spent a great deal of time in trying to discover some cer- 
tain indication in various actions of my own poodle of the 
presence or absence of general experiential concepts. I was 
never able to demonstrate them ; but I made many observations 
which rendered its nower to form them exceedingly improbable. 


60 Lectures on Hitman and Animal Psychology 

I taught the dog to close an open door in the usual way by 
pressing with the fore feet when the command ' Shut the door! ' 
was given. He learned the trick first of all on a particular door 
in my study. One day I wished him to repeat it on another 
door in the same room ; but he looked at me in astonishment 
and did nothing. It was with considerable trouble that I per- 
suaded him to repeat his trick under the altered circumstances. 
But after that he obeyed the word of command without hesi- 
tation at any other door which was at all like these two. It is 
obvious that when the animal had learned the trick of closing 
the door for the first time he had formed no general concept of 
4 door ' ; otherwise he would have found no difficulty in shutting 
others. His action plainly depended upon a quite individual 
association. Some force had to be used to form this, as is 
always the case with such performances. I stood the dog up 
on his hind legs against the open door, while I gave the word of 
command, until he responded to the order by voluntary move- 
ment. But you will say : might not the further step necessary 
to the formation of a concept have been taken when the 
animal learned to close the second door? After that he was 
able to close others for which he had not been specially trained. 
I do not believe that the assumption of the formation of a 
concept is necessary even here. It is evident that when the 
association between word of command, movement, and closing 
of the door had been formed at several doors the more special 
association between the particular door and the action of closure 
must have become obscured. The association of particular ideas 
had developed into a true similarity-association. There is not 
the slightest indication of the presence in the dog's mind of the 
principal characteristic of the formation of concepts, the con- 
sciousness that the particular object vicariously represents a 
whole category of objects. It had throughout only a very con- 
fused idea of the ' nature ' of a door. When I ordered it to 
shut a door which opened from the outside, it made just the 
same movement, opened the door, that is, instead of closing, 
and though I impatiently repeated the command, it could not 
be brought to do anything else, although it was obviously very 
unhappy at the ill-success of its efforts. Only after I had on 
several occasions let it out of the room, and made it shut the 

Alleged Formations of Judgments and Concepts 3^1 

door from outside, did the inability to close it from inside 
decide it to go out and repeat the attempt from the other side ; 
and then it would at once begin to scratch at the closed door 
to beg for readmission. 

What holds of concepts holds equally of the alleged judg- 
ments and inferences of our more intelligent domestic animals ; 
on closer consideration they may be all resolved into obvious 
associations, and they fail at the point where the sphere of ex- 
ternally suggested association touches that of intellection proper. 
I used to take my poodle on summer evenings into a garden 
lying across a stream, to which we had access by boat. The 
banks were very crowded, and boats were always plying to and 
fro. One day the poodle had stayed behind with some other 
dogs, and came to the bank too late, so that the boat was at 
some distance from the shore when he reached it. There was 
nothing left for him but to swim the stream ; and since it was 
fairly broad, and he had little love for water, that was very un- 
pleasant. Some days afterwards the same misadventure befell 
him. He ran up and down the bank whining dismally, and 
obviously very averse to repeating his bath. At that moment 
a boat-load of passengers pushed off; he sprang in and came 
dry-footed and extremely happy to the other bank. Hence- 
forth he not infrequently crossed in the same way. Now what 
was going on in his mind during this particular action? Did he 
think: ' my master has crossed the stream; this boat is also 
about to cross: therefore I shall overtake him if I step in ' ? Of 
course we car^ translate the animal's action into this chain of 
reasoning. Only we are not then dealing with the process in 
its mind, but with a logical interpretation of that process. But 
are not the interpretation and the process equivalent? And 
since the chain of reasoning would lead to the same result, how 
do we know that it did not actually take place in the dog's 
consciousness ? This instance is in truth a good illustration of 
the fact that the outcome of a logical reflection is not so com- 
pletely identical with that of a mere ideational association. 
The two processes present characteristic differences which ac- 
quire especial significance in cases like this, where we are deter- 
mining the nature of internal experiences. If the dog had sprung 
at once into a boat lying upon the shore and waiting to take 

362 Lectures on Human and Animal Psychology 

in passengers, we should still perhaps hardly have been justified 
in crediting it with logical reflection ; the association between 
the boats and their passage over the stream might have been 
formed in its mind through its frequent journeys across. Never- 
theless this association would have been considerably more 
complicated ; and its inclusion of a number of memorial 
elements, in the absence of the corresponding impressions, would 
have brought it very near the intellectual operations. But it 
never occurred to the animal to enter an empty boat and wait. 
Not till the boat pushed off did it associate this new crossing 
with previous journeys. So that the action bears every mark of 
a simple association. If on later occasions, and I am not sure 
about this, the dog did spring into an empty boat, or one just 
filling up, we need not refer this action to the sudden birth of 
logical reflection, but can explain it as a gradual extension of 
the associational series, resulting from practice. Many accounts 
of animal intelligence which are given without falsification or 
exaggeration in works on animal psychology would, we may be 
sure, oftentimes admit of a similar interpretation, if we were 
informed of all the stages in the animal's conduct. You could 
hardly imagine a better example of intelligence than the action 
of a dog which should cross a stream of his own accord with a 
number of passengers who were total strangers to him in order 
to visit a garden on the other side. And yet the act becomes 
simple enough when we can follow its gradual development in 


All these manifestations of animal 'intelligence' may be 
adequately explained as relatively simple associations. When- 
ever we are in a position to investigate the nature of the connec- 
tion of conscious processes involved, we look in vain for traces 
of logical reflection or real imaginative activity. We can now 
understand how it is that the animals lack one function which 
is characteristic of the intellectual processes, at the same time 
that it is their invariable concomitant, language. Animals can 
express emotion ; the more highly developed of them can to a 
limited extent give evidence of the connection of ideas with 
their emotions. But the expressive movements of animals never 

General Significance of Association 363. 

show that regular articulation, that reflection in organic struc- 
ture of the nature of intellectual ideation, which is characteristic 
of language proper. Animals possess certain elements of 
language, just as they possess certain elements of consciousness 
which might serve as the basis of intellectual function, but they 
do not possess language itself. So that the mere absence of 
this external mark would justify us in inferring the absence of 
those mental functions of which it is the mark. As a rule, it is 
not any physical obstacle, as is so often thought, which prevents 
animals from talking. In very many animals the development 
of the organs of speech has gone far enough to enable them to 
clothe thought in words, if the thought were there to clothe. 
The question why the animals do not talk is most correctly 
answered in the old way : because they have nothing to say. 
Only we must add that certain movements and sounds charac- 
teristic of feelings and ideas seem to be the forerunners of 
language ; and that animals give signs that in this connection, 
as in others, their mental life is the immediate precursor of our 

Indeed, the importance of association for the animal conscious- 
ness recalls what we have already said of its value for the human 
mind. When we began our consideration of the mental life of 
animals, we condemned the tendency of animal psychology to 
translate every manifestation of intelligence ' into an intellectual 
operation. The same reproach could be made against certain 
more or less popular views of our own mentality. The old 
metaphysical prejudice that man 'always thinks' has not yet 
entirely disappeared. I myself am inclined to hold that man 
really thinks very little and very seldom. Many an action which 
looks like a manifestation of intelligence most surely originates 
in association. Besides this, man is constantly translating acts 
of logical thought back again into customary associations, and 
so increasing the sphere and the intellectual consequences of the 
associational processes. By practice we can reduce anything to 
association. Trains of thought which at first involved consider- 
able intellectual labour are completed with increasing certainty 
and mechanical facility the oftener they are repeated. We can 
hardly overestimate the facilitation of logical operation and con- 
structive imagination which this tendency brings with it. All 

364 Lectures on Human and Animal Psychology 

the work spared by associative practice can be employed in new 
intellectual achievements. For this reason, thought proper is 
continually engaged with permanent mental associations at the 
same time that it is making new ideational connections. It is 
a process compounded of logical and associative connections of 
ideas. We may rightly give the name of thought to a train of 
ideas whose associations are manipulated by the attention for 
definite intellectual ends, and are only allowed to have weight 
within the bounds set by those ends. The best confirmation of 
this is given by the expression of thought in language. Though 
the general content of the expression is the result of an intellec- 
tual process, still the ready-made thought-forms furnished by 
association play no small part in the whole process. 

The fruitfulness of this interaction of association and intellec- 
tion helps us to understand how it is that many psychologists, 
so far from translating all mental activity into logical reflection, 
prefer, on the contrary, to regard thought and imagination as 
forms of association. We saw earlier that there are external 
and internal characteristics which clearly differentiate these 
processes. The associationist psychology can give no account 
of them. It ignores them, identifying imagination with memory 
and referring logical thought to logic as distinct from psycho- 
logy, as though the forms discovered by logic must not in the 
last resort rest upon psychological facts and laws. According 
to this school, the creations of imagination and intellect keep 
even pace with the activity of association. Dreaming and 
mental disturbance furnish, of course, a direct refutation of such 
a view; but the associationist theory makes light of that. It 
pays so little regard to the laws of ideation that it does not need 
to spend any time in considering exceptions. 


In spite of these essential differences between the intellectual 
processes and pure associations, there is an intimate correlation 
and mutual furtherance of the two in our own consciousness. 
That is possible simply because they are both stages in the 
same development The duty of association is to form those 
multifarious connections among the conscious elements which 

Man and the Animals 16 s 


enable us to comprehend a whole number of previous experiences 
into a resultant mental force, such as that which is employed in 
each separate act of voluntary attention, and on the basis of 
which the activity of association may be directed to the achieve- 
ment of definite intellectual ends. So that intelligence springs 
from association, and then turns round again to enrich it by new 
connections which will facilitate the employment of thought in 
the future. 

It is this relation between association and intelligence which 
must decide our answer to the final question which arises out 
of our investigation of the intellectual capacities of animals : are 
we to think that the gulf which separates the animals from man, 
which is on its mental side the gulf between association and 
intelligence, can ever be bridged ? 

In view of the facts of individual human development, we can 
hardly do anything else than reply with an unconditioned affir- 
mative. The boundary line between the purely associative pro- 
cesses which simulate intelligence and really intellectual actions 
may be crossed, because in actual fact it is crossed in the life- 
history of every one of us. From the stock of associations 
which we begin to lay in from our earliest childhood, there 
gradually developes the collective mental force of the individual 
personality, which manifests itself in self-consciousness, in active 
attention, and in voluntary control of ideation. And in this last 
especially we can clearly trace the influence of the increasing 
store of stable associations and the corresponding enrichment 
of affective and conative mentality which comes with it. 

But in its application to the relation of man to the animals 
our question falls into two special inquiries : is it probable 
that species or individuals of animals existing to-day will ever 
cross the boundary line ? and is it probable that man himself 
at some period of his development bridged the gulf which to-day 
separates him from the animals ? 

The first of these questions may be as certainly answered in 
the negative as the latter may in the affirmative. The step 
from association to intelligence proper is undoubtedly the 
longest ever taken in the course of mental evolution. When 
once the mind has reached the level of logical thought and con- 
structive imagination, it has before it that prospect of unlimited 

366 Lectures on Human and Animal Psychology 

advance which must inevitably at some point in time be realised 
in civilisation and history. That any species among the higher 
animals can make this tremendous progress is rendered alto- 
gether improbable by the general nature of their psycho- 
physical organisation. Besides which this organisation appears 
to be so far determinate that further modification cannot tran- 
scend very narrow limits. And perhaps the struggle for existence 
in the organic world would prevent any large number of quite 
differently constituted beings from rising by their own efforts to 
the height of civilisation and historical existence on a single 

It is otherwise with the second question. The laws of physical 
development put it beyond all doubt that man passed through 
lower forms of life before he reached his present stage of organi- 
sation. And the laws of mental development make the same 
conclusion at least exceedingly probable. Just as every one in 
the course of his own individual development takes the step 
from association to that intellectual conscious activity which 
arises from it, so must mankind at large have done at some 
point in the world's history. It was the first step from savagery 
to civilisation. And surely it is no disparagement of the value 
of mental development to regard it as having been from the 
beginning what we see it to-day, an evolution of mind from 
itself, proceeding under the conditions set by the environment 
in accordance with the universal laws of mental life. 




OUR original plan in these lectures was to begin with the 
examination of particular mental processes, sensations 
and ideas, feelings and voluntary actions, and then to turn our 
attention to the interconnection of all these processes in con- 
sciousness. But when we came to analyse complex states of 
mind, it seemed better to take their components separately and 
examine them as we examined the simpler phenomena of mind, 
although their isolation was no more than an abstraction. We 
have accordingly spoken in the preceding lectures only of the 
ideational side of consciousness. We could not leave feeling 
and will altogether out of account, because of their importance 
for the apperception and association of ideas But we said 
nothing of the relation of feeling and internal voluntary action 
to the other affective states of consciousness and to external 
voluntary action. We will now glance briefly at the more 
complex affective processes. 

The sense-substrate of all the more complex affective states 
we have found to be the common feeling. Like the particular 
sense-feelings which give rise to it, this is either a pleasurable 
or an unpleasurable feeling. Indefinite as these categories may 
seem, they are characteristic for the mental nature of affective 
activity in general. The physical excitations underlying the 
sensations to which affection attaches differ only in intensity and 
in the nature of the stimulus which arouses them. Sensations 
as such, therefore, present only intensive and qualitative differ- 
ences. But when the physical excitation also gives rise to an 
affective process, we find these two polar opposites of pleasure 

368 Lectures on Human and Animal Psychology 

and pain, the principal determinant of which is the intensity of 
stimulus. As we intensify any sense-impression, pleasurable 
feeling passes gradually into unpleasurable, and that into pain. 
While the intensity of the physical excitation increases con- 
tinuously, its affective concomitant changes from one quality 
into its exact opposite. 

At the same time the conditions of this qualitative change are 
really present in the physical excitation. The effect of stimula- 
tion upon the course of the physiological processes in the 
organism must be either favourable or inhibitory. Moderate 
stimulation is indispensable for the continuance of normal func- 
tion. Organs which are not used for a long time degenerate 
morphologically. Stimuli whose intensity is regulated by the 
normal capacity of the bodily organs themselves are therefore beneficial for the maintenance of life : they call for a uni- 
form expenditure of energy which is never pushed to the limit 
of exhaustion, but always leaves some small surplus of force in 
reserve. These normal internal stimuli also excite feelings of 
moderate intensity which unite to form the common feeling of 
comfort. When, on the other hand, the intensity of stimulus 
becomes so great that the organs are in danger of exhaustion 
or of entire destruction, there follows either a general disturbance 
or a temporary arrest of the vital mechanism. This excessive 
stimulation conditions not only a morphological and physio- 
logical derangement of the bodily organs, but also feelings of 
unpleasantness or pain. 

The reason for these differences in the affective character of 
the excitation, differences of pleasure and pain, is given, 
therefore, with the physical attributes of stimuli and the changes 
which they set up in the organs of sense. But the differences 
themselves are altogether mental in nature, manifesting them- 
selves simply as modifications of the general state of mind. 
Though originally produced by sense-stimuli, they contain no 
necessary reference to physical processes, so that we may have 
feelings which bear the same relation to the simple sense-feelings 
as does a train of logical thought to a particular sensation. 
Pleasure and pain mean alteration of our general psychical con- 
dition. They do not, therefore, require a sense-stimulus for 
their origination ; they may be called up by the mere idea of 

Connection of Affective States in Consciousness 369 

a sense-stimulus or by the intellectual content of a single idea 
or an ideational association. 

At the same time, the purely mental feelings always rest upon 
a sensory substrate ; just as on the intellectual side of mind the 
most abstract concept is still so far dependent upon sensation 
that it cannot be thought without translation into a vicarious 
sense -idea. 

This connection with sense can be demonstrated just as plainly 
for feeling as it can for idea. Feelings which attach to the most 
complicated ideational series have the same names as the simple 
organic feelings. ' Pain ' may signify either the excessive excita- 
tion of a sense-organ, or the higher degrees of general mental 
discomfort. We speak of the ' pain ' caused by the loss of a 
friend, or by the disappointment of our hopes, just as we do of 
the ' pain ' of a wound or of a diseased organ. We talk of 
' burning ' love, of ' pressing ' care, of ' gnawing ' remorse. In a 
word, our names for all kinds of affective states which have no 
direct connection with sense-impressions are derived either from 
peripheral sensations or from the feelings to which they give 

We may, of course, term this secondary meaning of the words 
purely figurative. We speak metaphorically of being ' torn ' 
by conflicting emotions. So 'agony' of mind, the ' weight ' of 
care, and similar expressions which are used both for bodily and 
mental conditions would be metaphors which we easily overlook 
because we are always using them and have no others at our 
disposal. Nevertheless there must be some psychological reason 
for this figurative characterisation of the affective consciousness ; 
there must be some sort of relation between the sense-feeling 
:"rom which the name is taken and the intellectual feeling to 
which it is applied. The most obvious relation would be a 
temporal association of the two processes. 

The question then arises whether mental pain is associated 
with bodily, and whether when pain ' presses ' and remorse 
' gnaws ' the sense-feeling of pressure or of gnawing pain is 
really present. 

B B 

370 Lectures on Human and Animal Psychology 

If we look closely at the intellectual feelings, especially in their 
more intensive stages, we can hardly doubt that they are in- 
variably accompanied by sense-feelings. These concomitant 
feelings often attain an intensity equal to that of the sense-feel- 
ings aroused by direct external or internal stimulation. Some- 
times they may be even localised with some degree of 
definiteness. They always evince a determinate quality which 
varies with the general affective condition, and which is reflected 
in the expressions which we employ to describe this. All ex- 
cessive feeling is attended by physical pain, whether diffused 
over the body generally, or restricted to a particular organ. 
Moderate excitations also affect the sense-feelings though less 
strongly, and are more definitely localised. References to this 
localisation of the sense-stimulus in particular affective states 
are found in ancient literature. Every passion was supposed to 
be seated in a particular organ ; and it must be admitted that 
where observation was wanting imagination took its place. 
Anger was placed in the liver, envy in the spleen, the higher 
emotions in the organs of the breast. Even to-day the heart is 
the seat of the most various affective states. Care and dis- 
appointed hope bring on heartache ; despair dies of a broken 
heart ; love through all its changes and chances has its source 
and centre in the heart, and ' faint heart never won fair lady.' 

There is really good reason for this relation of the heart's 
activity to the state of feeling ; for the cardiac nerves are those 
most easily excited by changes in our affective condition. 
Every affective excitation manifests itself in a weaker or 
stronger, quicker or slower, heart-beat. Joy and hope make the 
pulse quick and strong ; care and anxiety render it weak and 
slow ; terror arrests it altogether. And there are many indi- 
cations that other organs react to affective changes. It has 
often been noticed that violent anger results in a return of the 
bile to the blood, which means a derangement of the function of 
the liver. The lachrymal glands are very easily excited by the 
feeling of sorrow. And we should undoubtedly discover other 
similar interrelations were it not that they have no external 
symptoms. Besides the particular organ which is especially con- 

Emotion 3 7 r 

cerned in a particular affective state, there are always other organs 
more or less affected ; and it is the common feeling resulting 
from the sum total of these sensations that constitutes the sense- 
substrate of the total affective condition. The muscles, e.g., are 
almost always involved in this secondary excitation. We have a 
direct measure of the energy and tension, or of the exhaustion 
and relaxation of our muscles in sensation of movement ; and our 
general affective state is altogether different according as the 
limbs are motile and elastic or are heavy loads to weigh the 
body down. The feeling of the moment is here of the greatest 
importance. A feeling of joy and excitation makes movement 
easy and prompt ; a depressing feeling renders it slow and 

We must try to explain how it is that sense-excitations are 
always present in greater or less degree in affective processes. 
The view of the ancients that the excited organ is the direct 
seat of the feeling cannot, of course, be maintained. We know 
quite certainly that the parts of the body most closely related 
to mental activity are the central organs of the nervous system. 
It is here that the impulse must arise which has the sense- 
feeling as one part of its result. The symptoms observable in 
the peripheral organs simply indicate that this central excitation 
has a different seat at different times, and consequently 
produces different organic changes, now altering the pulse, 
now deranging the liver, now affecting the muscular system. 
In fact, we have before us a phenomenon which presents a 
certain analogy to reflex action : only that its source is not to 
be looked for in external stimulation, but in an activity of the 
central organs. 

These peripheral symptoms are of especial importance, as prov- 
ing that there is no complete separation of mental process and 
bodily function in the sphere of feeling any more than in that 
of ideation, but that the two are intimately interconnected. 


The affective processes which attach in consciousness to 
ideational connections are generally designated emotions. They 
are among the most important mental phenomena, exerting a 

37 2 Lectures on Human and Animal Psychology 

marked influence both upon ideation and voluntary action. They 
resemble feelings in that they are subjective processes not 
directly referred to external objects ; they differ from them in 
that they involve change m ideation and reactions in the organs 
of movement. Feelings, that is, are not accessible to external 
observation, or at least only become so when they pass over into 
emotions. Then they are reflected in certain expressive move- 
ments. These are further connected with reactions of the heart, 
the blood-vessels, the respiratory muscles, and certain secretory 
organs, which take on a special characteristic form in each par- 
ticular emotion. 

This twofold relation of emotion to feeling and ideation has 
led to a diversity of view as to its nature. It has been regarded 
both as an intensive feeling, and as a feeling originating from the 
train of ideas. Neither of these definitions does it full justice. 
The typical emotion has three stages : an initial feeling ; a 
subsequent change in the train of ideas, intensifying and quali- 
tatively modifying the initial feeling ; and (always supposing 
that the emotion is distinct and well defined) a final feeling, of 
greater or less duration, which may possibly give rise to a new 
emotion of which it forms the initial feeling. The principal 
difference between feeling and emotion, that is, consists in the 
second stage : the alteration in the train of ideas. The presence 
of this alteration enables us to divide emotions into tvvo classes, 
excitant and inhibitory. Instances of the former are joy and 
anger ; of the latter, terror and fear. At the same time all very 
intensive emotions are inhibitory in character, and it is only when 
they have run some part of their course that their excitant side 
comes to consciousness. On the physical side, the effect of 
emotion upon the train of ideas is accurately reflected in ex- 
ternal movement. The excitant emotion quickens ideation, and 
involves heightened mimetic and pantomimic movement, increase 
of cardiac activity, and dilatation of the blood-vessels ; the in- 
hibitory emotion paralyses, or at least relaxes, the muscles, slows 
the heart-beat, and contracts the vessels. All these physio- 
logical effects havetheir accompaniment of sense-feeling which 
intensifies the affective elements in the emotion. 

Less intensive degrees of emotion are called moods. It is a 
general rule that the duration of emotion varies inversely with 


its intensity ; so that moods are more permanent states of 
mind than emotions proper. Violent emotions are sometimes 
termed passions. The name indicates that strongly affective 
states, which oscillate between the feelings of pleasure and pain, 
tend invariably towards the side of the latter. ' Passion ' also 
implies that a particular emotion has become habitual. Hence 
the word is often used to denote a permanent condition which 
finds its expression in frequent outbursts of emotion. 

The most indefinite emotions are joy and sorrow. All the 
rest may be regarded as forms of one or the other of these two 
fundamental moods. When, e.g., sorrow is directed upon the 
external object which excites it, we call it care. We can only 
be careful about others ; and if we wish to express the fact that 
an object arouses no interest in us, we say that we do not ' care ' 
about it. The subjective opposite of care is melancholy. The 
melancholiac is centred in himself; he withdraws from the 
world to brood in solitude over his own pain. Care and 
melancholy become anxiety and dejection when they pass from 
emotions to permanent moods. Intermediate between these 
objective and subjective forms of sorrow, stand gloom and 
depression. We may be gloomy as to our fate in the world and 
depressed about a loss we have suffered, or we may be gloomy 
or depressed without any external reason simply because our 
mood will have it so. 

Joy, like sorrow, assumes different forms according to the 
direction which it takes. But we have not nearly so many 
words to express joyous emotion as we have to express sorrow. 
A joyous mood we call cheerfulness, or in its higher stages 
hilarity. But we cannot tabulate the joyous emotions as 
objective and subjective, as we could their opposites. It may 
be that our poverty of words points to a distinction in the facts 
of our affective life. The joyous emotions appear to be more 
uniform, less variously coloured, than the sorrowful. 

The emotions of joy and sorrow, whether their reference is 
mainly external or to the feeling subject, are always subjective 
in character ; the affective excitation of our own minds is al- 
ways the principal thing. A mood, on the other hand, may be 
objectified by our putting our own feelings into the external 
objects which excite them. If joy and sorrow are the expressions 

374 Lectures on Human and Animal Psychology 

of an internal harmony and disharmony, these objective emotions 
are the result of some external harmonious or inharmonious 
impression. Like and dislike are the most general forms of 
objective emotion, corresponding to joy and sorrow on the 
subjective side. They further imply a movement to or from the 
object ; what we like attracts us : what we dislike repels us. And 
this movement finds its expression in the various particular 
forms in which the general emotions occur. The attraction 
which a pleasing object has for us we call charm ; a thing is 
' charming ' which both pleases and attracts us. The opposite of 
charm is repulsion, a violent dislike which makes us turn away 
from an object in displeasure. Repulsion becomes aversion, and 
at a still higher stage anger, when it is turned directly upon the 
repellent object ; it becomes chagrin and mortification if the 
unpleasant mood can find no outlet. The extreme degree of 
anger is rage ; the extreme of mortification is exasperation. 
The opposite of chagrin is contentment ; when pleasantly con- 
cerned with external objects it becomes delight, when quietly 
occupied with its own affairs, happiness. 

The two opposite processes of charm and repulsion find a 
meeting point in indifference. Indifference has a tendency in 
the direction of unpleasantness ; when sense or thought is sated 
with the indifferent or perhaps originally attractive object, it 
passes over at once into repugnance. Repugnance is as much 
sense-feeling as emotion. In the latter shape, it has an objective 
form, antipathy, and a subjective, discontentedness. If the 
emotion becomes a permanent mood, we have weariness and 

In all these cases, emotion and mood are at once distinguish- 
able from sense-feeling by their connection with a train of 
strongly affective ideas. When we feel joy or sorrow, our mood 
is the result of some pleasant or painful experience which may 
be resolved into a number of ideas. If we are mourning the 
death of a friend, our consciousness is filled by affectionate 
memories, more or less clear or distinct, which co-operate to pro- 
duce the emotion. If we are made angry by some insolent 
remark, our first feeling is one of violent displeasure ; then our 
mind is flooded by a torrent of ideas connected with ourselves, 
the personality of our assailant, and the more immediate cir- 

Emotion 375 

cumstances of the insult. Most of them will not attain to any 
degree of clearness, but all are held together by the feeling of 
displeasure, which in its turn is intensified by the sense-feelings 
accompanying our expressive movements. 

A simple sense-idea which has no special relation to our past 
mental history will, therefore, hardly be able to excite an emo- 
tion, though it may call up quite intensive sense-feelings. 
Where an emotion appears we may assume the presence of 
memorial ideas, of experiences in which a similar sense-impres- 
sion was somehow concerned. The full and harmonious tone of 
a peal of bells sounds holiday-like to us, because we have been 
accustomed from childhood to interpret the chimes as har- 
bingers of holidays and religious festivals ; the blare of the 
trumpet reminds us of war and arms ; the blast of the horn 
brings up the green wood and the tumult of the chase ; the 
call of the cuckoo tells us that spring has come ; the chords of 
the organ suggest a congregation assembled for devotion. 

It is probably memory again which determines our affective 
reaction to colour-impressions, although in their case the ideas 
aroused are not so clear or distinct. Why is white the colour 
of innocence and festivity, black the colour of mourning and 
severity? Why do we choose blood-red to express energy and 
spirit, or purple to express dignity and solemnity ? Why do 
we call green the colour of hope? It would be difficult to trace 
the mood to its original source in each particular case. In 
many cases it probably arises from an obscure association 
of the colour with the occasions when custom prescribes 
its use. Purple has been the royal colour since time began ; 
and black is almost everywhere the colour of the mourner's gar- 

It is true that this association does not fully explain the con- 
nection between the sense-impression and the mood which it 
arouses. There must be some original reason for the choice of 
one particular sense-stimulus, and no other, as the expression 
of an affective state. It is perhaps justifiable to look for this 
reason in the relationship between the sense-feeling and the 
affective character of particular emotions. The sensation as 
such could then originally excite only a feeling ; but this might 
become emotion as soon as consciousness had at its disposal 

376 Lectures on Hitman and Animal Psychology 

affectively efficient memorial ideas, into which the sensation 
naturally entered as a normal constituent. 

Emotions exhibit peculiar modifications when their affective 
character is not determined, as in the cases hitherto considered, 
by impressions and ideas belonging to the present and thought 
as present ; but by ideas which refer to the future, whether in 
the way that an occurrence is definitely expected, or that some 
indefinite idea of the future gives rise to a feeling, and through 
it to an emotion. 

The most general of these expectations of the future is ex- 
pectation itself. In it we outrun the impressions of the present, 
and anticipate those which the future will bring. We look for- 
ward to its realisation ; and if this realisation is postponed, it 
becomes what we call strained expectation ; the bodily feeling 
of strain accompanies the emotion. In expectation the muscles 
are tense like those of a runner awaiting the signal for 
the race, although very possibly the expected impression de- 
mands no motor response whatsoever. Expectation becomes 
watching if the expected event may happen at any moment, 
and our sensory attention is wide awake to prevent its passing 
unnoticed. The tension is relaxed with the appearance of the 
expected impression. If the consequent perception fulfils our 
expectation, we have the emotion of satisfaction ; if not, that 
of disappointment. Satisfaction and disappointment constitute 
sudden relaxations of expectant attention. If expectation is 
prolonged, its tension will gradually disappear of itself : for, as 
you know, every emotion weakens with time. 

The opposite of disappointment is surprise. Surprise is the 
result of an unexpected event. In it we have ideas suddenly 
aroused by external impressions, and interrupting the current 
train of thought in a way which we did not anticipate, and 
which at the same time strongly attracts our attention. Sur- 
prise may be in quality pleasurable, painful, or altogether in- 
different. A special form of it is astonishment. Here the event 
is not only unexpected at the moment, but unintelligible for 
some time afterwards. Astonishment is therefore a kind of 
continued surprise. If it passes into a still more permanent 
imood, it becomes wonder. 

The feeling of rhythm, which is the single psychological 

Emotion 377 

motive in dancing, and ranks with harmony and disharmony as 
a psychological motive in musical composition, contains the 
elements both of expectation and satisfaction. The regular 
repetition in rhythmical sense-excitation makes us expect every 
succeeding stimulation, and the expectation is immediately 
followed by satisfaction. Rhythm therefore never involves 
strain, or if it does, it is simply bad rhythm. In pleasant 
rhythms satisfaction follows expectation as quickly as possible. 
Every impression arouses the expectation of another, and at the 
same time satisfies the expectation aroused by its predecessor, 
whose temporal relations it reproduces. Rhythm, that is, is an 
emotion compounded of the emotions of expectation and satis- 
faction. A broken rhythm is emotionally identical with disap- 

Hope and fear may be regarded as special forms of expecta- 
tion. Expectation is indefinite. It may refer to a desirable or 
undesirable, or pernaps to a relatively indifferent, ever.t. Hope 
and fear give definiteness to expectation : hope is the expecta- 
tion of a desirable result, fear that of something undesirable. 
It is hardly correct to call hope a future joy, and fear a future 
sorrow. The feelings can as little penetrate into the future as 
the senses. Hope and fear are the expectation of future joy 
and future sorrow, but not joy and sorrow themselves. Either 
of them may be unrealised, just as expectation may lead to 
satisfaction or disappointment. 

Fear of some immediate disagreeableness is called alarm. 
Fright bears the same relation to alarm as does expectation to 
surprise. Fright is the surprise occasioned by some sudden 
terrifying occurrence. It becomes consternation when the 
occurrence physically paralyses the individual experiencing it ; 
and it is called terror when he stands amazed before the event. 
Consternation is therefore the more subjective side of fright, and 
terror its objective side. If fear is continued, it becomes un- 
easiness. The uneasy mind is always afraid ; every occurrence 
alarms it. In other words, the emotion has become permanent, 
but at the same time somewhat less intensive. 

378 Lectures on Hitman and Animal Psychology 


The emotions both of the present and future assume the most 
varied forms according as the ideational content of the moment 
changes. Especially important are those attaching to certain 
intellectual processes and originating in the peculiar feelings 
which accompany them. We can distinguish four kinds of 
intellectual feelings : the logical, ethical, religious, and aesthetic. 
Attaching as they do to very complicated ideational connec- 
tions, they almost invariably pass over into emotions, and in that 
form exert upon our mental life an influence which far exceeds 
that of any other affective process. Their analysis belongs, of 
course, to the special sciences from which they take their name. 
But we will devote a few words to the logical emotions ; first 
because they are often overlooked altogether, and secondly 
because their relationship to the emotions of the future enables 
us to use them as illustrations of the passage of emotion in- 
general into the particular forms of intellectual emotion. 

It is well known that the rapidity of the course of thought 
exerts a considerable influence upon our general affective con- 
dition. It is not indifferent to us whether our ideas succeed 
one another at their normal rate, or proceed slowly with many 
inhibitions and interruptions, or pour in upon us in perplexing 
confusion. Each of these cases may be realised whether from 
internal or external causes. Our state of mind at the moment, 
the topic of our current thought, and external sense-impressions 
may all be of determining influence. The traveller in a new 
country is well content when his carriage takes him quickly 
from one impression to another, not so quickly that he cannot 
assimilate what he sees, but not so slowly that he is always 
wishing himself farther on amid new scenes. He is not so 
satisfied if he is lumbering along in a heavy waggon, passing 
for days together through the same scenery when he longs to 
be at his journey's end, or is curiously anticipating novel ex- 
periences. Nor is he quite happy when the railway takes him 
swift as an arrow through a country rich in historical associa- 
tion, and he tries in vain with deafened ears and tired eyes to 
fix some of its features in his memory. 

This general result can be produced by internal causes just as 

Intellectual Feelings 379. 

well as by the variation of external impressions. If you have 
to solve a mathematical problem in a short time, your thoughts 
trip each other up ; you are in a hurry to get on, but are obliged 
to go back, because you have been following out a second 
thought before you had brought the first to its conclusion. And 
it is not less disagreeable to be arrested in the middle of your 
task because your thought halts, and you cannot answer the 
next question. On the other hand, work becomes a recreation 
when one result leads certainly and easily to another. 

We have therefore the three emotions of confused, inhibited,, 
and unimpeded thought. The two last are closely related to 
the emotions of effort and facility. Correlated with these are 
the sense-feelings attaching to ease and difficulty of muscular 
action. They are generally present to some degree in the 
corresponding emotions, even when the causes of these are 
wholly mental. The feeling of effort is a weight which presses 
upon the affective condition, and whose removal is accompanied 
by a sudden feeling of pleasure. This characteristic feeling of 
relief affects us mainly by way of contrast to our previous mood. 

Special forms of the emotions of unimpeded and inhibited 
thought are those of enjoyment and tedium. In enjoyment our 
time is so well filled by external or internal stimuli to ideational 
activity that we hardly notice its passage, if we do at all. The 
nature of tedium is indicated by its name. Our time is unoccu- 
pied, and passes slowly because we have nothing else to think 
of. Tedium, therefore, has a certain affinity to expectation, but 
it is an expectation that has remained indefinite. It does not 
expect or anticipate any particular occurrence, but simply waits 
for new events of whatever kind they may be. A long con- 
tinued expectation always passes into tedium, and an intensive 
tedium is hardly distinguishable from strained expectation. 

Related to the feelings of effort and facility are .those of failure 
and success. Investigation and discovery are attended by 
feelings which show a close resemblance to those of effort and 
facility. The feelings of agreement and contradiction are some- 
what different. They originate in the comparison of simul- 
taneous ideas, which in the one case are accordant, and in the 
other refuse to be connected. 

Doubt, which we have already discussed under the heading 

380 Lectures on Human and Animal Psychology 

of oscillatory feelings (Lecture XIV., p. 219), is not the same as 
contradiction. The doubter cannot decide which of two alter- 
natives is the correct one ; he is in contradiction with himself. 
The conflicting ideas are nothing real, but simply products of 
his own thought, so that there is always the possibility that the 
contradiction in doubt may be resolved by experience or more 
mature consideration ; and so far doubt is related to the emo- 
tions of the future. This relationship becomes still more 
apparent in a special form of doubt, the feeling of indecision. 
When we are undecided, we are in contradiction with ourselves 
as to which of different roads we shall follow, or which of differ- 
ent actions we shall choose. Indecision is therefore a doubt 
implying reference to action and resolved by it. 




WE have seen that the movement among ideas which is 
characteristic of emotion in general is always attended 
by physiological movements, which exhibit specific differences 
according to the intensity and quality of the particular emotion. 
These expressions of the emotions have more than a symptomatic 
interest : they are genetically important. It is through them 
that we are able to understand the relation of emotion to the 
development of external voluntary action. Emotion bears the 
same relation to this as feeling does to the internal will-process. 
The transition from volition to external voluntary action runs 
parallel with that from feeling to emotion. But just as not 
every feeling develops into a volition, so emotion need not 
necessarily or invariably lead to a voluntary act. To take a 
special instance, the control of emotion which is natural to the 
morally and intellectually mature consciousness consists for the 
most part in its inhibition at the boundary line which separates 
it from external voluntary action. In the savage and the animal 
any emotion that is at all intensive passes over irresistibly into 
action. And even where the inhibition is effective, the internal 
tension always finds relief in movements whose only differentia 
is that they are not intended to bring about any determinate 
result. In this way arise the ' pure ' expressions of emotion, 
which are simply symptomatic of a particular internal affective 
state. They are rudiments of true voluntary actions. 

Among the regular expressions of emotion are the mimetic 
movements. They are the most characteristic of any for the 
nature of the particular emotion. Physiologically considered, 

382 Lectures on Human and Animal Psychology 

they correspond to definite reflex movements in the facial sense- 
organs. Thus the mimetic movements of the mouth, which are 
so important for the expression of the affective state, resemble 
the reflexes set up by the action of taste-stimuli (acid, bitter, 
sweet, etc.). When a man ' looks sour/ the lips are drawn out 
laterally, so that there is more space between them and the 
sides of the tongue, which are especially sensitive to acid. In 
the ' bitter ' expression the posterior portions of the tongue and 
palate, the parts most sensitive to bitter, are held apart. ' Sour ' 
and ' bitter looks,' i.e., depend on reflex movements which serve 
to prevent the contact of certain ill-tasting substances with the 
portions of the organ most sensitive to them. With the ' sweet ' 
-expression it is just the reverse. The tip of the tongue is the 
part most sensitive to sweet. The expression consists in a 
sucking movement, calculated to bring the tip of the tongue 
into as complete contact as possible with the sweet substance. 
We may imagine that all these movements depend upon a 
uniform connection of certain nerve- fibres and nerve-cells, the 
reflex movement being gradually restricted by that process of 
regulation which we discussed above (Lecture VIII., p 128). 
Direct evidence for this latter supposition is furnished by the 
fact that in early life the mimetic movements are more diffuse 
and indefinite than they become later on ; the movements of the 
mouth, e.g., are invariably accompanied by general facial con- 
tortions, and often by movements of other parts of the body. 

But the mimetic movements appear not only as the response 
to special sense-stimuli, for which they are teleological reflexes, 
but also as the expression of internal emotion. Unpleasant 
excitations, of whatever kind, will manifest themselves in ' sour ' 
and 'bitter' looks. The 'bitter' expression varies with the 
different degrees of contempt, abhorrence, and loathing ; the 
' sour,' which culminates in weeping, may denote mental as 
well as physical pain and affective disturbance. So that the 
facial expression becomes symbolic, so to say ; it is the sensible 
index of a mental condition. This assumes, of course, that the 
sensible expression and the sense-excitation producing it are 
more or less closely related to the emotion. And that is the 
case. All emotions, you remember, are accompanied by sense- 
feelings, though these may only become clearly perceptible 

Expression of the Emotions 383 

when the emotion is very intensive. Now these mimetic move- 
ments mean movement-sensations from the muscles, and they 
in turn give rise to sense-feelings which call up clearly enough 
the peripherally excited sensations to which they correspond. 
When we are looking 'sour' or 'bitter' or ' sweet,' we think 
that we are actually tasting some acid or bitter or sweet sub- 
stance ; because whenever these stimuli affect us the reflex 
movement follows, and so the sensation of the mimetic move- 
ment is fused with the sensation of taste proper. 

The process by which these movements develope, then, will 
be somewhat as follows. Every affective excitation is attended 
by bodily movements. Some of these gradually obtain an 
advantage over the rest, those, i.e., whose affective tone is 
similar to that of the emotion. This is a process of restriction 
of movement, completely analogous to that of the gradual 
restriction of reflex movement discussed above. It is true that 
the mimetic movements and the sense-feelings attaching to 
them are few in number as compared with the infinite diversity 
of emotion and mood. They can do no more than indicate the 
general class to which a particular affective state belongs. Still 
they admit of a certain amount of variation, as different facial 
expressions are combined or modified in detail. But the mimicry 
becomes more and more indefinite and equivocal as the 
emotion grows more intensive. 

Those mimetic movements which serve as a means for the 
expression of emotion and mood cannot obviously be regarded 
as true reflexes, whose invariable antecedent is the operation of 
sense-stimuli. They may with better right be termed impulsive 
movements, if we understand by ' impulse' the effort of con- 
sciousness to induce the physical condition appropriate to a given 
psychical condition. The reflex need not involve any conscious 
process at all ; in impulse some such process appears as a 
necessary condition, either antecedent to the external movement, 
or at least simultaneous with it. Do not misunderstand the 
ascription of a symbolic meaning to impulsive movements as 
compared with the same mimetic movements in their purely 
reflex function. We do not mean that they once were simply 
reflexes, and that the symbolic meaning has gradually been 
developed from their former significance. Observation decisively 

384 Lectures on Hitman and Animal Psychology 

negatives any such view. We have, on the contrary, every 
reason to suppose that the movements were first impulsive and 
later became reflex. The new-born child, which has never tasted 
acid or bitter or sweet, makes the corresponding mimetic move- 
ments quite unmistakably. When it cries, the ' sour ' and 
' bitter ' expressions appear, alternately or in combination. 
Before its lips have ever closed on its mother's breast, it makes 
sucking movements, and so 'looks sweet.' In the course of 
some weeks there develops the mimetic movement of laughing, 
the index of pleasurable mental excitation. 

These phenomena indicate quite clearly that the human child 
when it first comes into the world possesses feelings and 
emotions ; and that even at this early stage of life the emotions 
find expression in movements whose affective character is 
related to that of the emotions themselves. There is pre- 
supposed either a previous mental development, or a connate 
adaptation of bodily movement to mental state. There has 
obviously been no such development in the course of the indi- 
vidual life. We must, therefore, assume a connection which for 
the individual is original, i.e., connate. 

How is this to be explained ? The most obvious thing to do 
is to derive the association from the organic interconnection of 
nerve-fibres and nerve-cells. We may assume that the great 
majority of the sensing organs are intimately connected within 
the central nervous system with the motor fibres running to the 
mimetic muscles. But there is always the possibility that these 
connections are further developed in the course of the individual 
life ; movements which at first were diffuse and indefinite 
gradually becoming restricted. And observation raises the 
possibility to certainty : we find a continuous and continually 
increasing restriction of the mimetic movements. At the same 
time, we found ourselves obliged, in dealing with the general 
theory of the reflex process, to assume the existence of a certain 
disposition or tendency due to the original interconnexion of 
fibres in the central nervous system. The theory explained the 
increasing limitation of the reflex response in the life of the 
individual by the supposition that the connexion of sensory and 
motor nerves is the most direct possible, i.e., that it represents 
the path most usually followed by an excitation-process. 

Impulsive and Volitional Action 385 

But when we said just now that the connection of the mimetic 
movements with emotion was 'original for the individual,' there 
was implied the possibility of pushing our investigation beyond 
the limits of the individual life. The question now becomes 
problem in evolution. 

You know that Darwin based his hypothesis of the origin of 
species by ' natural selection ' upon two principles, the prin- 
ciple of variability and the principle of inheritance of individual 
characteristics. It is surely evident that these cannot be meant 
as really explanatory principles, but only as general rubrics, 
under each of which are included a whole number of problems 
to be solved. For our present purpose, however, it is enough 
to remark that, whatever their ultimate causes may be, they are 
undoubtedly as valid for mind as they are for body. Suppose 
that both conditions, variation and inheritance, have been at 
work for an indefinite time, and that the physical peculiarities 
of the organic world have differentiated it more and more ; 
there will also be constant differences to be found in mental 
disposition or tendency. The perfection and differentiation of 
species as regards body and as regards mind constitute, that is, 
two parallel processes of development. When certain nerves, 
muscles, and central organs habitually function in response 
to psychophysical impulses, their physical development must 
necessarily follow suit ; while, on the other hand, furtherance 
of physical development means increase of mental function. 

If we apply this hypothesis in our special case, it seems quite 
adequate to explain the appearance of connate impulsive actions 
observed in the new-born child. There is no reason why in 
the course of many generations certain nervous fibres and ner- 
vous cells should not advance in development and others recede, 
new ones be produced and old ones disappear. Even as between 
different individuals of the same species the number of these 
elements may differ very considerably. And the differences of 
family, race, and species arise through the summation of these 
individual variations by inheritance ; while upon the develop- 
ment of the separate parts of the nervous system and its terminal 
organs depends further the capacity of a simultaneous excita- 

C C 

3 86 Lectures on Human and Animal Psychology' 

tion of different parts of the body, i.e., the inclination towards 
combined movements of some particular kind. 

We may therefore regard the conditions of the development 
of impulsive movements as at once physical and psychical. Let 
us suppose that there exists an organism with the very simplest 
nervous system, consisting, say, of a few cells and connecting 
fibres. The impulsive movements called forth in such a creature 
by sense-stimuli will be altogether irregular. But very soon 
particular sensory fibres, which, owing to their position or for 
some other reason, are more frequently stimulated from without 
than their neighbours, will begin to develop more strongly. 
The immediate result will be a corresponding development of 
the motor fibres most directly connected with them. In this 
way a connexion will be formed, which may be perpetuated ; 
i.e., which will be present in the individual's descendants from 
the very first. Regarded from the psychical side, this process 
appears as a gradual restriction of the effect of emotion to those 
actions which call forth feelings similar to the emotion in 
affective tone, and which thus enter into intimate association 
with it. The association cannot, of course, be inherited as such. 
But since the corresponding physical connection within the 
nervous system is transmitted from one generation to another, 
the impulsive movement in the individual is just as reflexly 
certain a response to the central excitations underlying emotion, 
as it is to external sense-impressions whose effect for feeling is 
analogous. In this way the affective associations which have 
been gradually acquired in the course of a long generic develop- 
ment may be present from the first, and require but little further 
development by individual practice. 

At this point it becomes evident that no hard and fast line 
can be drawn between impulsive and expressive movement. 
Every impulsive action is a consequence, and therefore an 
expression of emotion. The animal which, impelled by its 
desire for food, throws itself upon its prey, is thereby giving 
expression to a state of mind dominated by emotion just as 
certainly as the man who expresses his grief by tears. The 
only difference is, that in expressive movement in the narrower 
sense the external action has no special purpose ; it has no 
direct effect upon the satisfaction of the pleasurable or painful 

Impulsive and Volitional Action 387 

feeling connected with the emotion. In this sense the expressive 
movements are rudiments of impulsive movements. But very 
frequently the more active emotions, such as anger or the 
pleasure in a coveted object, pass directly over into impulses 
and impulsive actions proper. Anger, e.g., becomes transformed 
into the instinct of revenge, and this finds its expression in 
movements which seek to satisfy the revengeful feeling by an 
injury done to the object of anger. Impulse, that is, bears the 
same relation to emotion in the internal experience as impul- 
sive action bears to expressive movement. And just as in the 
development of mental life impulsive action is the earlier, and 
pure expressive movement, a mere relic of it, necessarily 
the later, so the universal animal impulses, the impulses of 
nutrition, of sex, of revenge, of protection, etc., are indubitably 
the earliest forms of emotion. Or if we wish to express the 
same thought in somewhat different language : the emotions 
are impulses which have become complex, but which in pro- 
portion to their complexity have lost their characteristic of 

We are able, then, to distinguish impulsive action from pure 
expressive movement by the fact that the former has a definite 
purpose, which is consciously attained or at least attempted, 
while the latter, though it shows some faint indication of a 
purpose, does not imply the least consciousness of intention 
to attain it. In saying this, we are at the same time charac- 
terising impulsive action as voluntary action. This, and this 
alone, is the criterion of voluntary action : that the thought of 
the end to be realised accompanies or precedes it. Impulsive 
action, therefore, is simple voluntary action in the sense ex- 
plained above (Lecture XV., p. 228). 

Again, when a feeling is transformed into an emotion, it takes 
part in an ideational movement which is itself accompanied by 
feelings. Generally some particular idea stands out in this 
movement as the efficient cause of the process ; and arouses the 
appropriate impulsive action either simultaneously with its own 
appearance or directly afterwards. If at this stage a number of 
partial emotions combine to form one compound affective state, 
there may plainly be present together in consciousness a 
number of conflicting motives. And so there is developed, in 

388 Lectures on Human and Animal Psychology 

natural order from the simple, the compound voluntary action, 
or act of choice. At any one of these stages voluntary action 
may be mechanised to reflex action : the steps have been 
described above in the course of our description of the separate 
conscious processes (Lecture XV.). The pure expressive move- 
ments also fall under this law of mechanisation ; their ac- 
companiment of the emotions has ceased to be a matter o; 
consciousness and volition. 


Movements which originally followed upon simple or com- 
pound voluntary acts, but which have become wholly or partly- 
mechanised in the course of the individual life or of generic 
evolution, we term instinctive actions. ' Instinct ' is derived 
from instinguere, to incite or impel ; and in meaning-, as by 
definition, it comes very near to impulse. The only difference 
between the two consists in this, that ' impulse ' is generally 
used to denote the simpler purposive movements ; ' instinct ' 
to denote the more complex impulsive actions, which pre- 
suppose a long course of individual or generic practice. 
Instinctive action, therefore, stands midway between reflex 
movement and pure voluntary action. Thus the mimetic 
movement which follows the application of an acid stimulus to 
the tongue will be counted as a reflex ; hardly regarded as 
instinctive. But the involuntary movement of defence that a 
man makes when a stone is thrown at him we shall be inclined 
to term an instinctive action. It is evident enough that it must 
often be difficult to draw any very definite line between move- 
ments which have become entirely mechanical and those that 
still contain the impulsive element. Under certain circumstances 
the mimetic reaction to acid may be impulsive. Indeed, this 
\vill happen fairly often, i.e., whenever there is at once asso- 
ciated with the acid taste the impulse to keep the tongue away 
from the stimulus. On the other hand, the movement of 
defence may appear as a simple reflex, occurring before the 
impression of danger has come to consciousness at all. This 
uncertainty of definition, combined with the current psycho- 
logical restriction of the concept of will to choice, explains how 

Theories of Instinct 389 

it is that the chapter on instinct is one of the most debated 
fields in the science, notwithstanding that the now universal 
recognition of the genetic view of animal life in general has 
removed the principal obstacle to the comprehension of the 
more complex animal instincts. 


Still even to-day the theories of instinct form a regular 
museum of conflicting opinions. Some regard it as a purely 
mechanical result of the physical organisation, a compound 
reflex movement, only different from the simple in that the 
motor responses to particular stimuli are more complicated and 
extend over a longer period of time. Others look upon the 
instinctive action of animals as a manifestation of connate ideas. 
A third view considers it as voluntary action, involving con- 
sciousness of end or purpose, but characterised by diminution in 
the clearness of ideas. The two last hypotheses have in modern 
times been gradually superseded by a fourth and fifth, which 
have grown up under the influence of the theory of evolution. 
These, together with the first (the pure reflex hypotheses), may 
be regarded as the standard theories at the present day. The 
first of them makes instinctive actions ' mechanised rudiments 
of manifestations of intelligence.' It emphasises the opinion, 
especially with reference to animal instincts, that this mechani- 
sation has been going on for countless generations. The second, 
of which Darwin is the representative, explains instinct as 
inherited habit, determined principally by the influences of the 
environment and the struggle for existence, but also to some 
slight extent by intelligence. Like all habit, instinct has been 
subject to change ; but natural selection has brought it about 
that these changes have always been purposive, advantageous 
to the species. 

We may reject at once as wholly untenable the hypothesis 
which derives animal instinct from an intelligence which, though 
not identical with that of man, is still, so to speak, of equal 
rank with it. At the same time we must admit that the ad- 
herents of an intellectual theory in a more general sense are 
right in ascribing a large number of the manifestations of 

390 Lectzires on Human and Animal Psychology 

mental life in animals not, indeed, to intelligence, as the 
intellectualists sensu stricto do, but to individual experiences, 
the mechanism of which can only be explained (as we saw 
above) in terms of association. The precautions which the 
spider takes in fastening the threads of her web, and in selecting 
a suitable spot for it, point quite decisively to associative 
mental activity. The same is true of the many alterations 
made by honey-bees in the ordinary structure of their comb 
when they are disturbed by pieces of glass or other objects 
introduced into the hive. Indeed, it is probably impossible to 
adduce a single instance of instinct in which the actions of the 
animal do not afford evidence of some amount of individual 
experience. At the same time, there is another and parallel 
class of actions to be taken account of, which, although wholly 
purposive, cannot either be interpreted as the outcome of 
teleological reflection, or be explained from impressions and 
associations experienced during the individual life. When the 
bird builds her nest, or the spider spins her web, or the bees 
construct their comb, these are distinctly purposive actions ; 
indeed, they are more purposive than the other actions of the 
same animals which are explicable in terms of individual 
experience. If it were really teleological reflexion that led the 
bird to build her nest, the spider to spin her web, and the bees 
to make their comb, we should be compelled to attribute to 
these animals a degree of intelligence which the experience of 
a single life could hardly be expected to develope even in man. 
Another argument that makes against this explanation is 
the regularity with which the same actions are repeated by the 
different members of a single species in cases where no con- 
nexion can be demonstrated between the various individuals 
such as might possibly account for the uniformity. Of course, 
there is an intimate connexion existing between the inhabitants 
of the same hive or ant's nest, and between parents and young 
in the species in which the family holds together for some little 
time. But in numberless instances the animal begins its life in 
total independence of its fellows. When the caterpillar hatches 
out of the egg, its parents are long since dead ; nevertheless, it 
spins the cocoon that they did. And, lastly, to interpret 
instinctive action as intelligence would in very many cases be 

Theories of Instinct 391 

to predicate of it a prevision of the future. It is hard to suppose 
that this prevision is conscious when there are neither analogous 
experiences given in the previous life of the individual, nor any 
way by which they could be communicated to it. The night- 
flying phalcBna covers the eggs which it has laid with a layer of 
fur made from its own hair, to protect them against the cold, 
before the winter has come. The caterpillar changes into a 
chrysalis without any experience of the metamorphosis which 
it is to undergo. 

We cannot better demonstrate the impossibility of a deriva- 
tion of instinctive action from conscious reflection than by 
quoting an illustration from an earlier author, in which all the 
contrail ictions which the theory involves are brought together 
into short compass. The caterpillar of the emperor moth spins 
at the upper extremity of its cocoon a double arch of stiff bristles, 
held together above only by a few fine threads. The cocoon, 
i.e., opens at the very least pressure from within, but is able to 
resist quite strong pressure from without. Autenrieth writes of 
this in his AnsicJiten ilber Natur- und Seelenleben : ' If the cater- 
pillar acted from reflection and with understanding, it must, on 
human analogy, have pursued the following train of thought : 
that it had reached its chrysalis stage, and would therefore be 
at the mercy of any unlucky accident, without possibility of 
escape, unless it took certain precautionary measures in advance ; 
that it would have to issue from its cocoon as imago without 
having organs or strength for breaking through the cover it had 
spun as caterpillar, and without possessing any secretion, like 
other insects, which would if emitted eat through the threads of 
silk ; and that consequently, unless it took care to provide as 
caterpillar a convenient exit from its cocoon, it must certainly 
come to a premature end in imprisonment. On the other hand, 
it must have clearly recognised during its work upon the cocoon 
that, in order to have free egress as imago, it would only be 
necessary to construct an arch which could resist attacks from 
without while opening easily from within ; and that these con- 
ditions would be fulfilled if the arch were made of stiff threads, 
inclined together in the median line, and with their ends left free. 
At the same time it must have realised that the plan could be 
carried out if the silk employed for the construction of the other 

392 Lectures on Human and Animal Psychology 

parts of the cocoon were employed with special care and skill at 
the upper end. Yet it could have learnt nothing of all this from 
its parents : they were dead long before it had issued from the 
egg ; it had had no practice or experience, for the spinning of 
the cocoon happens only once in a lifetime ; it could not imitate 
a neighbour, for the species is not a social one. And during 
the whole of its existence as caterpillar its understanding could 
have been but very little developed : it crawled about on the 
branch where it first saw the light, devouring leaves, an occupa- 
tion which required no consideration, since the food was there 
waiting for it ; it clung fast with its feet, perhaps, to avoid falling 
to the ground, and crept under a leaf to escape the rain ; it got 
rid of its old uncomfortable skin some few times by involuntary 
contractions of its entire body, but without making any cocoon : 
and that was the whole of its life, the sum of its opportunities 
for the exercise of intelligence.' 

Instinctive action, then, cannot be explained either in terms 
of conscious reflection or from individual associations : the 
hypothesis requires an amount of prevision on the part of the 
animal which is psychologically impossible. But the opposite 
theory, recently defended by Herbert Spencer, that instinctive 
action is simply compound reflex action, determined by the laws 
of the physical organisation, is equally untenable. That the 
caterpillar secretes silk, the spider the material for her web, and 
the bee wax, is just as much a matter of physical necessity as is 
the emission of any other secretion. But that these substances 
when secreted are worked up in such definite and artistic forms 
is altogether inexplicable from the facts of physical organisation. 
That accounts for the material which the animal has at its dis- 
posal, but not for the form, which is the real result of its work. 

Worse still, if that is possible, is the view which stands mid- 
way between the intelligence and reflex theories, and which 
regards connate ideas as the motives of instinctive action. The 
bee is supposed to have in its mind from the first a pattern of 
its hexagonal cell, the spider a pattern of the meshes of her web, 
the caterpillar a picture of its cocoon, and the bird one of the 
nest it is to build ; and each of these animals must necessarily 
translate its idea into reality. The older philosophical idealism 
found in such a hypothesis a welcome support for the doctrine 

Theories of Instinct 393 

of innate ideas. But it contradicts everything that our analysis 
of the human consciousness Has taught us. It is impossible to 
prove the existence in our own minds of ideas which do not 
spring from the experience of the individual life. The congeni- 
tally cleaf has no knowledge of tone, the congenitally blind none 
of colour. And the probability that complex ideas can be 
innate is infinitely less. Besides, the observation of instinct 
does not by any means give unqualified support to the hypo- 
thesis. If there is so definite an image of the hexagonal ceil in 
the bee's mind, how is it that all the cells of the hive are not 
made of the same size? You see, there must be present in its 
consciousness not the idea of a single cell, but that of the whole 
number of cells belonging to the colony, if its action is to 
become intelligible in every respect. The bird builds its nest of 
certain determinate materials, from which it never varies except 
in cases of necessity. Does the innate idea of the nest include 
the ideas of every twig and straw used in its construction ? It 
is evident that this theory becomes entangled in difficulties no 
less grave than those which proved fatal to the hypothesis of 
intelligent action. It requires the assumption not of a single 
innate idea, but of a whole connected series, in a word of an 
innate activity of thought with a large store of experience 
behind it. 

Only two hypotheses remain, therefore, as really arguable. 
One of them makes instinctive action a mechanised intelligent 
action, which has been in whole or part reduced to the level of 
the reflex ; the other makes instinct a matter of inherited 
habit, gradually acquired and modified under the influence of 
the external environment in the course of numberless genera- 
tions. There is obviously no necessary antagonism between 
these two viexvs. Instincts may be actions originally conscious 
but now become mechanical, and they may be inherited habits. 
This compromise would have a great deal to recommend it, if 
we might slightly alter the first theory, and make instinct, 
according to it, partly a matter of mechanised volition and 
partly of action which is still determined by psychical motives. 
If we are ourselves to appeal to the facts for a decision in favour 
of one of the two views or for a verdict for or against a combi- 
nation of both, we shall do well to keep in mind the rules laid 

394 Lectures on Human and Animal Psychology 

down above in connection with our consideration of the mani- 
festations of animal 'intelligence' (Lecture XXIIL). Never have 
they been so sadly sinned against as in this particular chapter 
of psychology on the nature of animal instinct. The first ran, 
you remember, that we must always set out from known facts 
of the human consciousness ; the second, that simple principles 
of explanation are always to be preferred to complex. 

We must, therefore, go on in the following lecture to discuss 
briefly the instinctive actions in man. When we have done 
that, we may pause to look back once more upon the very 
difficult phenomena presented by animal instinct. 




BY an instinctive action we understand, as remarked above, 
something purposive, but involuntary, half impulsive and 
half reflex. It cannot be doubted that in this sense many 
human activities come under the category of instinctive action. 
We laugh and weep, we make the most complex mimetic 
movements, without, or even against, our wish or our know- 
ledge. Most of our movements are determined by emotion, 
and volition manifests itself quite as often in the moderation or 
inhibition of movements as in their independent initiation. Not 
seldom the will simply makes definite the direction of a move- 
ment ; its execution is left to instinct. When we walk, it is 
generally volition that prescribes the road ; but step follows step 
instinctively. Many actions at first require practice and the 
exercise of voluntary effort, but when once they have become 
familiar, may be performed under almost exclusively instinctive 
control. The child learning to write will laboriously copy every 
stroke of the pen ; the ready writer needs only the intention to 
write some particular word, and it stands before him on the 
paper. The novice at the piano must strain his attention upon 
every note, in order to find the appropriate key ; the practised 
player translates the printed page mechanically into the proper 
movements. Any movement that has become altogether 
habitual is made instinctively. An impulse of will is, of course, 
necessary at the outset ; but its effect extends to a whole series 
of actions, and each particular one takes place without effort 
and without knowledge : the series once started is continued to 


396 Lectures on Hitman and Animal Psychology 

its end with the same unconscious certainty and purposiveness 
as the reflex. The voluntary movements of early childhood are 
uncertain and awkward ; practice has not had time to transform 
them into instinctive acts. And the same is true of the adult 
whenever he wishes to perform some as yet unaccustomed 
action, of however simple a character. Precision and grace of 
movement, then, depend upon certainty of instinct, not upon 
firmness of will. 

This transformation of voluntary into instinctive activity is 
greatly furthered by the influence of the environment. From 
the first days of life we are surrounded by our fellow-men, and 
imitate their actions. And these mimetic movements are in- 
stinctive in character. As soon as the child's consciousness is 
aroused from its first sleepy passivity, it begins to perceive the 
expressions of others' emotions, and to respond to them by 
similar emotions with corresponding impulses. The continued 
imitation by which a child comes to learn the language that is 
spoken round it is impulsive, not voluntary. Even the peculiar 
word-formations, of child-language are not," as is often wrongly 
held, invented by the child, but borrowed by it from its environ- 
ment, from the words of nurse and mother, who in their inter- 
course with it adapt themselves to its level of mental develop- 
ment and capacity of articulation. And with them, again, this 
formation of special baby- words and imitative sounds is to a 
very slight degree a matter of purposed invention ; for the most 
part the adaptation and imitation are themselves instinctive. 
Voluntary act and instinctive movement, suggested by environ- 
ment and example, cross and recross in human conduct from 
the beginning to the end of life. And if the sum of action 
resulting from personal choice and intellectual reflection were 
laid in one scale, and that proceeding from instinct and imita- 
tion in the other, there can be little doubt that the beam would 
incline on the side of instinct. Suppose a bird were to become 
interested in zoological investigation ; he might well regard man- 
kind as the richest of all creatures in instincts. Man shares 
with the birds the instinct to live in wedlock ; like the fox, he 
educates his children ; he has the beaver's impulse to build 
houses, and the bee's custom of founding states and sending 
forth colonies ; while he has in common with the ant a pleasure 

Acquired Ins finds 397 

in war, in slave-making, and in the domesticating of useful 

There is, it is true, one immense difference. In man all these 
instincts, at least in the form which they have assumed in the 
course of history, are the fruits of a continuous intellectual 
development, not a trace of which is demonstrable among the 
animals. And a great gulf is set up also by the fact that within 
the limits of these general norms of life individual volition has 
ample space for the determination of its particular conformity 
to them. Still, if human conduct as a whole is divided into the 
two great departments of voluntary and involuntary action, 
there can be no doubt that for the vast majority of us the 
principal incentive to those very acts which constitute the 
universal criterion of the genus homo is not reflection and free- 
will, but instinctive imitation of our neighbours. Reflection and 
volition begin as a rule only when the general norms of life 
have to be applied in the particular case. How the individual 
builds his house, or where he lives, may be a matter of pro- 
tracted consideration 'for him. But that mankind at large build 
houses and seek shelter seems to him to be as natural and right 
as it probably does to the bee to construct its hexagonal cells. 
And even the question of the particular disposal of his own life, 
which is so tremendously important for the civilised man, 
generally troubles the savage but little. He builds his hut or 
pitches his tent as his fellows do, and as his forbears did before 
him. So that human life is permeated through and through 
with instinctive action, determined in part, however, by intelli- 
gence and volition. As for that, all forms of psychically con- 
ditioned action are mixed processes. It hardly ever happens 
that a fact of consciousness admits of complete subsumption to 
any of the categories that psychological abstraction enables us 
t?> set up. Like mental life in general, it contains a mixture of 
vuious elements. 


Instincts which, so far as we can tell, have been developed in 
this way during the life of the individual, and in the absence of 
definite individual influences might have remained wholly un- 
developed, may be called acquired instincts. You can see from 

398 Lectures on Human and Animal Psychology 

what has been already said that all and each of them, from the 
instinctive finger-movements of the practised pianist down to 
the instinct to build a shelter and wear clothes for protection 
against the weather, spring from two conditions, one physio- 
logical and the other psychological. The former consists in the 
property of our nervous organisation gradually to mechanise 
complex voluntary movements ; the second, in the operation of 
the mimetic impulse, which is probably natural to all animals 
that live in any kind of society, but is especially powerful in 
man. This impulse is itself an instinct ; mimetic movements 
are, as a rule, impulsive, and not volitional. But it is at the 
same time the fountain-head of many other instincts, and 
especially of those whose development is furthered by a social 
mode of life. It is a necessary corollary from these remarks 
that the first of the two conditions will, be effective, even if the 
second be absent, in the case of the acquired instincts sensu 
stricto, instincts developed during the individual life as a result 
of individual practice, such as the instinctive movements of the 
skilled pianist. These are purely matters of physiological 
practice ; so that it is not difficult to understand that the move- 
ments may occasionally become quite reflexive. The hypothesis 
which is most nearly adequate to this special case, then, will be 
the fourth of those which we reviewed above as professing to 
account for instinct in general : that of the passage of intelligent 
into reflex action. I say ' most nearly ' : for the expression ' in- 
telligent action ' is not admissible in the present instance, any 
more than in the other contexts in which we have discussed it. 
In most cases there are no acts of intelligence involved at all, 
but only associations ; and in any case intelligent action must 
have been reduced to association before it could become mecha- 
nised. The piano-player has first of all to form a stable associa- 
tion between the printed note and the movement of touch. But 
this association gradually lapses from consciousness, and the 
interconnection of movements becomes purely mechanical. 

The operation of the second condition, the psychological 
impulse toward imitation, is to be seen, often enough in com- 
bination with the physiological factor, in the case of the .social 
instincts. The fact which lies at the root of the imitative im- 
pulse s this, that as a rule any action resulting from psychical 

Connate Instincts 399 

motives excites in all individuals of the same species an emotion 
similar to that experienced by the agent himself. And similarity 
of emotion means similarity in its external expression. The 
simplest manifestations of the imitative impulse, then, will be 
found in the different forms of violent emotional expression. 
The passionate gestures of a speaker are reflected in the in- 
voluntary movements of his audience. As we look at a terrified 
or sorrowful face, our own features assume a cast in keeping with 
the feelings it expresses. In all these cases the imitative move- 
ments are purely instinctive. On the other hand, if the strange- 
ness of the presentation is such as to evoke an act of will on our 
part, the instinctive reaction passes over into some less simple 
form of action. This is seen in all the human social instincts, 
where the sphere of instinct borders on that of custom. The 
phenomena here are of a kind so mixed and complex, that their 
instinctive element is usually entirely overlooked. 


To be distinguished from these acquired human instincts are 
others, which are connate. They are, perhaps, more modified in 
man than in the other animals by civilisation and education ; 
but they are still indispensable for the origin of the most 
important vital functions. There are in particular two funda- 
mental instincts of organised nature, the impulses of sex and of 
nutrition, which appear unchanged in man, as connate instincts. 
The investigation of the conditions of connate instinct in general 
is exceedingly difficult. But that, of course, is so much the 
more reason for starting out from the facts of the human con- 
sciousness, which furnish the only directly accessible observa- 
tional material. 

Do connate instincts spring from connate ideas, or do they 
depend upon intellectual processes ? You will see at once that 
such hypotheses as those could never have been set up had not 
mankind been left out of account in their formulation. Or is 
the impulse to imitation in some way or another a factor in their 
constitution, as in that of the acquired social instincts ? To this 
question also we may return a negative answer, without more 
words. Are we, then, to look on these manifestations of original 

400 Lectures on Human and Animal Psychology 

instinct as something analogous to the mechanised voluntary 
actions that now resemble reflexes ? Certainly, if you observe 
the first sucking movements of a new-born mammal, those that 
appear before it has satisfied its hunger by actual sucking of 
milk, you will not find much to object to in the term ' reflex.' 
But none the less it is impossible to suppose that these reflexes 
have originated in a similar way to the mechanised movements 
(say those of the pianist) that have come about by practice. No ! 
so long as we confine ourselves to the life of the individual, there 
can be no question that they are original, and not acquired. It 
looks, therefore, as though we had found an exceptional case to 
.support the reflex theory, which has proved untenable every- 
wnere else. 

But we must not decide in its favour too hastily ; we must 
go to observation for refutation or confirmation. The reflex 
theory assumes that the sucking movements of the new-born 
mammal are not only involuntary, but unconscious. Like 
reflex movement in general, they are purely physiological in 
nature : they show an entire absence of psychical motives. 
Now, though such an assumption might look reasonable enough 
on the study-table, it is hardly a theory that any one would 
hold who had ever really seen the movements of a hungry 
infant. Every feature and every gesture betoken the presence 
of unpleasant feelings. Plain enough to read in its crying 
and movements is the inarticulate complaint : ' I am hungry.' 
Give it anything that can be sucked, a finger or a corner 
of its pillov. All movement ceases ; sucking, and only sucking, 
is the business of the moment. It is not long, of course, before 
the restlessness comes back again, only to be finally overcome 
by 'the satisfaction of hunger. 

It is wholly impossible that all this is a matter of purely 
physiological reflexes. If emotional expressions have any 
significance at all, the infant's movements can only be inter- 
preted as psychically conditioned actions, i.e., manifestations of 
impulse. No doubt we must suppose that in these first impul- 
sive movements there is not present a shadow of the idea of 
th end towards which the impulse is directed. But that is 
not at all necessary for the origin of emotional and impulsive 
expression. Sensations, with the feelings attaching to them, 

Connate Instincts 401 

are altogether adequate to the result. And they are given 
in the sensation of hunger, which is physiologically conditioned, 
and the unpleasant feeling connected with it. 

At the same time, there is one part of the effect that these 
causes do not suffice to explain, the very phenomenon which 
gives to these impulsive movements their character of pur- 
posiveness, and renders it possible for them to attain their 
end : namely, the sucking movement of the lips, which is 
in no sense a characteristic of unpleasant emotions in general. 
None the less we may regard this as a special emotional 
expression, inseparably associated in the human infant with 
the intensively toned sensation of hunger. And if the move- 
ment is one of expression, its purposiveness becomes intel- 
ligible. For while expressive movements are the means of 
expression of individual emotions, their general nature and in 
particular their characteristic of purposiveness result from a 
process of development extending beyond the individual : their 
physiological conditions are inherited or, what in this connec- 
tion is the same thing, were acquired in the course of earlier 
generations, reaching back into an unlimited past. And this 
shows us the grain of truth that is contained in the reflex 
theory. The sucking movements of the new-born child are 
reflexes, in the sense in which expressive movements in general 
are reflexes. Their purposiveness, like that of the reflexes, is 
due to an organisation acquired in the course of generic, not 
individual, evolution. But they differ from the reflexes proper 
in this, that they are accompanied by emotions in the mind, 
and that their performance is regulated by these emotions 
It is just the combination of these two characteristics that 
constitutes the peculiarity of the connate impulse. It stands 
midway between the reflex and the acquired impulsive action : 
related to the former in that its ultimate basis is physiological, 
and to the latter in that it springs directly from psychological 
conditions, which may at any time interfere to modify its 
original character. 

D l) 

4.O2 Lectures on Human and Animal Psychology 


If we survey all those phenomena of human conduct which 
are referable to instinct, we see that the simplest conditions 
of instinctive action in general are to be found in the cases 
where it is the result of individual practice. Here the action 
simply indicates a disposition of the physical organisation, 
which has been induced by movements often repeated in the 
past. The performance of a definite complex act and its con- 
nexion with an adequate sense-stimulus have become more 
and more matters of course ; till at last they are rendered 
completely mechanical. In the second place come the ac- 
quired social instincts, whose conditions are complicated by 
the development of the social emotions and the corresponding 
.mimetic actions. Lastly, the connate instincts oblige us to 
assume that the disposition of the physical organisation plus 
the mechanisation of complex movements correlated with it, 
if induced through a number of generations, leaves behind it 
permanent physical effects, common to all individuals alike ; 
so that certain impulsive movements, subserving the ele- 
mentary necessities of life, take on the reflex form. They 
may then constitute a starting-point for fresh developments, 
through which the impulse can arrive at a special degree of 
perfection in special individuals. 

The effect of ' practice ' and ' habit ' can only be due to 
after-effects of excitation, of the kind assumed by us for the 
explanation of instinctive movements. And since the expres- 
sions of instinct are par excellence ' customary ' or ' habitual ' 
actions, their subsumption to the general law of practice needs 
no justification. That law runs as follows : the more fre- 
quently a voluntary action is repeated, the easier is it to perform, 
and the greater is the tendency of its constituents (if it is a 
complex act) to take on the reflex form, i.e., to arrange them- 
selves in a connected series of movements, which runs on 
mechanically when once initiated by an adequate stimulus. 

The formulation of this law shows us at once that its basis 
must be physiological. The goal attained by the process of 
practice is simply the mechanisation of movements which were 
originally dependent upon psychical antecedents. That must 
mean that mechanical, i.e., physiological, alterations of the 

Practice, Imitation, and Inheritance 403 

nervous system are at the bottom of the whole matter. We 
are still so much in the dark as regards the real nature of 
nervous processes, that we need not be surprised to find the 
exact physical and chemical character of these alterations 
quite unknown. If we know nothing more about them, we 
are at least certain that they exist ; the witness of the actual 
results of practice cannot be called in question. There is 
hardly any movement of the human body, however difficult, 
which we cannot, by continued practice and repetition, reduce 
to a mechanical certainty so complete, that it will be performed, 
even without any intention on our part, as the necessary 
reaction to certain sense-stimuli. Very remarkable instances 
of this mechanisation of complex actions by practice occur 
now and again in the conduct of ' absent-minded ' persons. 
It is quite a common experience to begin a customary action 
at a time which is altogether unsuited for it, the stimulus 
having been given by some familiar impression. We may 
intend to pass by our own house or the place of our daily 
business, but suddenly discover that we have mechanically 
followed our usual route, and entered the building without 
in the least meaning to do so. Some years ago I was occupied 
with certain physiological experiments on the frog, each of 
which involved the performance of a fairly complicated opera- 
tion. It happened one day that I had taken up a frog for the 
purpose of making a quite different experiment. I suddenly 
found, to my great astonishment, that, instead of making tin* 
intended experiment, I had performed the customary operation. 
Now we certainly cannot regard acts of this kind as pure 
reflexes. The impressions are not only physiological stimuli, 
but psychological motives as well. But the reaction to them 
is impulsive : the familiar visual impression calls up the sensa- 
tions, feelings, and movements associated with it. The move- 
ment could not, however, become instinctive in that way 
unless the succession of movements had been thoroughly prac- 
tised physiologically. The greater the extent of this practice, 
the more effective is the inhibition of the conscious realisation of 
what we are doing, which puts a stop to the unintended action. 

There are experiences of the most different kinds, then, 
which put the physiological effects of practice beyond the 

404 Lectures on Human and Animal Psychology 

shadow of a doubt. But there is yet another proof of their 
reality in the functional properties of the nervous elements. 
If you excite a motor nerve by a stimulus so weak that it 
only just occasions a contraction of the muscle attached to 
it, and continue to apply this same stimulus at intervals just 
sufficient to avoid exhaustion, you will find (especially if the 
nerve is in good condition) that the contraction gradually 
increases in amount. This increase of excitability by stimula- 
tion can be best seen in the reflex movements that follow the 
stimulation of a sensory nerve connected with the cord, 
supposing always that the experiment is made under condi- 
tions which preclude the adverse influence of fatigue. The 
molecular changes in the nervous elements on which the 
increase of excitability depends are, as was said above, still 
unknown. But we can get some idea of them by taking a 
few common illustrations of the facilitation of a movement 
by its repetition. As a carriage-wheel, for instance, turns 
round the axle, the rough surfaces are gradually worn smooth ; 
the frictional resistance is diminished. A watch, as you all 
know, goes better the more regularly it is wound up : and 
so on. Similarly, we may suppose, repetition facilitates the 
functioning of the nervous elements by removing all manner 
of obstructions and inhibitions. Now a complex muscular 
movement consists of a definitely arranged sum of simple 
movements, every one of which depends upon some element- 
ary excitation-process. Each preceding excitation in such 
a series serves as the adequate stimulus for the succeeding 
one. This means that the effect of practice consists not only 
in the facilitation of every particular component of the com- 
plex process, but also in that of the definite combination 
of elementary movements which go to make it up. 

You can easily see that this law of practice possesses a 
significance for the physical basis of our mental life which 
extends far beyond the sphere of instinctive action. Not only 
the combinations of certain movements, but the associations 
of sensations and ideas in general, are rendered stable by 
practice. Contiguity- and similarity-associations alike bear wit- 
ness to its influence. The former are directly correlated with 
the habitualness of certain excitatory processes in a sense- 

Practice, Imitation, and Inheritance 405 

centre, the effect of which is to facilitate the genesis of sensa- 
tion when the same impression is repeated ; the latter depend 
upon our habituation to a particular connection of simulta- 
neous or successive excitations. Regarded from this point of 
view, that is, instinct appears as an extension of association 
to the motor sphere. 

These laws of practice suffice for the explanation of the 
acquired instincts. The occurrence of connate instincts renders 
a subsidiary hypothesis necessary. We must suppose that the 
physical changes which the nervous elements undergo can be 
transmitted from father to son. Later generations will then 
be affected in two ways : they will from the first acquire 
familiarity with certain complex movements more easily, owing 
to connate dispositions of the nervous system ; and they will 
react to particular stimuli by reflex movements of mechanical 
certainty, owing to particular nervous dispositions of a more 
definite and clearly marked kind. The assumption of the 
inheritance of acquired dispositions or tendencies is inevitable, 
if there is to be any continuity of evolution at all. We may 
be in doubt as to the extent of this inheritance : we cannot 
question the fact itself. It is in particular the inherited reflexes 
of the human infant, so important for the development of its 
instinct of nutrition, that belong to those constituents of ori- 
ginal disposition which reach far back to the beginnings of 
generic development. But more individual gifts, the trans- 
missibility of certain talents is unquestionable, also appear to 
lend probability to the view that the propagation of definite 
dispositions takes place, at least within certain limits. Disposi- 
tions of this kind, however, are not the products of any very 
long development ; and are probably to be looked on rather 
as dispositions facilitating the practice of new functions than as 
ready-made systems of reflex arcs. It is of great importance, 
by the way, in this matter of the transmissibility of more 
special gifts, that the disposition of associations and the direc- 
tion of instincts be in complete agreement. A connate talent, 
especially if its field of exercise is internal rather than external, 
depends at least as much upon the disposition to form certain 
associative connexions as upon the facility of certain complex 
forms of movement. But in every case the point to remember 

406 Lectures on Human and Animal Psychology 

is that it is the disposition, not the actual functional capability, 
which is connate. Every instinctive action, however original it 
may be, the taking of food by the infant, e.g., must to a 
certain extent be acquired afresh by the individual. Far more 
practice, then, will be required for the realisation of the connate 
talent, which has so short a period of development behind it. 
Readiness of movement and many-sidedness of ideational con- 
nexion are the promise of the connate disposition ; the fulfil- 
ment of the promise comes later in life. Ideas cannot be 
inherited any more than complex volitional actions. Talent 
and instinct alike are latent until external stimulation calls 
them into actual life. 

We have now reviewed the conditions of origin of the human 
instincts. How does the matter stand with the analogous 
phenomena presented by the animal kingdom ? Are they de- 
ducible from the same conditions, perhaps with the difference 
that the various factors are concerned in different amounts ? 
Or must we look for other and peculiar explanatory reasons ? 

Different in expression as animal and human instincts are, 
their fundamental similarity can hardly be doubted. The first 
question must be answered in the affirmative, the second in the 
negative, though the negation cannot be absolute : for the 
conditions of human life are such that in it certain influences 
tend to disappear, and may accordingly be left out of account, 
which acquire a very considerable importance in the life of 
animals. To see the necessity of this admission, we have only 
to cast a glance at instincts which enable an animal, e.g., a 
caterpillar, to provide not only for itself, but also for its larval 
condition and even for a still later condition, that of the imago, 
without the aid of the example of other animals or of any 
previous experiences of its own. Can the principles which our 
explanation has adopted explain this, that a caterpillar living 
in a pomegranate cuts a way out of the fruit just before its 
transformation, and then makes this particular part of its home 
fast with silk thread to the nearest branch, that it may not fall 
to the ground before the transformation is complete ? Many 
similar instances are quoted by Darwin in a posthumous essay 

Relation of Animal to Human Instinct 407 

on instinct, published as an appendix to Romanes's work 
Mental Evolution in Animals. Here belongs, too, the case 
of the caterpillar of the emperor moth, which we employed 
as an argument against the intellectualistic hypothesis (p. 391). 
These are all connate instincts ; so that the closest analogy to 
them in our own experience would be furnished by the sucking 
movements of the hungry infant. But these are sufficiently 
simple to be referred to a ready-prepared reflex mechanism. 
Can the same be said of the complicated animal actions which 
conform so wonderfully in the different species to the special 
conditions of life ? And, granted that it can, how far will the 
previous life-history of the species enable us to explain the 
origin of the particular reflex mechanism ? 

We do not know the details of this life-history. And there- 
fore we must give up any hope of a real genetic explanation 
of instinct. All that we can do is, first, to test the general 
question of the possibility of the origin of reflexes, which do not 
simply involve a definite and unchangeable co-ordination of 
movement and stimulus, but a co-ordination which may vary 
with variation of its special conditions ; and, secondly, to 
inquire whether the term ' reflex ' is really applicable to the 
facts as stated. Now it is true that brainless animals exhibit 
reflexes of the kind that varies with its special conditions. 
A frog the whole of whose brain has been removed, with the 
exception of the optic lobes, not only tries to escape when its 
skin is stimulated, but avoids obstacles placed in its way. But 
in other respects the movement has all the characteristics of a 
reflex. Apply this to the present case. There can be no doubt 
that here, too, variations occur in accordance with special con- 
ditions ; if only for the reason that (as our previous discussion 
shows) the movements, like those of the hungry infant, are not 
purely reflexive, but expressive of emotions, the expression 
being mediated by preformed purposive connections within the 
nervous centres. So that, strange as the instinctive action of 
an animal like the caterpillar may at first sight appear, it yet 
differs only in degree from the action of the human infant, 
which we have found comparatively easy of explanation. ' 

1 For a description of the various animal instincts, cf. G. H. Schneider, 
Der thierische Wille (1880). Unlike so many works upon the same subject, 

408 Lectures on Human and Animal Psychology 

There still remains one point which requires further elucida- 
tion. Hitherto, relying upon the facts of human experience 
\ve have been bridging the difference between connate and 
acquired instincts in this way : we have supposed that father 
can transmit to son the physiological dispositions that he has 
acquired by practice during his own life, and that in the course 
of generations these inherited dispositions are strengthened and 
made definite by summation. But is it possible to conceive of 
any specific life-history into which there could be crowded such a 
multitude of tendencies as should finally give rise to a succession 
of instinctive actions so complicated as those of the caterpillar, 
of the emperor moth, or even of the bird of passage, which 
flies south in winter without precept or example ? Surely the 
analogy of the practised pianist fails us here. But is it really 
applicable even to the connate instincts of man ? Do not they 
imply besides the action of will, which is required to introduce 
what afterwards becomes habit, a compelling force residing in 
the external conditions of life ? We do not know from what 
beginnings the functions of nutrition have been developed in 
man, except in so far as the facts of structural evolution admit 
of functional inference. But his general mental attributes en- 
able us to assume quite definitely that the earliest development 
and consolidation of habit took place under the conjoint and 
unfailing influence both of external circumstances and of volun- 
tary actions proceeding from feelings. 

And this leads us to the principle which Darwin enunciated 
as of prime importance for the development of instinct and for 
the course of evolution in general, the principle of adaptation 
to environment. There can be no doubt that this adaptation 
and voluntary action constitute the two universal determinants 
of the development of animal impulses. The first supplements 
the second ; volition must have an object towards which it is 
directed. The converse, of course, need not necessarily be the 
case. In the vegetable kingdom specific alterations are gradually 
effected by the sole operation of the environment, influencing 
the functions of growth or favouring certain peculiarities which 
are thus more readily and certainly perpetuated. And this 

this volume may be recommended as giving an impartial and accurate 
account of observed facts. 

Relation of Animal to Human Instinct 409 

passive adaptation will naturally be found among animals as 
well, since they share with plants all the physiological functions 
which are capable of modification by it. But Darwin's explana- 
tion of the development of instinct as being mainly the result 
of passive adaptation seems to contradict the facts. Instinctive 
action is impulsive, that is voluntary action and however far 
back we may go, we shall not find anything to derive it from 
except similar, if simpler, acts of will. The development of any 
sort of animal instinct, that is to say, is altogether impossible 
unless there exists from the first that interaction of external 
stimulus with affective and voluntary response which constitutes 
the real nature of instinct at all stages of organic evolution. 
We may possibly succeed in deriving a complicated form of 
instinct from a more simple one ; but we can never explain in- 
stinct in terms of something which is as yet neither instinct nor 

External conditions of life and voluntary reactions upon them, 
then, are the two factors operative in the evolution of instinct. 
But they operate in different degrees. The general development 
of mentality is always tending to modify instinct in some way 
or another. And so it comes about that of the two associated 
principles the first, adaptation to environment, predominates 
at the lower stages of life ; the second, voluntary activity, at 
the higher. This is the great difference between the instincts of 
man and those of the animals. Human instincts are habits, 
acquired or inherited from previous generations ; animal instincts 
are purposive adaptations of voluntary action to the conditions 
of life. And a second difference follows from the first : that the 
vast majority of human instincts are acquired : while animals, 
apart, of course, from the results of training, which do not con- 
cern us here, are restricted to connate instincts, with a very 
limited range of variation. This makes it to a certain extent 
intelligible that the older psychology, failing to see the close 
connection of habit and practice with instinct, usually ascribed 
instinctive action to the animals alone and denied it to man. 
The corollary from that connexion really is, that animal instinct 
is more predominantly reflexive, more exclusively constituted of 
purposive movements given with the connate physical organisa- 
tion. If the complexity of a number of instinctive actions in 

4io Lectures on Huuian and Animal Psychology 

animals seems to contradict that view, you must remember that 
throughout the animal kingdom they remain relatively uniform. 
We might almost say that the whole organisation of the central 
nervous system seems in many cases to be determined by certain 
associations that have been established by instinct. 




NEW and peculiar conditions for the development of in- 
stincts are to be found in the common life of animals. 
This is, of course, a product of social instincts ; but it 
reacts in the most various ways upon the original impulses 
which occasioned it. 

At the very lowest stages of animal life we see every crea- 
ture seeking its like. Many of the medusae and molluscs, 
many insects and fish, unite temporarily in swarms or schools. 
In all such cases it is not the individuals, but the species, which 
know one another. At the same time, the origin of the social 
impulse can only be looked for in a feeling of inclination, how- 
ever primitive, which attracts animals of the same species 
towards each other by the intermediation of certain sense- 
impressions, perhaps of smell or sight. At a higher stage of 
development this feeling of inclination shows itself as an indi- 
vidual attraction of animal to animal. But this is not found until 
we come to the higher birds and mammals. Dogs, as you of 
course know, manifest very pronounced likes and dislikes. If 
two poodles, e.g., are kept in the same house, there may spring 
up between them a kind of friendship ; the survivor mourns the 
loss of his comrade. Horses from the same stable become 
similarly attached to one another. Most remarkable are the 
friendships which arise between animals of different species as 
a result of living together. Even a dog and a cat may become 
friends. The inclination in all these cases is purely individual 
The dog will pick out his comrades from a score of other dogs ; 
and however gracious his behaviour to a particular cat, will 
chase all her companions with true canine hostility. 

412 Lectures on Human and Animal Psychology 


When the feeling of individual inclination combines with the 
sexual impulse, we have the phenomena of animal marriage. 
We can only speak of marriage when the union of male and 
female for the fulfilment of the sexual functions is a permanent 
one, demonstrably based upon individual inclination. There is 
no trace of it among the invertebrates or the lowest vertebrates. 
Although the ' insect-states ' are really extended families, there 
is no proof that their individual members know one another as 
such, or are held together by any permanent mutual inclination ; 
indeed, the facts enumerated in Lecture XXIII. make such a 
hypothesis exceedingly improbable. 

On the other hand, marriage is a very common phenomenon 
among birds and mammals. That our domestic animals furnish 
so many exceptions to the rule is probably the result of domesti- 
cation. By close association with man, the animal loses touch 
with its kind. Most animals are monogamous, although poly- 
gamy is a well-known institution among birds. Polyandry does 
not appear to have been observed in animals ; it is confined to 
certain savage tribes. 

We have many reliable observations to show that with many 
birds the marriage contract is a matter of free choice. Males 
and females that are kept together in a cage by no means always 
pair. There are preferences and aversions shown, for reasons 
often inexplicable to ourselves. Male song-birds contend for 
the females in song. The birds of paradise are said to spread 
their gorgeous plumes till the female chooses the wooer that 
pleases her best. Animals of a fiercer disposition do not get 
through their wooing so peaceably ; the males generally have to 
fight pretty vigorously for the object of their choice. Lions and 
tigers wage bloody war for the possession of a spouse ; and stags 
will wound one another to the death in their struggle for a doe. 
The males of polygamous species are especially ferocious in 
this quarrel for wives : you know that two cocks in the same 
yard are impossible. In this matter of choosing a mate, peace- 
ably or otherwise, recourse is always had to the special weapons 
and peculiar ornamentation that so frequently characterise the 
male animal : think of the antlers of the deer, the spur of the 

Animal Marriage 413 

gamecock, the tusk of the boar, the lion s mane, the varied 
plumage of many birds. All the birds that breed with us, 
pies, storks, swallows, sparrows, doves, and what not, are mono- 
gamous. The nest is nearly always the family residence ; male 
and female build it together, and share in the tending of eggs 
and young. Only the swallows have separate nests for male 
and female. Besides the common fowl, the ostrich and the 
cassowary are polygamous. 

The marriage-relation among animals takes on a different 
form in monogamy and polygamy. The cock looks after his 
hens, seeks food for them ; they follow his call. But the hen 
does nothing for the cock except obey him. On the other hand, 
she watches, feeds, and protects the chicken, while the cock does 
not trouble himself about them. In monogamy it is generally 
otherwise. A pair of pigeons share all there is to do between 
them. Male and female take turns in sitting, and both alike see 
to the feeding of their young. It is obvious that these differ- 
ences depend upon differences in the feelings of individual 
attraction ; and we shall therefore be right in explaining by 
their aid the difference between the monogamous and poly- 
gamous form of the marriage-relation. 

The stability of animal marriage seems in general to be pro- 
portional to affection for the young. And this again becomes 
stronger the more careful and lasting attention the brood 

o o 

requires. A secondary reason for the continuance of the union 
after the young have ceased to require care is the need of 
mutual aid and protection. This holds especially of animals 
which construct nests or lairs, or which live in holes. So far, 
therefore, animal marriage is intimately connected with the 
conditions of specific physical organisation. But it would be 
unwarrantable to ascribe all its phenomena to this source. 
Individual inclination certainly determines choice in animals as 
well as in man ; accidental contiguity is a second and different 
cause. If there is an intimate connection between the fulfilment 
of a mental impulse and the satisfaction of a physical necessity, 
that is no more than we find everywhere else in our investigations 
of life and mind. 

414 Lectures on Human and Animal Psychology 


In marriage individuals are held together by the feeling tha\ 
they belong to one another. If this feeling is extended so far 
as to embrace a large number of animals, we have an animal 
society. Most birds and mammals tend to unite in flocks or 
herds. Domestication may overcome this tendency ; but it is 
seldom absent in the wild or feral state. Even dogs that have 
run wild not infrequently get together into a pack. Our oxen 
and sheep have retained the impulse to social life even under 
domestication. Many animals herd only for a particular purpose, 
especially for plunder or food ; but even then the herd is very 
often composed of a definite group of individuals. Migrant birds 
flock only when about to migrate ; the passage is made by 
thousands together. At its conclusion the individuals separate, 
to reunite in the following autumn. In the meantime the only 
connexion that the members of the same flock have with one 
another is that of locality ; they generally settle down near one 
another. A flock of daws likes to settle, if possible, all together 
in the same ruin ; storks of the same flight nest in neighbouring 
villages. There seems to be evidence in all these cases that the 
primitive feeling of inclination which brings the members of a 
species together at the very lowest stages of animal life is rein- 
forced by individual inclinations, though there may be but few 
in every flock or herd that are held together by them. 

We are taken a step further by animals which construct 
interconnected lairs or holes, destined to contain not one family 
only, but the progeny of a whole colony. This inclination is a 
direct outgrowth from the impulse of the individual to build. 
The otter tends to settle down in the neighbourhood of other 
otters. The same is true of the hamster and the beaver. Some- 
times the partitions between nest and nest are broken down, 
and the whole system thrown open ; as is the case with the 
hiding-places of rats and mice. 

These forms of social union are common enough. A special 
place among them must be assigned to the so-called insect-states ; 
' so-called,' for they are not really ' states.' The expression 
applied to these animal communities has done more to mislead 
than to explain. It has led to the assumption that all their 

Animal Societies and States 415 

phenomena are to be interpreted in terms of those of human 
governments and institutions ; which in its turn tempted ob- 
servers to parallel the division of labour in these societies, 
conditioned by the facts of physical organisation, with class 
distinctions in human society, and so to explain their observa- 
tions by reading into them their own thoughts and feelings. 
We have already had illustrations of this method of procedure 
in our review of ant-life (Lecture XXIII., p. 343). 

Insect-states are really extended families. The dwelling- 
places of the colonies are nests of a more or less complicated 
structure, according to the size and composition of the society. 
In most cases those orders of animals which contain the species 
that live in states include others which have not carried their 
social life farther than the stage of simple nest-building. In 
some species of wasps, e.g., the digger-wasps and solitary wasps, 
the males and females live separately, though the female digs a 
hole in the mortar or wood of a wall, in which she lays her egg, 
putting in small caterpillars along with it, to serve as food for 
the newly hatched larvae. The nest of the common wasp is more 
-extensive. In the spring the female builds in a tree, on a roof, 
or in the ground a few hexagonal cells of vegetable material, 
lays an egg in each of them, and feeds the newly hatched larvae 
until they crawl out. After this the young assist in the work 
of building, and a nest is gradually constructed, the female 
depositing an egg in every new cell. The females that develope 
at this period are themselves incapable of laying eggs ; their 
whole energy is consumed in the business of nest-building, and 
their sexual organs remain immature. These sexually unde- 
veloped females are accordingly called workers. Not till towards 
the end of the summer are eggs laid which produce males and 
perfect females. These males fertilise the females in the autumn. 
When cold weather sets in they and the workers die ; but the 
females survive the winter, and at the return of spring begin to 
make nests and lay eggs. The female solitary wasp generally 
begins her work in some narrow hole in a wall, into which she had 
crept for the winter ; and the colony issues forth later, when the 
space has grown too small for it, to build a larger nest. What 
is true for the wasp holds also for the humble-bee (botnbus), a 
relative of the ordinary honey-bee. The female is fertilised in the 

416 Lectzires on Human and Animal Psychology 

autumn, survives the winter, and begins an underground nest in 
the spring, in the building of which she is assisted by the females 
or workers which are first hatched. Towards the end of the 
summer sexually mature insects make their appearance ; and 
with the coming of winter the whole colony perishes, with the 
exception of the females which seek shelter underground. 

There are two features in particular of these communities of 
wasps and humble-bees which for a long time defied any attempt 
at interpretation : the presence of sexless workers among perfect 
males and females, and the constancy with which the appear- 
ance of the latter two forms was restricted to the end of the 
summer. The first problem was solved so soon as it became 
known that the workers were not really sexless, as had been 
supposed, but simply immature females. This arrest of develop- 
ment could easily be explained from the expenditure of force 
necessary for building the nest ; and experiment showed that a 
more abundant supply of food did really suffice to change the 
workers into ordinary females. The second problem was 
answered by the discovery, first made in the case of the honey- 
bee, that the laying of male or female eggs depends entirely on 
the nature of their fructification by the female herself. After 
fertilisation by the male, the female retains the injected seed in 
a small pocket opening into the canal by which the eggs escape. 
This arrangement is of immense importance, because in these 
insects all eggs, even those which are not fertilised by the male, 
are capable of development. The fertilised eggs produce female, 
the unfertilised male insects. It is now plain why the humble- 
bee and the wasp at the beginning of summer lay only eggs 
which develop into females : the female fructifies her eggs as 
long as she retains any of the seed which she received from the 
male the autumn before. When this store is exhausted, the eggs 
can only produce males. But even of the fructified eggs it is 
only those last laid that can become perfect females, because it 
is only after the completion of the nest and the production of a 
sufficient number of workers that the larvae are well enough fed 
to attain to their complete development. So that what looked 
at first sight like a preconceived design in these simplest insect- 
states proves to be the necessary result of physical organisation 
and of the relatively simple instincts which accompany it. 

Animal Societies and States 417 

Taking the wasp as our guide, we shall not, perhaps, find it 
so very difficult to explain the organisation of the bee-state. 
The female bee, the ' queen ' as she is called, also lays fertilised 
and unfertilised eggs. But she lays both kinds from the first, and 
distributes them among the cells of the hive, which the workers 
have built from wax of their own production. The cells are of 
two kinds, wide and narrow. The wide are for the unfertilised 
e gg s > which develope into males or drones ; the narrow for the 
fertilised, which develope into workers. Besides this, the queen 
lays a few fertilised eggs in specially wide cells. The larvae from 
these are fed more abundantly than the rest ; they become per- 
fect females or queens. Sometimes the workers will take a larva 
from an ordinary cell into a royal cell which is not quite finished ; 
then by means of good nourishment it becomes a queen. In 
the spring, as soon as the brood of queens is beginning to- 
approach maturity, the hive becomes restless ; and on the first 
fine day a part of its inhabitants swarms out in quest of a new 
abode. This first swarm is quickly followed by others ; so that 
a single hive may found several colonies in the course of the 
summer. The old queen always goes with the first swarm, 
leaving the hive before the brood of new queens are out of their 
cells. The first of these latter to appear remains queen of the 
hive ; the others fly off with a portion of the workers to found 
other colonies. If two queens of the new brood make their 
appearance simultaneously, they fight till one or the other is 
overcome and killed, unless she avoids her danger by swarm- 
ing out in time. The hive, therefore, never contains more than 
one sexually mature female, though the number of drones is 
very various, ranging from none at all to nearly a thousand. 
The drones are not confined to the limits of their hive. In the 
spring they fly out on every warm day and meet the young 
queens. But in the autumn, as soon as provisions become scarcer, 
they are expelled by the workers, and perish on the first cold 

What distinguishes the bee's hive from the societies of wasps, 
hornets, and humble-bees, therefore, is a more hard and fast 
division of labour. In containing only one female, the hive re- 
sembles the nests of these other insects. But it is essentially 
different from them in its mode of origin. The wasp's nest is 

E E 

418 Lectures on Human and Animal Psychology 

begun by a female, so that her solitariness is a matter of course. 
But the bee's hive is a society from its foundation, a society 
which grows, but without undergoing any radical change. The 
solitariness of its queen is partly the result of force. But it is 
this very interconnection of bee-states, the fact that each is a 
colony from some pre-existing one, that enables us to under- 
stand the mode of origin of the bee-societies and their difference 
from the associations of related insects. The natural history of 
the building of every wasp's nest is just a repetition of the same 
processes. But the bee-state stands in connection with its parent 
state, that again with its own, and so on. It has, in other words, 
a historical relation to the past and the future. If we suppose 
that there occur in such a society expressions of the universal im- 
pulse towards imitation, it is a necessary corollary that a colony 
will not have to begin its life from the beginning, but carries to 
its new home the customs acquired by previous generations, 
whether these are transmitted in inherited organic dispositions 
or are perpetuated by being directly handed down from the 
older insects to the younger. But it would be altogether gratu- 
itous to assume that the organisation of the hive has always 
been what we find it to-day. We know from experience that 
the habits of animals may change. One can wean the domestic 
insects from swarming and founding colonies by enlarging 
their hive as circumstances require. Populous bee-states will 
now and then give up the work of collecting honey, and take to 
plundering the smaller hives in their neighbourhood. And if 
we see changes like these in the habits of animals going on 
under our eyes, there is nothing to prevent our concluding 
that the peculiarities of the bee-society have arisen gradually 
and slowly, and that its customs have been fixed and settled 
both by inherited physical dispositions and by imitation. This 
conclusion is all the more probable in that the mode of 
origin of the beehive of to-day contains indications that its 
primitive mode of origin was something different. The first social 
union of the insects, you see, cannot possibly have branched off 
from any pre-existing society. How could it have been brought 
about ? 

The question is answered by the condition in which we still 
find certain of the insects that are most closely related to the 

Animal Societies and States 419 

honey-bee. Every female wasp founds her own family ; every 
female bee must originally have founded her own family in the 
same way. Worker and queen at once, she prepared by her- 
self the first cells for her brood. Now an alteration in these 
conditions may have been brought about by the greater length 
of life of the bee-communities. When more than one female had 
appeared in a single hive, jealousy made any peaceable common 
life impossible ; death and exile were the only alternatives for 
the weaker faction, and the latter would have been already 
suggested whenever the crowded condition of the nest pre- 
vented any further increase of population. So far everything 
is intelligible. But how is it that the queen voluntarily de- 
posits drone-eggs in the wide cells, and worker-eggs in the narrow 
ones, and that the workers kill the royal larvae if the weather 
is unfavourable to swarming ? These customs, too, we have 
every reason to suppose, are matters of gradual development, 
products of the natural evolution of instinct. The size of the 
cells in which the larvae develop, e.g., would have to be settled 
by reference to their wants. At first all the cells might have 
been made of one size. It would soon be found that the more 
poorly nourished larvae, which were destined to become workers, 
required less space than those which turned into queens or 
drones. When once the right size had been hit upon, it might 
be adhered to in the future, since the bee-state is in touch with 
past traditions which lay down rules of conduct for its 
members. The younger generation had only to follow the 
example set them by their elders. For this reason the bee- 
state never needs to return to the primitive stage, and model its 
organisation from the very beginning. There is no exaggeration 
in saying that it is based, like our own civilised states, upon the 
work of all preceding generations. 

Ant-communities differ from those of bees chiefly in the number 
of females which they support. For the greater part of their 
lives the males and females are winged ; and they are larger 
than the wingless neuters which make up the great bulk of the 
population. These neuters, like the worker-bees, are immature 
females. With the ants, too, the division of labour seems at 
times to extend even to the workers : especially with the ter- 
mites, or white ants of Africa and Southern Asia. These anta 

420 Lectures on Human and Animal Psychology 

build hills which often attain the height of several feet. Their 
workers are of two classes, the workers proper, to which the 
peaceful avocations of the colony are entrusted ; and the soldiers, 
whose duty it i? to attack strange nests or defend their own 
from attack. This difference in instinct is probably correlated 
with difference in the physical strength of the individuals. And 
everything that we know of the intellectual capacities of these 
insects would lead us to suppose that the division of labour is 
not consciously agreed upon. A very similar instinct is dis- 
played by the Amazon ant, which carries off the larvae from the 
nests of weaker species, and makes workers or ' slaves ' of them. 
This instinct is rooted in the general aversion that the different 
species of ants manifest towards one another, and has been 
gradually developed from the mob- fights in which the feeling 
of mutual dislike often culminates. Another specific instinct of 
ants is the custom of keeping plant-lice as ' domestic animals,' 
for the purpose of feeding themselves and their larvae from 
the liquid secreted in the abdomen of these animals. There is 
nothing strange in such an expression of the nutritive im- 
pulse : the plant-louse, being one source of food among others, 
would naturally be carried with the rest into the ant-hill. 

The phenomena presented by these animal states can only be 
seen in their proper light, if we keep in mind at the same time 
the mental capacities of the individuals which compose them. I 
have already pointed out to you that the exaggerated ideas 
of the early bee and ant naturalists as to insect intelligence 
must be considerably modified in view of the results of obser- 
vations made under careful experimental conditions. The 
members of a bee- or ant-community do not know one another 
individually. And the feeling of inclination which holds them 
together is of a collective, indefinite nature, standing on a far 
lower developmental plane than the analogous feelings of birds 
and mammals, which lead to marriage or to the formation of 
less extended associations. The power of communication is 
also extremely limited, confined in all probability to certain 
manifestations of the imitative impulse. Numerous proofs ot 
the comparatively low development of the mental life of the 
individual in these insect-communities have been collected by 
Sir John Lubbock, to whose work upon ants, bees, and wasps 

Animal Societits and States 421 

I must refer you, in the absence of any observations of my own. 1 
His investigations show clearly enough the immense advantage 
of experiment in this field over simple observation. Lubbock ap- 
proached every single question with pre-conceived ideas, derived 
from observation of the general results of instinct and naturally 
tending towards an overestimation of the intellectual capacities 
of the insects. But experimental tests always gave the same 
result, that the impulses of the common instinct left hardly any 
place at all for the exercise of individual intelligence or the ex- 
pression of individual feelings of inclination. And even 
Lubbock's conclusions require one further limitation ; this con- 
cept of intelligence still plays far too great a part in his 
pages. The very modest performances which he ascribes to 
intelligence are entirely explicable in terms of comparatively 
simple associations. And that implies that the feelings and 
impulses operative in the instinctive actions of the insects in 
question are of an extremely primitive kind. So that when 
we talk of their having feelings of inclination and aversion, or of 
their impulse towards imitation, we must be careful not to 
regard these feelings and impulses as identical with the analogous 
processes of our own consciousness, still less with these pro- 
cesses plus the products of our reflection upon them. We have 
before us no more than the first obscure movings of feelings and 
emotions, which we do not find in their clearly conscious form 
till we reach the higher animals or even man, but which for 
that very reason act with all the greater certainty and uniformity 
at this low level of development. We are fatally inclined to 
make the same mistake with regard to the elementary psychical 
factors, feelings and impulses, which lead to the formation of 
animal societies that we make in foisting our own point of 
view upon their complex results, these communities themselves. 
We talk of the organisation of insects into a state, of queens and 
workers, of soldiers and slaves, even of the rearing of domestic 
animals. And so we tend to read into their loves and hates, 
acts of succour and of imitation, conscious processes completely 
analogous to those which the terms call up in our own minds. We 

1 Ants, Bees, and Wasps : a Record of Observations on the Habits of 
the Social Hymenoptera, by Sir John Lubbock, Bart., M.P. (Int. Sci. 

422 Lectures on Human and Animal Psychology 

must remember that we are really in face of very primitive 
forms of mentality, which may be every bit as different from its 
more highly developed stages as is a single cell from a complex 

But if we are always obliged to measure the animal mind 
by the standard of our own consciousness, applying this as best 
we can where the conditions are so different, the other side 
of the matter is not less important. We must look into these 
facts of animal psychology for light upon the phenomena of the 
human mind. Another fatal tendency on the part of the 
psychologist is to measure every human action by the highest 
standard applicable to it. We look at it from the standpoint of 
intellectual reflection, and then make this reflection, our own 
affair entirely, the condition of its origin. Man lives in wed- 
lock ; he combines with his fellows to form a community ; he 
founds states. All this as he does it presupposes an immense 
sum of intellectual work, accumulated through countless genera- 
tions and implying the development of the higher feelings. 
In every particular case of human action this accumulated store 
is drawn upon. But it is surely wrong, in the light of the 
instances which the animal kingdom furnishes of the manifes- 
tation of social impulses, that a part played in the constitution 
of human society by original, natural impulse should be so 
entirely overlooked as it not infrequently is. Why, even in man 
it is only the special development which the phenomena have 
undergone, not their existence or their origin, which is the result 
of civilisation. The witness of animal psychology tells with 
all possible directness for the naturalness of the first beginnings 
of human social life. The investigation of the interaction of the 
two factors, nature and civilisation, in their gradual develop- 
ment forms the subject-matter of other disciplines upon which 
we cannot enter, social psychology and social science. 




WHEN we were considering will in its significance as an 
elementary psychical phenomenon, we found that the 
facts comprehended under the term constituted the 
links in a chain of development. The lower stages of this de- 
velopment, simple voluntary acts, were classed together as 
manifestations of impulse ; the higher stages, acts of choice, as 
those of volition proper. In reviewing the expressions of in- 
stinct we have become familiar with a whole number of pheno- 
mena whose invariable mental condition is some impulsive act, 
while at the same time the peculiarities of the physical organi- 
sation exercise a determining influence upon their development. 
It now remains to consider briefly the second and higher form of 
voluntary activity, volition proper, in its relation to the entirety 
of conscious processes. 

We took our best examples of instinctive action from the 
animal kingdom. In the present investigation of volition, on 
the other hand, we are exclusively restricted to the human con- 
sciousness, although it is certain enough that instances of 
volitional action are not infrequent in the animal world, and 
especially among its more highly organised members. But the 
problem of volition, or, as it is generally called in consequence 
of the popular restriction of the concept of will to the sphere of 
choice, the problem of will, is practically confined to man, for 
this reason, that the one question which is of decisive import- 
ance for our understanding of the nature of voluntary action and 
its relation to the other facts of our inner experience, a ques- 

424 Lectures on Hitman and Animal Psychology 

tion which has long divided psychologists and philosophers 
alike into two hostile camps, is one that must be answered by 
an appeal to our own minds. It is the question of the causality 
of will. 

An impulsive action is one, as we have seen, which is univo- 
cally conditioned ; there is only one motive present in conscious- 
ness. Volitional action arises from the choice between different 
motives, clearly or obscurely conscious. In impulse, therefore, 
the feeling of our own activity is less developed than in volition ; 
whilst, since this latter involves a decision as between various 
conflicting motives, the feeling of our own activity rises in it to 
that Q{ freedom. 

But if freedom is a result of the possession of will, of the 
choosing or selective will, how does it come about that the re- 
lation of the two is so often transposed ? Instead of saying, ' I 
am free, for I can will,' we are apt to say, ' I can will, for I am 
free.' Is not this a confusion of cause and effect? It is plain 
enough that our consciousness of freedom can only have its 
source in the power of willing. The prisoner is not free, be- 
cause his will is without effect. He would gladly be out of 
prison : but that is wishing, not willing. A firm belief in our 
power to do is an indispensable condition of willing, which is 
just the decision to act. How, then, are we to explain the fact 
that the consciousness of freedom, whose root is in the will, thus 
denies its origin, and makes itself out to be the cause of that 
from which it has really resulted ? 

We know that we are free when we act of our own power, 
unimpeded by any external obstacle. Action by our own power 
we term volitional action, and regard as the consequence of our 
freedom. But what do we suppose to be the cause of this free- 
dom ? There appears to be at this point a sudden break in the 
chain of cause and effect. We say the very concept of freedom 
excludes any idea of causality. For if it were dependent upon 
some cause or other, it would cease to be what it is, freedom. 
Freedom and necessity mutually exclude each other. 

Notice now the steps by which we have arrived at this con- 
clusion. We should not be justified in saying the very concept 

The Causality of Will 425 

of will excludes any idea of causality. For the fact that we do 
not know all the causes of a volition cannot be regarded as 
necessarily implied in the concept. What is done, then, is this. 
Freedom, the concept of which excludes causality, is interpolated 
as a middle term, on the one side of which volition is subject to 
causality, while on the other it is independent of it. For it is 
now subjected to a special causality, the causality of freedom, 
while made independent of general causality, the causality of 
natural processes. 

It is this view of the matter which has given rise to the con- 
flict between ordinary determinism, which maintains the uni- 
versal validity of the law of causation, and indeterminism, which 
postulates freedom. ' The will cannot be free,' says the deter- 
minist, 'for a free will would not accord with the actual causal 
connexion of world- processes. Natural law would be replaced 
by miracle. No! every action, however free it seems, must have 
its cause. It is a necessary occurrence, and the agent cannot 
help himself.' ' The will is free,' replies the indeterminist, ' for 
we have an immediate consciousness of its freedom. Natural 
necessity and personal freedom are opposites. But the latter is 
vouched for by the inner voice of conscience, requiring from 
the agent responsibility for his every action.' 

The opponents of the freedom of the will, that is, assert that 
its assumption is nonsense ; its advocates maintain that it is 
necessary. Which party in the dispute is right ? 

We must insist, in the first place, that all the ethical arguments 
which have been brought to bear upon the question of the free- 
dom of the will are out of place. They rnay move us ; the}- may 
incline us to the hypothesis of the freedom of human volition : 
they cannot prove anything. Even if a denial of the freedom of 
the will imperilled the validity of conscience and shook the 
foundations of our whole ethical system, still, if clear proof 
could be adduced that the will is not free, science would have to 
take its course. But happily that is not the case. Whichever 
theory holds the field, practice may stay quietly at home. You 
may remember what Kant said : ' Every being who can act 
only under the idea of freedom is in his action really free ; that 
is, he is governed by all the laws which freedom would neces- 
sarily bring with it, just as really as though his will were proved 

426 Lectures on Human and Animal Psychology 

to be free to the satisfaction of theoretical philosophy.' The 
undeniable fact that we have a consciousness of freedom makes 
fatalism impossible, unless, indeed, this consciousness itself be 
regarded as included in the universal causal nexus. For this 
consciousness of freedom tells us that we have the power to act 
without being consciously impelled by any constraining force, 
external or internal ; it does not tell us that we act without a 
cause. The defenders and the opponents of the freedom of the 
will have not seldom been at one in their confusion of constrain- 
ing force and cause. Really, the two are wholly disparate con- 
cepts. We cannot say the earth is constrained to move, but we 
can say man is constrained to die. Only a being who knows 
that he is free can be constrained. The fatalist makes the mis- 
take of destroying freedom and putting constraint in its place, 
constraint being in actual fact a condition which arose out of 
freedom, and cannot be conceived of without it. 

So that if we take the concept of freedom in its proper sense, 
we shall say, ' The will is free,' for everything that stands in the 
way of a purposed voluntary action is felt by consciousness as 
constraint, while will seems to it the very opposite of this con- 
straint. Freedom and constraint are reciprocal concepts ; they 
are both necessarily connected with consciousness ; outside of 
consciousness they are both imaginary concepts, which only a 
mythologising imagination could relate to things. If we say, 
' The earth is subject to constraint because it moves round the 
sun,' we might just as well go on to assert that the sun is free, 
because it moves the planets. 

Herbart remarks somewhere: ' If we regard ourselves as not 
free, we are really not free ; but if we ascribe freedom to our- 
selves, it by no means follows that we are so in reality.' We 
may say with equal justice : ' If we know the cause of a pheno- 
menon, it necessarily follows that this really has a cause ; but if 
we do not know the cause of a phenomenon, it by no means 
follows that it has no cause.' But it is this last and erroneous 
inference which the adherents of absolute indeterminism draw 
when they conclude, from the premise that we cannot discover 
in consciousness all the causes which determine the will, that the 
will itself is the first cause of our actions. 

It was attempted to support this negative proof from con- 

The Causality of Will 427 

sciousness by a further positive argument. In nature, we are 
told, every occurrence presupposes a previous condition of things 
of which it is the inevitable consequence. This previous con- 
dition must itself have a predecessor, and so forth. But for the 
beginning of this infinite series we must postulate a primary, 
spontaneous impulse, if the origin of the world is to become in- 
telligible at all. Now if it is once shown that one point stands 
outside of the universal causal nexus, there is no difficulty in 
conceiving of any number of causally connected series arising in 
the progress of the world's development, and each possessing 
its own particular beginning. If I now undertake the perform- 
ance of some voluntary action, this fact, with all its consequences, 
means the beginning of a new series, each term of which is 
determined by natural causes except the first one, which is be- 
yond their reach. 

There are two weak points in this argumentation. In the 
first place, the assumption of a first beginning of things seems to 
be impossible for consciousness, whether pictorially or concep- 
tually represented ; and secondly, even if a first beginning of 
the world had to be assumed, the hypothesis that similar be- 
ginnings could take place in the midst of the course of the 
world's development would be an analogical inference, destitute 
of all positive foundation. 

The fundamental error in these and other arguments for or 
against the freedom of the will goes deeper. It consists in con- 
sidering the entire question simply under the concept of natural 
causation. The very first requisite is a treatment of it as a 
question of psychological experience. If we regard it from this 
point of view, we see at once that the psychical causes, whether 
of a voluntary act or of any other manifestation of consciousness, 
are never wholly discoverable, for two reasons : first, because tht-y 
lie outside consciousness, and belong to an inaccessible series of 
past experiences ; and secondly, because they form part of a more 
general conscious nexus, of which the individual mind constitutes 
only one link. The general direction of the individual will is, 
you see, determined by the collective will of the community in 
which its possessor lives. And it is particularly in this connec- 
tion that we find reason for the belief that the causality of our 
mental life cannot be subsumed without more ado under the 

428 Lectures on Human and Animal Psychology 

familiar laws of natural causation, such as that of the equivalence 
of cause and effect. 


An attempt to construct the history of a nation or of mankind 
at large in terms of the laws of natural causation would not 
only be vain in practice : it would be wrong in principle. If the 
individual can say that, in place of acting as he did in some par- 
ticular case, he might have acted otherwise, we must also be able 
to say of every event in history that it might have happened 
differently. In both cases the necessity of natural causation is 
wanting. For historical events and for the voluntary actions of 
an individual we can only adduce determining motives ; we 
cannot prove constraining reasons. In this regard the concepts 
of historical occurrence and of voluntary action are exactly 
equivalent. The only difference is, that one refers to a com- 
munity, the other to an individual. 

The general will of a community consists simply in the ex- 
pressions of the wills of a large number of individuals. The 
individual and his voluntary action are enclosed within concen- 
tric circles of more and more general volition. First comes the 
general will of the little community in which he most immedi- 
ately belongs ; then he, with this will, is subject to the will of a 
larger community ; with this, again, to a still more comprehensive 
will ; and so on. The relations in which the individual is thus 
placed are the principal determinants of his voluntary actions. 
But the general will of a community is usually in its turn 
determined by the wills of the more energetic individuals, which 
are acquiesced in by the individual wills of the majority. 

It is a rule written upon the face of history that the frequency 
of expressions of volition is inversely proportional to the mag- 
nitude of their effects. National action by which the course of 
history is suddenly changed is a matter of rare occurrence. 
Events which we can refer to the action of the general will of 
considerable communities constitute, as it were, the milestones 
of history. In the intervals between them the general will is for 
the most part inactive ; though there are changes occurring 
within the community, oscillations in this direction or in that, 
they are not of vital import: they are like the variations of an 

Relation of the Individual to tke General Will 429 

individual will in obedience to the impulses and emotions to 
which a man's manner of life exposes him. The determination 
of the general will by those of a few prominent individuals has 
given place to its direction by a crowd of hardly noticeable in- 
fluences, affecting each and all alike, directly or indirectly, by 
way of external condition or internal modification. 

The principal determinant of the individual will is, as we have 
seen, the will of the community. In stirring times the course of 
events carries the individual with it, while in those periods of 
history when the general will is inactive the community remains 
in what we may call a state of equilibrium. But the social con- 
dition resulting from previous history, from external natural 
causes, and from the intervention of particularly strong indivi- 
dual wills in the ordinary progress of things, must, of course, 
itself contain motives of determining influence upon the voluntary 
action of the individual ; so that it is only to be expected that in 
the long intervals elapsing between historical events of the first 
magnitude the practically constant condition of society will bring 
with it a certain uniformity in the voluntary actions of the 
individuals composing it. 

This general influence is confirmed by statistical facts. We 
find that the annual number of crimes, suicides, and marriages 
may remain constant for decades together, in civilised countries 
where the condition of society resulting from their past history 
is also approximately constant. Ouetelet showed that the num- 
ber of marriages every year is more regular even than the 
number of deaths, to which, of course, except in cases of suicide, 
the will has nothing to say. The same statistician proved also 
that so long as the course of justice, the prosecution and punish- 
ment of crime, remain unaltered in any nation, the crimes 
committed show a marvellous constancy in number, character, 
and distribution with regard to age and sex. And the same 
regularity obtains for suicide. It extends even to the manner 
of death chosen. Every year approximately the same number 
of men hang, shoot, poison, and drown themselves. From all 
this constancy we cannot but conclude that the historically de- 
termined social condition of a people is a dominant influence in 
the voluntary actions of the individual citizen. 

And our conclusion finds still further confirmation in observa- 

43 Lectures on Human and Animal Psychology 

tions of a different nature, which afford us the means of isolating 
certain of the factors which combine to constitute the state of a 
society. If we compare the slight deviations from absolute 
regularity which the statistical tables show with the relations 
which help to determine that state, we are able in some measure 
to trace them to their causes. Thus it is demonstrable that 
famine increases the number of crimes against property and 
decreases the number of marriages. Violent epidemics, like 
cholera, bring with them a temporary decrease in the number of 
marriages, followed shortly after their disappearance by a still 
more marked increase. This latter phenomenon is to be 
ascribed to the increased mortality occasioned by the epidemic. 
Society seems to be hastening all unconsciously to fill up the gaps 
that death has made in its ranks. However irregular the actions 
of the individual, those of the community present the completest 
uniformity. But this regularity appears as the product of a blind 
necessity. Actions of every kind follow a definite numerical law, 
which no volition of the individual can avail to change. 

But if in this summation of individual actions there is no trace 
of anything that could be ascribed to the influence of an indivi- 
dual will, are we not bound to conclude that this influence is 
illusory? Is not the exception to natural law only an apparent 
one, which disappears when our observations extend over a 
sufficiently wide field ? Yes ; this conclusion has been drawn. 
The statistical figures prove, it has been said, that voluntary 
actions are dependent in measurable degree upon a series of 
external factors. Will within us, that is. corresponds to acci- 
dent in the natural world without. Neither is a phenomenon 
without laws ; but both are phenomena whose laws cannot be 
deduced from the particular instance. In this way, it has been 
thought, the problem of the freedom of the will is solved by 
appeal to experience ; and the solution is determinism. 

But there is nothing in the facts of statistics to warrant such 
a conclusion, in the remotest degree. They simply show that 
the influence exerted by the condition of society constitutes 
one of the causes which determine the will. Whether it is the 
only cause, or whether there is not a whole number of co-ordinate 
causes to be found elsewhere, on those questions they have noi 
a word to say. 

Relation of the Individual to the General Will 431 

In extending our observations from the individual to a large 
community, we eliminate all the causes which condition the in- 
dividual alone or only some small section of the community. 
It is the same procedure as is employed in physics. To elimi- 
nate chance influences which might vitiate the result of an 
observation, a large number of observations are taken. The 
more observations there are, the more probable it becomes that 
the separate sources of disturbance, which work in both directions 
as plus and minus, will compensate one another ; so that the 
average of the whole number will give us a result in accordance 
with the real fact under observation. But when we argue that, 
because statistics enable us to cancel out the influences that 
are restricted to the individual, therefore these influences do not 
exist, that is as bad as it would be to say in physics that the 
accidental errors, eliminated in the total number of observations 
were not present in the particular case. The physicist can 
afford to neglect them, simply because they possess no signifi- 
cance for him ; the psychologist cannot. The question before 
him is whether there exist, in addition to the influence exerted 
by the social state of a community, further determinants of 
volition of a more individual character. He must not neglect 
the deviations from the norm shown in the particular case; for 
their presence constitutes the proof that such secondary deter- 
minants really exist. 

Statistics itself teaches us that the effect of individual condi- 
tions determining voluntary action can really be traced in 
different degrees in the different circles of a community. The 
number of crimes, suicides, and marriages varies with age, sex, 
income, profession, etc. As soon, that is, as statistics goes more 
into details, it points to influences of a more special kind, de- 
pending upon the special nature of the state of society in that 
particular circle of the community. Still the utmost t