SOCIOLOGY
AND PHILOSOPHY
by
EMILE DURKHEIM
Translated by
D. F. POCOCK
With an Introduction by
J. G. PERISTIANY
Senior Lecturer in Social Anthropology at the
University of Oxford
COHEN & WEST LTD
30 PERCY STREET, LONDON, W. 1
1953
COPYRIGHT
COPYRIGHT
ISBN 0-203-09236-8 Master e-book ISBN
This edition published in the Taylor & Francis e-Library, 2009.
To purchase your own copy of this or any of
Taylor & Francis or Routledge’s collection of thousands of eBooks
please go to www.eBookstore.tandf.co.uk.
CONTENTS
 
   
 INTRODUCTION
  ix
 TRANSLATOR’S NOTE
  xxiv
 PREFACE TO THE ORIGINAL EDITION
  xxv
 
I
 
INDIVIDUAL AND COLLECTIVE
REPRESENTATIONS
 
1
II   THE DETERMINATION OF MORAL FACTS   16
III   REPLIES TO OBJECTIONS:   32
   
1.   The Condition of Society and the Condition of
Social Opinion
  32
    2.   Individual Reason and Moral Reality   33
   
3.   The Feeling of Obligation: the Sacred Character
of Morality
  35
    4.   The Moral Authority of the Collective   37
    5.   Philosophy and Moral Facts   38
    6.   The Subjective Representation of Morality   40
IV
 
VALUE JUDGMENTS AND JUDGMENTS OF
REALITY
  42
I
TIONS1
IF analogy is not a method of demonstration in the true sense of the word, it is nevertheless
a method of illustration and of secondary verification which may be of some use. It is
always interesting to see whether a law established for one order of facts may not, mutatis
mutandis, be found to apply elsewhere. This comparison may also serve to confirm it and
give a greater understanding of its implications. In fact, analogy is a legitimate form of
comparison, and comparison is the only practical means we have for the understanding of
things. The fault of the biological sociologists was not that they used it but that they used
it wrongly. Instead of trying to control their studies of society by their knowledge of biology,
they tried to infer the laws of the first from the laws of the second. Such inferences
are worthless. If the laws governing natural life are found also in society, they are found
in diferent forms and with specific characteristics which do not permit of conjecure by
analogy and can only be understood by direct observation. However, if one had already
by the use of sociological methods begun to determine certain qualities of social organization,
it would be perfectly legitimate to inquire afterwards whether these qualities did not
show some partial similarities with the animal organism as established by the biologist. It
might be assumed that all organisms must have certain characteristics in common which
are worth while studying.
It is, however, much more natural to look for analogies which may exist between the
laws of sociology and those of psychology, because these two cover neighbouring fields.
Like the individual, the collective life is composed of representations, and it may therefore
be presumed that collective and individual representations are in some ways comparable.
We shall in fact try to show that both maintain the same relations with their respective
substrata. Far from justifying the belief that reduces sociology to nothing more than a corollary
of individual psychology, this similarity will on the contrary set in relief the relative
independence of the two worlds and the two sciences.
1
The psychological conceptions of Huxley and Maudsley, which reduce the mind to nothing
more than an epiphenomenon of physical life, have no longer many defenders; even
the most authoritative representatives of the psycho-physiological school have formally
rejected them and endeavoured to show that such conceptions are not implicit in their
principles. The fact of the matter is that the cardinal idea of this system is purely verbal.
Certain phenomena have a limited efficacy—that is to say, they only affect surrounding
phenomena slightly; but the idea of an added phenomenon which has no purpose, which
does nothing, lacks all positive content. Even the metaphors which the theoreticians of
1
First published in Revue de Métaphysique et de Morale, vol. vi, May, 1898.
INDIVIDUAL AND COLLECTIVE
REPRESENTA
2  Sociology and Philosophy
this school employ most frequently to express their meaning turn against them. They say
that the mind is a simple reflection of the underlying cerebral processes. It is a light which
accompanies, but does not constitute, those processes. But a light is not a ‘nothing’, it is a
reality which testifies to its presence by its peculiar effects. Objects are not the same and do
not act in the same way according to whether they are in the light or not; light can even alter
their characteristics. In the same way the act of understanding, however imperfectly, the
organic process which one wishes to make the essence of psychic facts constitutes a new
reality which is not without importance, and which manifests itself by noticeable signs.
The more this faculty for understanding what happens within ourselves is developed, the
more the subject’s movements lose that automatism which is the characteristic of physical
life. The agent endowed with reason does not behave like a thing of which the activity can
be reduced to a system of reflexes. He hesitates, feels his way, deliberates, and by that distinguishing
mark he is recognized. External stimulation, instead of resulting immediately
in movements, is halted in its progress and is subjected to a sui generis elaboration; a more
or less long period of time elapses before the expression in movement appears. This relative
indetermination does not occur where there is no thinking mind, and with thought it
increases. It would appear, then, that the reason (la conscience) is not as inert as has been
supposed. Indeed, how could it be otherwise? All that is has its specific way of being and
has its peculiar properties. But every property expresses itself by manifestations which
could not occur had that property been other than it is, for it is by this behaviour that the
property is defined. Thus whatever name one may give to the reason, it has certain qualities
without which it would not be recognizable. Consequently from the moment that it exists
things cannot occur as though it did not.
The same objection can be put in the following form. It is a commonplace of both science
and philosophy that everything is subject to becoming. But change produces effects.
Even the most passive participates in the stimulus which it receives, if only by resisting. Its
life and direction depend in part upon its weight, its molecular constitution, etc. If, then, all
change implies a certain causal ability in what is changing, and if the mind once produced
is incapable of producing anything, we are forced to the conclusion that from the moment
it exists it ceases to be subject to change. It would stay as it is; the series of transformations
of which it is a part would stop with it, and beyond it there would be nothing more.
It would be in a sense the final stage of reality, finis ultimus naturae. It is not necessary to
point out that such a notion is untenable, for it contradicts all the principles of science. The
way in which representations are destroyed would become equally unintelligible from this
point of view, for any compound which dissolves is always, to a certain extent, a factor in
its own dissolution.
It seems to us useless to discuss any further a system which, taken literally, is contradictory
in its terms. Since observation has revealed the existence of an order of phenomena
called representations, distinguishable by certain characteristics from all other natural phenomena,
it is scarcely methodical to treat them as though they did not exist. Undoubtedly
they are caused, but they are in their turn causes. Life itself is nothing but a combination of
mineral particles; nobody, however, tries to make it an epiphenomenon of inoranic matter.
Once this proposition has been accepted we must also accept the logical consequences. But
it is a fundamental proposition which seems to have escaped several psychologists and we
shall endeavour to throw a little light upon it.
 
3
The reduction of the memory to an organic fact has become almost classical. The representation,
it is maintained, has no power of retaining itself as such. When a sensation,
image, or idea is no longer presented to us it ceases to exist, without leaving the slightest
trace. The organic impression which preceded the representation does not, however, disappear
completely. What remains is a modification of the nerve elements involved which
will predispose them to vibrate again as they vibrated on the first occasion. Subjected
to any further stimulus this same vibration will be reproduced; there results in the mind
the psychic state which appeared before, in the same conditions, at the time of the first
experience. The memory results from and consists in this process. That this renewed state
appears to us as a revivification of the first is then only an illusion. Indeed, if the theory is
exact it is an entirely new phenomenon. It is not the old sensation reawakened after lying
dormant for some time; it must be an entirely new sensation in so far as nothing remains
of the original one. If it did not by a recognized process locate itself in the past, we would
really believe that we had never before experienced it. The only thing which is the same
in the two experiences is the state of the nerves, the necessary condition of the second representation
as of the first. This theory is not maintained only by the psycho-physiological
school. It is also explicitly admitted by many psychologists who believe in the reality of
the mind and wish to see in mental life the highest form of reality. Thus Léon Dumont
says: ‘When we no longer entertain an idea it no longer exists, even in a latent condition.
Only one of its conditions remains and serves to explain how, with other conditions, the
same thought may be renewed.’A particular act of memory results ‘from the combination
of two elements: (i) a condition in the organism complemented by (ii) a force coming from
without’.1
Rabier, writing in almost the same terms, says: ‘The action of memory is a new
stimulus to the habituated conditions, having the effect of restoring a state of the nerve
centres (an impression) similar to, although usually weaker than? that which produced the
original representation.’2
William James is more formal still. ‘The phenomenon of “retention”’,
he says, ‘is not a fact of the mental order at all. It is a purely physical phenomenon,
a morphological feature, the presence of these “paths” namely, in the finest recesses of the
brain’s tissue.’3
The representation follows the restimulation of the affected area just as it followed the
original stimulus, but in the interval it has completely ceased to exist. Nobody insists more
than James upon the duality of these states and upon their heterogeneity. There is nothing
common to them except the traces left in the cerebral tissues by the first excitation, which
make the second more smooth and rapid.1
The consequence, moreover, follows logically
from the principles of this explanation.
But how is it that James has failed to see that thus he is in agreement with Maudsley,
whose theory he has earlier rejected with contempt?2
If at each moment of time the psychic
life consists exclusively in the actual condition present in the mind, one is justified
1
‘De l’habitude’, in Revue Philosophique, I, pp. 350–1.
2
Leçons de Philosophie, I, p. 164.
3
Principles of Psychology, I, p. 655.
1
Ibid., p. 656.
2
Ibid., pp. 145, 188. [He rejects Spencer on p. 145, Comte on p. 188, and deals rather roughly with
Maudsley on p. 656.—D.F.P.]
Individual and Collective Representations
4  Sociology and Philosophy
in saying that the mental life is reduced to nothing. It is agreed that the field of the mind’s
activity is, as Wundt says, very limited; its elements can be enumerated. If, then, they
are the only psychic factors in our conduct, we are forced to conclude that our conduct is
entirely dependent upon physical causes. Our direction is guided not by the few ideas that
hold our attention, but by the residues of our past: the habits which we have contracted, the
prejudices, the tendencies which motivate us and for which we cannot completely account
to ourselves—in a word, all that constitutes our moral character. If, then, nothing of all this
is mental, if the past cannot exist within us except in material form, it is indeed the organism
that leads the man. For however much of the past the mind can conceive at any given
moment it is nothing in the light of that which cannot be conceived and, further, the number
of entirely new impressions is infinitesimal. Moreover, pure sensation, in so far as it exists,
is of all intellectual phenomena the one to which the term epiphenomenon could the most
properly be applied. It is clear that it depends very closely on the disposition of the organs,
as long as another mental phenomenon does not intervene and modify it, in which case it
ceases to be pure sensation.
Let us, however, go further and consider what in fact happens in the mind. Could it be
said at least that the mental states have a specific nature, that they are subject to special
laws, and that if their influence is slight, due to their numerical inferiority, they are, for all
that, none the less original? The actual effect upon the action of the vital forces would not
amount to much, but it would be something. But how could it be possible? The very life
of these features consists in the sui generis manner in which they are grouped. It would
be necessary for them to be attracted and to associate together according to the affinities
of their intrinsic natures rather than the characteristics and organization of the nervous
system. If, then, the memory is organic, these associations must be reflections of equally
organic connexions. For if a representation cannot be evoked except through the antecedent
physical condition, and this latter cannot be revived except by a physical cause, ideas
cannot be linked unless the corresponding points in the cerebral mass are materially linked.
This, in fact, is expressly maintained by the partisans of the theory. In deducing this corollary
from their principle we have done no violence to their thought, since we attribute to
them nothing that they do not admit, as indeed they are logically bound to do. As James
himselfsays,1
‘the psychological law of association [of objects thought of through their
previous contiguity in thought or experience2
] would thus be an effect within the mind of
the physical fact that nerve currents propagate themselves most easily through those tracts
of conduction which have already been most in use’. And M. Rabier: ‘In an association of
ideas the suggestive feature a has its corresponding condition, the neural impression A; the
suggested feature b has another neural impression, B. This being accepted, to explain how
these two modifications, and consequently these two states of mind, follow each other is
only a short step, which is to conclude that the neural agitation has spread from A to B; and
because the movement has followed this trajectory once it will find the same path easier
in the future.’1
1
Op. cit., I, p. 563.
2
From the original not quoted by Durkheim.
1
Op. cit., I, 195.
 
1
But if the mental association is only an ‘echo’ of the physical association, what use is
it? Why does not the nervous movement immediately determine the muscular movement
without the intervention of this phantom-like mind? Will our earlier arguments be raised to
the effect that this ‘echo’ has its reality and that a molecular vibration in conjunction with
a conscious mind is not the same as one without the mind, that, in fact, a new entity is in
existence? The defenders of the epiphenomenalist theory use the same argument. They
also know very well that an unconscious cerebral process differs from what they call a conscious
cerebral process. The problem is, however, whether this difference is in the nature
of the process itself—in, for example, the intensity of the neural agitation—or whether it
is due to the addition of the mind. If this addition is not redundant, the new factor which
has now been added, the mind, must have some activity peculiar to itself and can, in fact,
produce effects which cannot occur in its absence. But if, as is supposed, the laws that
govern it are the same as those that govern neural matter, they can only be a useless repetition
of the latter. We cannot even suppose that this combination, while only reproducing
certain cerebral processes, nevertheless gives birth to a new state with a relative autonomy
which is more than a pure succedaneum of some organic phenomenon. For according to
the hypothesis such a state cannot last unless it is essentially based in a polarization of the
cerebral cells. What is a state of mind that has no persistence?
Generally speaking, if a representation can exist only when it is supported by certain
conditions of intensity and quality in the neural elements and disappears if this is not sustained,
it can have no reality of its own which does not derive directly from its substratum.
As Maudsley and his school have maintained, it is a shadow which no longer exists when
the object it reproduces is no longer there. From which one concludes that there is no real
mental life and consequently no real field for psychology. For, in these circumstances, if
one wished to understand mental phenomena and the ways in which they reproduce and
modify each other, one would not study them in themselves, but rather the anatomical conditions
of which they are the more or less faithful reflections. We could not even say that
they react upon and mutually modify each other, since these relations are only apparent.
When we say of reflections in a mirror that they attract, repel, or succeed each other, etc.,
we accept these terms as metaphorical—we do not ascribe actual life to these movements.
So little importance is attributed to these manifestations that we do not even wonder about
their processes and disappearance. It is apparently quite natural that an idea which preoccupied
us at one moment should cease to exist the next; it can only have seemed to exist if
it can so easily vanish.
If the memory is solely a property of the tissues there is no mental life, for the mind
does not exist outside the memory. This is not to say that intellectual activity is only the
reproduction of earlier states of mind, but that for those states of mind to undergo an intellectual
elaboration different from that implied in the laws of living matter they would have
to have an existence relatively independent of their physical substratum. Otherwise they
would be grouped as they are born and reborn, according to purely physical affinities.
Some have tried to find an escape from this intellectual nihilism by imagining some sort
of essence superior to phenomenal determinations. People talk vaguely of a mind which is
distinct from the materials furnished by the brain and which elaborates them by sui generis
processes. But what else is a mind that is not a system and sequence of particular thoughts
but a hypostatized abstraction? Whether they exist or not, science is not concerned with
Individual and Collective Representations
2  Sociology and Philosophy
essences or pure forms. For a psychologist the phenomenon of representaion is only an
assembly of representations. If these representations die as soon as they are born, then in
what does the mind consist? We must choose: either epiphenomenalism is correct or else
there is a memory that is a specifically mental phenomenon. But we have already seen that
the first position is untenable, and consequently the second solution must be accepted if we
are to remain consistent.
2
But we are forced to this conclusion for another reason.
We have seen that, if the memory is exclusively one of the properties of neural matter,
ideas have no power of mutual evocation; the order in which they occur to the mind can
only reproduce the order in which their physical antecedents are restimulated, and this can
be done only by physical causes. This proposition is so clearly implied in the premises of
the theory that it is formally admitted by its adherents. It is a proposition that makes any
psychic activity an abstraction lacking any reality, and it is directly contradicted by the
facts. There are cases—and important ones—when the sequence of ideas does not appear
to be explicable by such a theory. It is understandable, no doubt, that two ideas cannot
appear together or follow each other immediately unless their substrata in the brain have
some material connexion. Consequently there is no a priori impossibility in the restimulation
of the one, following the line of least resistance, being spread to the other and thus
determining the reappearance of its psychic manifestation. But there are no organic connexions
known that can explain in what way one idea can evoke a similar idea simply
because of their similarity. Nothing that we know about the brain leads us to suppose that
a vibration at A will tend to spread to B simply because there is a similarity between the
representations a and b. For this reason any psychology that sees the memory as a purely
biological fact is unable to explain associations of resem-blance except by reducing them
to associations of contiguity—that is to say, by denying them all reality.
This has been in fact attempted.1
It is argued that if two thoughts are similar they must
have at least one element in common, and this identically repeated feature has in both cases
the same neural element for support. This element is thus related to the two different groups
of cells that correspond to the parts of the two representations that differ, since it has co-operated
with both. Consequently it acts as a link between them, and thus the ideas are linked.
If, for example, I see a piece of white paper the idea I carry away includes the impression
of whiteness. The stimulation of the particular cell responsible for that colour induces a
nervous current which will radiate round but which will tend to follow the paths to which
it is habituated. That is to say, it will spread to those points that have previously been in
comunication with the first. But only those will satisfy this condition that have produced
representations similar in this one respect with the first. Thus it is that the whiteness of the
paper makes me think of the whiteness of snow. Thus two similar ideas are associated, but
the association depends not upon similarity but solely upon the material contiguity.
1
The reference is to James, Principles of Psychology, I, p. 690. There is no p. 690. The reader is
referred to chapters xiv and xvi, which deal respectively with Association and Memory.—D.F.P.
 
3
This explanation rests on a series of arbitrary postulates. First of all, there is no justification
for the supposition that representations are formed of defiite elements or atoms that
without losing their individuality go to the making up of various different representations.
Ideas are not made up of bits and pieces which they exchange according to circumstances.
The whiteness of paper is not the same as that of snow and the two appear in different representations.
Will it be said that they both depend upon the general impression of whiteness
which is common to both? If so, we should have to admit that the impression of whiteness
in general constitutes a sort of distinct entity which, grouped with other entities, gives
birth to a particular sensation of whiteness. However, there are no facts to justify such an
hypothesis. Everyhing goes to show that, on the contrary—and it is interesting that James
has contributed more than anyone in drawing attention to the proposition—psychic existence
is a continual stream of representations that blend into each other so that no one can
say where one begins or another finishes. No doubt the intellect comes to make certain
divisions, but it is we who introduce them into the psychic continuum. This process of
abstraction allows us to analyse what is, in fact, an indivisible complex. According to the
hypothesis which we have just discussed the brain could make these analyses itself, since
these divisions would have an anatomical basis. We know the difficulty we meet with when
we try to give the products of abstraction some sort of precarious form and individuality
by the use of language, quite apart from entertaining the supposition that this dissociation
corresponds to the original nature of the facts!
The physiological assumptions at the base of the theory are even more questionable.
If we concede that ideas can be decomposed into parts, we should have to admit further
that to each of the parts corresponds a particular neural element. Thus we should have one
part of the cerebral mass devoted to sensations of red, another to sensations of green, etc.
Even this would not be enough, for there would have to be particular substrata for each
shade of green, red, etc. For according to the hypothesis two colours of the same shade
cannot evoke each other unless their point of resemblance corresponds to one and the same
organic point, since all psychic similarity implies spatial coincidence. Such a geography of
the brain belongs to the world of the novelette rather than to that of science. No doubt we
know that certain intelectual functions are more closely bound to certain regions than to
others, but that these localizations are in no way precise or rigorous is demonstrated by the
phenomenon of substitution. To go further and to suppose that each representation dwells
in a particular cell is a gratuitous postulate, and one which the conclusion of this study will
demonstrate to be impossible. What is to be said in favour of the hypothesis that maintains
that each of the finest elements of the representation (allowing for the moment that they
exist) is no less narrowly localized? Thus the representation of the paper on which I am
now writing is literally scattered throughout the recesses of my brain. There must not only
be a place for the impression of colour, for the form and the texture, but the idea of the
colour in general will be in one place, the particular shade in another, and elsewhere the
special characteristics which this shade takes on in the present individual case when it is
before my eyes, etc. Surely it can be seen, quite apart from any other consideration, that if
mental life is divided to this degree and made up of these myriads of organic elements, the
unity and continuity which it presents are incomprehensible?
We could also ask how it comes about, if the resemblance of two representations is due
to the presence of one and the same element in both, that this element appears as doubled.
Individual and Collective Representations
4  Sociology and Philosophy
If we have an image ABCD, and another AEFG evoked by the first, and if consequently
the total process may be formulated as (BCD) A (EFG), then how is it that we see two A’s?
It will be argued that this distinction arises as a result of the differential elements given at
the same time: as A is involved at one and the same time with the complex BCD and the
complex EFG, and as these complexes are distinct from each other, logically we are bound
to admit that A has been doubled. This, however, only explains why we are bound to postulate
this duality; it does not for all that explain why it is, in fact, that we see it. While we
may conjecture that one image is related to two different complexes of circumstances, it
does not follow that we should see it doubled. At this very moment, let us suppose, I have
before my eyes on the one hand this piece of white paper and on the other the snow outside
on the ground. There are in my mind two representations of whiteness and not one only.
To reduce similarity to a partial identity is to make an artificial simplification. Two similar
ideas are distinct, even in those aspects which constitute the similarity. The elements which
they are said to have in common are separate in the one and in the other; we do not confuse
them even while we compare them. It is the sui generis relationship which is established
between them, the special combination which they form by virtue of this resemblance, the
particular characteristics of this combination, which give us the impression of similarity;
combination presupposes plurality.
One cannot then reduce resemblance to contiguity without mistaking the nature of
resemblance and forming hypotheses, at the same time physiological and psychological,
which cannot be justified; from this it follows that the memory is not a purely physical fact
and that representations as such have permanence. If, in fact, they vanished entirely as soon
as they had left the present consciousness, if they only survived as organic impressions, the
similarity which they might have with a particular present idea would not suffice to bring
them to life again, since no relationship of similarity can exist either directly or indirectly
between this surviving physical trace and a presently existing mental condition. If at the
moment that I see this piece of paper there exists in my mind nothing of the snow which I
saw previously, the first image cannot work upon the second nor the second on the first; the
one cannot evoke the other by the mere fact of similarity. The phenomenon is, however,
no longer unintelligible if the memory is a mental fact, if past representations exist as such
and if the act of remembering consists, not in a new and original creation, but in a new
emergence into the light of consciousness. If our psychic life is not annihilated at the same
time that it unfolds, if there is no solution of continuity between our earlier and our present
states of mind, then there is no impossibility in the proposition that they can work upon
each other and that the result of this mutual action can, in certain conditions, so increase the
intensity of the earlier ones that they come once again to consciousness.
It has been objected, it is true, that resemblance cannot explain association of ideas,
since it cannot appear until ideas have already been associated. It is argued that resemblance
is recognized because the ideas in question have become associated and therefore
cannot be the cause of this relationship. But this argument confuses resemblance itself
with the perception of that resemblance. Two representations can be similar, as the things
which they express, without our knowing it. The principal discoveries of science consist
precisely in this perception of previously unnoted analogies between ideas that are known
to everyone. Why should not this unperceived resemlance not produce effects which would
serve to characterize it and make it apparent? Images and ideas work upon each other
 
5
and these actions and reactions must necessarily vary with the nature of the representations;
particularly they must change according as the representations which are in this
manner brought together resemble each other, differ or contrast. There is no reason why
resemblance should not develop a sui generis property by virtue of which two conditions
separated by an interval of time should be made to come together. In order to admit this
as a reality it is not at all necessary to imagine representations as things having a separate
existence; it is merely sufficient to admit that they are not non-entities, that they are phenomena
but endowed with reality, with specific properties which behave in different ways
with each other according as they have, or have not, common properties. One could find in
the natural sciences many examples of the same thing. When bodies of different densities
are mingled those with a similar density tend to group together and separate themselves
from the others. Among living things similar elements have such an affinity that they tend
to lose themselves in fusion and become indistinct. Certainly this phenomenon of attraction
and coalescence can be explained by mechanical reasons and not by a mysterious attraction
which like has for like. But why cannot the grouping of similar representations in the
mind be explained in an analogous manner? Why should there not be a mental mechanism
(not exclusively physical) which should explain these associations without introducing any
occult faculty or scholastic entity?
Even at the present stage of knowledge it is possible to see roughly the path along which
such an inquiry would be guided. A representation does not appear without affecting the
body and mind. The very fact of its birth presupposes certain movements. In order to see
a house which is in front of me, I have to contract the muscles of the eye in a particular
manner and incline the head according to the height and dimensions of the building; furthermore,
the sensation, once it exists, in its turn determines certain movements. Now if
this has already happened once, if that is, the same house has been seen before, the same
movements were performed. The same muscles were moved, and in the same manner to a
certain extent, in so far, that is, as the subjective and objective conditions of the experience
are repeated identically. There exists, then, a definite connexion between the image of this
house in my memory and certain movements. Since these movements are the same as those
which accompany my present sensation of the house, through them is established a link
between my present and my past perception. Brought into being by the first they reawaken
the second; it is a well-known fact that by arranging the body in a particular attitude one
can evoke the corresponding ideas or emotions.
Nevertheless, this first factor is not the most important. However real the relation
between ideas and movements may be, it is not at all precise. The same system of movements
can serve very different ideas without being modified proportionately, and also the
ideas which are thus evoked are always of the most general kind. By arranging the subject’s
limbs in a suitable position one can suggest the idea of prayer in general but not of a particular
prayer. If it is true that any condition of the mind is involved with movement, it must
be added that, the further representations are from pure sensation, the more the element of
movement loses its importance and positive significance. The superior intellectual functions
presuppose the inhibition of movement, as is proved by the predominant role played
among them by concentration and the nature of concentration itself, which consists in as
complete a suspension as is possible of all physical activity. However, a simple denial of
the faculty of movement does not suffice to characterize the infinite diversity of the pheomIndividual
and Collective Representations
6  Sociology and Philosophy
ena of ideation. The effort which we make to refrain from action is no more bound to this
conception than to another, if the second has exacted of us the same degree of attention
as the first. But the link between the present and the past can also be established with the
help of purely intellectual intermediaries. All representations from the moment that they
come into being affect, apart from the organs, the mind itself. That is to say, they affect the
present and past representations which constitute the mind, if it is admitted that past representations
do persist with us. The picture which I see at a given moment reacts upon my
manner of seeing, my aspirations and desires. My sight of this picture is then, in a sense,
responsible for these diverse mental elements. If I see the picture again it will act in the
same way on these same elements, which persist unchanged except for the modifications
which time has perhaps brought about. It will excite them as it did on the first occasion,
and through them this stimulus will be communicated to the previous representation with
which, from now onwards, they are related and which is thus revived. For unless we are
to deny to psychic conditions all effective force, there is no reason why they should not be
able to transmit their energy to the mental conditions with which they are related, just as the
movement in one cell can be transmitted to neighbouring cells. As regards representational
life, these phenomena of transference make it all the more easy to conceive that it is not
formed of separate and distinct atoms but is a continuous whole of interpenetrating parts.
We offer the reader this sketchy explanation only as an indication. Our aim is above
all to demonstrate that there is no impossibility in resemblance being in itself the cause
of association. It has so often been argued that the socalled impossibility of this makes it
necessary to reduce similarity to contiguity and mental memory to physical memory, that
it seemed to us important to show that this original difficulty was by no means an insuperable
one.
3
Thus not only is the concession that representations can persist as such the sole means of
escape from an epiphenomenalist psychology, but also the existence of the association of
ideas by resemblance demonstrates this persistence.
It has, however, been claimed that these difficulties have been escaped only at the price
of incurring a greater. It is argued that if representations can persist as such they must do
so outside the consciousness, for we have no conception of all the ideas, sensations, etc.,
which we have experienced in the past and which we are likely to remember in the future.
If consciousness is implicit in the nature of representation, then one must conclude that the
idea of an unconscious representation is inconceivable and a contradiction in terms.
But by what right do we thus limit the psychic life? If the argument is only over words,
it is perhaps legitimate but scarcely fruitful. Because it is convenient to call the conscious
states of mind psychological it does not follow that outside consciousness there are only
organic or biochemical phenomena. It is a question of fact which only observation can
answer. Would it be argued that, if one were to withdraw consciousness from representation,
what remains would be incapable of representation to the imagination? But in this
way there are thousands of authentic facts which could equally well be denied. We do not
know what an imponderable material environment is, nor can we conceive an idea of it;
 
7
nevertheless the hypothesis is necessary for the explanation of the transmission of light
waves. How many well-established cases suggest that thought can travel over a distance?
The difficulty which we may have in conceiving so disconcerting an idea is not sufficient
reason for us to deny its reality, and we shall in all probability have to admit the existence of
waves of thought: a notion which is beyond, and even contradicts, our present conceptions.
Before the existence of invisible light rays able to pierce opaque bodies was demonstrated,
it had been easily proved that they were irreconcilable with the nature of light. One could
multiply these examples. Because, then, a phenomenon is not easily presented to the mind,
that is no justification for denying its existence when it manifests itself in definite effects
which are representable and which serve it in the capacity of signs. Such phenomena are
thought of, not in themselves, but by the effects which characterize them. There is not a
science that has not been forced to make this detour in order to arrive at the facts it studies.
Science goes from without, from the external and immediately sensible manifestations, to
the interior characteristics of which these manifestations betray the existence. A nervous
current or a light ray is, to begin with, an unknown quantity recognized as present by this
or that particular effect, and the task of a science is precisely to determine the exact content
of this initial conception. If, then, we are forced to say that certain phenomena can only be
caused by representations, that these phenomena are the outward signs of representational
life, and if, on the other hand, the subject in whom these representations appear is ignorant
of them, we shall say that unconscious psychic states can exist, however hard it may be for
the imagination to conceive their existence.
Instances of this are innumerable if, at least, we understand by consciousness the apprehension
of a given state by a given subject. In fact, within each one of us a multitude of
psychic phenomena occur without our apprehending them. We say that they are psychic
because they make themselves apparent by their characteristic signs of mental activity,
recognized by hesitation, tentativeness and the adjustment of movement to a preconceived
end. If when an act directed towards a particular goal takes place we are not sure that it is
intelligent, one wonders by what faculty the intelligence is able to distinguish itself from
what is not itself. The experiments of M.Pierre Janet have proved that many acts, while
bearing all the signs of being conscious, are not in fact so. For example, a subject who had
just refused to execute an order complies docilely when his attention is distracted just at the
moment when the words of command are given. It is evidently a complex of representations
which dictates his attitude, for the order cannot have its effect unless it has been heard
and understood. However, the patient does not know what is going on, he does not even
know that he has obeyed. If at the moment when he is about to perform the required gesture
his action is pointed out to him, it is, for him, the most surprising discovery.1
In the same
way, when one forbids a hypnotized subject to see a person or object in front of him, the
order will have effect only if it is impressed upon the mind. Nevertheless the consciousness
is not aware of it. There are cases also of unconscious counting and complex calculations
performed by an individual who had no idea of what he was doing.1
These experiments,
1
See Janet, L’Automatisme psychologique, p. 237 et seq. [See also, for James’s discussion of Janet,
James, op. cit., I, pp. 203 et seq. For James’s discussion of the question ‘Can states of mind be
unconscious?’ see op. cit.,
I, pp. 162 et seq. D.F.P.]
1
Janet, ibid., p. 225.
Individual and Collective Representations
8  Sociology and Philosophy
which have been varied, have, it is true, been conducted with abnormal subjects, but they
only reproduced in amplified form what normally occurs in all of us. Our judgments are
influenced at every moment by unconscious judgments; we see only what our prejudices
permit us to see and yet we are unaware of them. We are always to a certain extent in a
state of distraction, since the attention, in concentrating the mind upon a small number of
objects, blinds it to a greater number of others; all distraction has the effect of withdrawing
certain psychic states from consciousness which do not cease to be real for all that,
since they continue to function. How many times is there a positive contradiction between
the actual state of a thing as it is and as it appears in the mind? We imagine that we hate
someone when in fact we love him, and the reality of this love shows itself in acts that
are apparent to others while we believe ourselves to be under the influence of completely
opposite sentiments.2
Furthermore, if all that is psychic were conscious, and all that is unconcious physiological,
psychology would return to the old method of intropecion. For if the reality of mental
states is the same as our consciousness of them, the conscious mind would suffice for the
complete understanding of this reality, since they would be the same thing, and there would
be no need for recourse to the complicated and roundabout methods at present in use. We
can, in fact, no longer regard the laws of phenomena as superior to the phenomena and
directing them from without. The laws are immanent in them and are their manner of being.
If, then, mental facts are only as we see them and only act as we are conscious of them acting
(which is the same thing), their laws are given at the same time. In order to understand
them we have only to look at them. As for those factors of mental life which are unconscious
and consequently cannot be studied in this way, they will have to come within the
field not of psychology but of physiology. There is no need for us to demonstrate the fallacy
of so facile a psychology; there is no question that the interior world of the mind is still, to
a great extent, unexplored; that discoveries are constantly being made and many more are
yet to be made and that, consequently, it will call for more than a little application of the
conscious mind to make them. It is useless to argue that those representations that pass for
unconscious are only perceived incompletely and confusedly; for this confusion can have
only one cause, simply that we do not see all that these representations comprehend—that
there are real and effective elements which are not, consequently, purely physical facts,
and which are not, however, obvious to the consciousness. This obscure consciousness is
a partial unconsciousness, and we must once again remember that the limits of consciousness
are not the limits of all psychic activity.
2
According to James this is no proof of lack of consciousness. If I imagine that I hate or am indifferent
when in fact I am in love, I have merely misnamed a condition of which I am fully conscious. I
must confess that I do not understand this. If I misname a condition, it is because my consciousness
of it also is false and does not express all the characteristics of this condition. Nevertheless these
characteristics which are not conscious still function. They are then, in a way, unconscious. My feelings
have all the constituent traits of love since they affect my conduct; but I do not recognize them,
so that in a sense my passions direct me one way and the knowledge which I have of them, another.
The two phenomena are not coterminous. Nevertheless it is difficult to see in an inclination like love
anything other than a psychic phenomenon. (See James, op. cit., I, p. 174.)
 
9
In order to avoid this word ‘unconscious’ and the attendant difficulties which the mind
faces in conceiving its content, one might say that unconscious phenomena are attached to
centres of secondary consciousness dispersed throughout the organism and unknown to the
primary centre, while normally subordinate to it.Also one could say that consciousness can
exist without any apprehension by a given subject’s ego. We cannot at the moment discuss
these hypotheses1
which, although plausible, leave unassailed the proposition which we
wish to establish. All that we wish to say is that certain phenomena occur in us which are
of a psychic order and which are nevertheless not known by our conscious selves. Whether
they are known to some other unknown ‘self’ or whether they are outside the realm of all
apprehension is not for us a matter of primary importance. All we wish to be conceded is
that representational life extends beyond our present consciousness and, as a consequence,
that the conception of memory as a fact of the psychological order is an intelligible proposition.
All that we are trying to make clear here is that such a memory exists without going
into all the possible ways in which it can be conceived.
4
We are now in a position to conclude.
If representations, once they exist, continue to exist in themselves without their existence
being perpetually dependent upon the disposition of the neural centres, if they have
the power to react directly upon each other and to combine according to their own laws,
they are then realities which, while maintaining an intimate relation with their substratum,
are to a certain extent independent of it. Certainly their autonomy can only be a relative
one; there is no realm of nature that is not bound to others. Nothing could be more absurd
than to elevate psychic life into a sort of absolute, derived from nothing and unattached
to the rest of the universe. It is obvious that the condition of the brain affects all the intellectual
phenomena and is the immediate cause of some of them (pure sensation). But, on
the other hand, it follows from what has been said earlier that representational life is not
inherent in the intrinsic nature of nervous matter, since in part it exists by its own force
and has its own particular manner of being. A representation is not simply an aspect of the
condition of a neural element at the particular moment that it takes place, since it persists
after that condition has passed, and since the relations of the representations are different
in nature from those of the underlying neural elements. It is something quite new which
certain characteristics of the cells certainly help to produce but do not suffice to constitute,
since it survives them and manifests different properties. To say that the mental condition
does not derive directly from the cell is to say that it is not included in it, that it forms itself
in part outside it and is to that extent exterior to it. If it was directly derived it would be
within it, since its reality would derive from no other source.
1
The idea of an unconscious representation and that of a consciousness without the ego are basically
equivalent. When we say a mental fact is unconscious we mean simply that it is not apprehended. The
question is merely which is the more suitable expression. From the point of view of the imagination
both are equally difficult. It is no easier for us to imagine a representation without the thinking subject
than to imagine a representation without consciousness.
Individual and Collective Representations
10  Sociology and Philosophy
When we said elsewhere that social facts are in a sense independent of individuals and
exterior to individual minds, we only affirmed of the social world what we have just established
for the psychic world. Society has for its substratum the mass of associated individuals.
The system which they form by uniting together, and which varies according to their
geographical disposition and the nature and number of their channels of communication,
is the base from which social life is raised. The representations which form the network of
social life arise from the relations between the individuals thus combined or the secondary
groups that are between the individuals and the total society. If there is nothing extraordinary
in the fact that individual repesentations, produced by the action and reaction between
neural elements, are not inherent in these elements, there is nothing surprising in the fact
that collective representations, produced by the action and reaction between individual
minds that form the society, do not derive directly from the latter and consequently surpass
them. The conception of the relationship which unites the social substratum and the social
life is at every point analogous to that which undeniably exists between the physiological
substratum and the psychic life of individuals, if, that is, one is not going to deny the existence
of psychology in the proper sense of the word. The same consequences should then
follow on both sides. The independence, the relative externality of social facts in relation to
individuals, is even more immediately apparent than is that of mental facts in relation to the
cerebral cells, for the former, or at least the most important of them, bear the clear marks
of their origin. While one might perhaps contest the statement that all social facts without
exception impose themselves from without upon the individual, the doubt does not seem
possible as regards religious beliefs and practices, the rules of morality and the innumerable
precepts of law—that is to say, all the most characteristic manifestations of collective
life. All are expressly obligatory, and this obligation is the proof that these ways of acting
and thinking are not the work of the individual but come from a moral power above him,
that which the mystic calls God or which can be more scientifically conceived.1
The same
law is found at work in the two fields.
Furthermore, it can be explained in the same way in the two cases. If one can say that,
to a certain extent, collective representations are exterior to individual minds, it means
that they do not derive from them as such but from the association of minds, which is a
very different thing. No doubt in the making of the whole each contributes his part, but
private sentiments do not become social except by combination under the action of the sui
generis forces developed in association. In such a combination, with the mutual alterations
involved, they become something else. A chemical synthesis results which concentrates
and unifies the synthesised elements and by that transforms them. Since this synthesis is
the work of the whole, its sphere is the whole. The resultant surpasses the individual as
1
If the characteristics of obligation and constraint are so essential to these eminently social facts, it is
to be expected that they will be found, if less obviously, in other social facts. It is impossible for phenomena
of the same nature to differ to the extent that some penetrate to the individual from without
and others are the result of a different process.
We should like here to correct a false interpretation that has been put upon our thought. When we
said that obligation and constraint are the characteristics of social facts we had no intention of giving
a summary explanation of the latter. We wished simply to point out a convenient sign by which the
sociologist can recognize the facts falling within his field.
 
11
the whole the part. It is in the whole as it is by the whole. In this sense it is exterior to the
individuals. No doubt each individual contains a part, but the whole is found in no one. In
order to understand it as it is one must take the aggregate in its totality into consideration.1
It is that which thinks, feels, wishes, even though it can neither wish, feel, nor act except
through individual minds. We can see here also how it is that society does not depend upon
the nature of the individual personality. In the fusion from which it results all the individual
characteristics, by definition divergent, have neutralized each other. Only those more general
properties of human nature survive, and precisely because of their extreme generality
they cannot account for the specialized and complex forms which characterize collective
facts. This is not to say that they count for nothing in the resultant, but they are only its
mediate conditions. Without them it could not emerge, but they do not determine it.
The exteriority of mental facts in relation to the cerebral cells is due to the same causes
and is of the same nature. Nothing, in fact, justifies the supposition that any representation,
however elementary, can be directly produced by a cellular vibration of a given intensity
and tone. But there is no sensation which is unrelated to a certain number of cells. The manner
of cerebral localizations admits of no other hypothesis, for the images are definitely
related only to more or less extended zones. Perhaps, as the fact of substitutions seems to
show, the whole brain participates in the elaboration from which they result. At least, this
seems to be the only way in which we can explain how it is that sensation is dependent
upon the brain while at the same time constituting a new phenomenon. It is dependent
because it is formed as a result of molecular modifications; how could it be made otherwise
and whence could it derive? But it is at the same time another thing because it results
from a new and sui generis synthesis into which these modifications enter as elements, but
in which they are transformed by the very fact of their fusion. Certainly we do not know
exactly how these combined movements do give rise to a representation, but neither do we
know how it is that movement can be translated into heat by being arrested in its course or
how heat is translated into movement. However, there is no doubt about the transformation
itself; what then is there more impossible about the first? More generally this objection
would strike at the root of all change, for between an effect and its causes, a resultant and
its elements, there is always a qualitative distance (écart). It is for metaphysics to find the
concepts which will render this heterogeneity in an acceptable form; for us it is sufficient
that its existence cannot be contested.
But if each idea (or at least each sensation) is due to the synthesis of a number of cellular
conditions, combined according to laws by forces which we do not as yet know, it is
obvious that it cannot be limited to one particar cell. It escapes from each because none
is sufficient of itself to bring it into being. Representational life cannot be divided among
and ascribed to particular neural elements, since several of these elements combine for its
generation; but it could not exist without the whole formed by their union, just as the collective
could not exist without the whole formed by the union of individuals. Neither the one
nor the other is made up of particular parts that can be attributed to the corresponding parts
of their respective substrata. Each mental condition is, as regards the neural cells, in the
same condition of relative independence as social phenomena are in relation to individual
1
See Le Suicide, pp. 345–63.
Individual and Collective Representations
12  Sociology and Philosophy
people.As it cannot be reduced to a simple molecular modification it is not subject to modifications
of this kind, which can happen in isolation at different points in the brain. Only
those physical forces that affect the entire group of cells that support it can affect it also.
But in order to survive, it is not in need of constant support and, as it were, constant recreation
by a continuous stream of nervous energy. To recognize this limited autonomy of the
mind is basically the same as the essential and positive content of our notion of spirituality.
There is no need to conceive of a soul separated from its body maintaining in some ideal
milieu a dreamy and solitary existence. The soul is in the world and its life is involved with
the life of things, or we could say that all our thoughts are in the brain. We must add that
within the brain, while they may be more related to certain areas of it than to others, they
cannot be rigidly localized or situated at definite points. This diffusion in itself is sufficient
proof that they constitute a specifically new phenomenon. In order that this diffusion can
exist, their composition must be different from that of the cerebral mass, and consequently
they must have a manner of being which is special to them.
Those, then, who accuse us of leaving social life in the air because we refuse to reduce
it to the individual mind have not, perhaps, recognized all the consequences of their objection.
If it were justified it would apply just as well to the relations between mind and brain,
for in order to be logical they must reduce the mind to the cell and deny mental life all
specificity. But then one falls into the dire difficulties that we have already indicated. Following
the same principle, one would be bound to say that the properties of life consist
in particles of oxygen, hydrogen, carbon and nitrogen, which compose the living protoplasm,
since it contains nothing beyond these particular minerals just as society contains
nothing more than the individuals.1
Here the impossibility of the conception which we are
opposing will perhaps appear with even greater clarity than in the earlier instances. How
can living movements be based in non-living elements? How are the characteristic properties
of life distributed among these elements? They cannot be equally divided since they are
different. Oxygen cannot play the same role as carbon or be invested with the same properties.
No less inadmissible is the contention that each aspect of life is embodied in a different
group of atoms. Life cannot be thus divided; it is one, and consequently cannot be based on
anything other than the living substance in its totality. It is in the whole, not in the parts. If,
then, to understand it as it is, it is not necessary to disperse it among the elementary forces
of which it is the resultant, why should it be different for the individual mind in relation to
the cerebral cells and social facts in relation to individuals?
In fact individualistic sociology is only applying the old principles of materialist metaphysics
to social life. It claims, that is, to explain the complex by the simple, the superior
by the inferior, and the whole by the part, which is a contradiction in terms. The contrary
principle does not seem to us to be any less questionable. One cannot, following idealist
and theological metaphysics, derive the part from the whole, since the whole is nothing
without the parts which form it and cannot draw its vital necessities from the void. We
must, then, explain phenomena that are the product of the whole by the characteristic properties
of the whole, the complex by the complex, social facts by society, vital and mental
facts by the sui generis combinations from which they result. This is the only path that a
science can follow. This is not to say that there is a solution of continuity between these
1
At least, individuals are the only active elements. More correctly, society also comprises things.
 
13
various stages of reality. The whole is only formed by the grouping of the parts, and this
grouping does not take place suddenly as a result of a miracle. There is an infinite series of
intermediaries between the state of pure isolation and the completed state of association.
But as the association is formed it gives birth to phenomena which do not derive directly
from the nature of the associated elements, and the more elements involved and the more
powerful their synthesis, then the more marked is this partial independence. No doubt it
is this that accounts for the flexibility, freedom and contingence that the superior forms of
reality show in comparison with the lower forms in which they are rooted. In fact, when
a way of doing or being depends from a whole without depending immediately from the
parts which compose that whole, it enjoys, as a result of this diffusion, a ubiquity which to
a certain extent frees it. As it is not fixed to a particular point in space it is not bound by too
narrowly limited conditions of existence. If some cause induces a variation, that variation
will encounter less resistance and will come into existence more easily because it has, in a
way, a greater scope for movement. If certain of the parts reject it, certain others will form
the basis (point d’appui) necessary for the new arrangement without, for all that, being
obliged to rearrange themselves. That at least is how one can conceive how it is that one
organ is able to perform different functions, different parts of the brain can substitute for
each other, and one social institution can successively further the most varied ends.
Also, while it is through the collective substratum that collective life is connected to the
rest of the world, it is not absorbed in it. It is at the same time dependent on and distinct
from it, as is the function of the organ. As it is born of the collective substratum the forms
which it manifests at the time of its origin, and which are consequently fundamental, naturally
bear the marks of their origin. For this reason the basic matter of the social consciousness
is in close relation with the number of social elements and the way in which they are
grouped and distributed, etc.—that is to say, with the nature of the substratum. But once a
basic number of representations has been thus created, they become, for the reasons which
we have explained, partially autonomous realities with their own way of life. They have the
power to attract and repel each other and to form amongst themselves various syntheses,
which are determined by their natural affinities and not by the condition of their matrix. As
a consequence, the new representations born of these syntheses have the same nature; they
are immediately caused by other collective representations and not by this or that characteristic
of the social structure. The evolution of religion provides us with the most striking
examples of this phenomenon. It is perhaps impossible to understand how the Greek or
Roman Pantheon came into existence unless we go into the constitution of the city, the
way in which the primitive clans slowly merged, the organization of the patriarchal family,
etc. Nevertheless the luxuriant growth of myths and legends, theogonic and cosmological
systems, etc., which grow out of religious thought, is not directly related to the particular
features of the social morphology. Thus it is that the social nature of religion has been so
often misunderstood.
It has been believed that it is formed to a great extent by extra-social forces because the
immediate link between the greater part of religious beliefs and the organization of society
has not been perceived. By this reasoning one would have to exclude from psychology
everything beyond pure sensation. For if sensation, this primary store of the individual
mind, cannot be explained except by the condition of the brain and the organs, once it
exists it forms itself according to laws which neither morphology nor cerebral physiolIndividual
and Collective Representations
14  Sociology and Philosophy
ogy can adequately account for. From this derive images and these, in their turn, group to
form conceptions. As these new states are added to the old, as they are separated by more
intermediaries from the organic base upon which, nevertheless, all mental life rests, they
become less immediately dependent upon it. They do not cease to be psychic facts, for it is
in them that one can best observe the characteristic attributes of the mind.1
Perhaps these comparisons will make clear why we insist so much upon a distinction
between sociology and individual psychology.
It is simply a matter of introducing and acclimatizing in sociology a conception parallel
to that which is tending to prevail more and more in psychology. During the last decade
a great innovation has been made in that science. Interesting efforts have been made to
establish a psychology which is in fact psychological without any other qualifying adjective.
The old introspectionists were content to describe mental phenomena without trying
to explain them; psycho-physiology explained them but dismissed their distinctive traits as
negligible. A third school is being born which is trying to explain them without destroying
their specificity. For the first mental life certainly had a nature of its own, but it was one
that lifted the mental out of the world and above the ordinary methods of science. For the
second school it was nothing in itself, and the role of the scientist was to pierce the superficial
stratum in order to arrive at the underlying realities. Neither school saw anything
more than a thin curtain of phenomena which, according to the first, was easily apparent
to the eye of the conscious mind and, to the second, was lacking in any consistency.
Recent experiments have shown us that it is far better to conceive of it as a vast system of
sui generis realities made up of a great number of mental strata superimposed upon each
other, far too profound and complex for the conscious mind to pierce, far too specialized to
be accounted for by purely physiological considerations. It is thus that this spirituality by
which we characterize intellectual facts, and which seemed in the past to be either above
or below the attentions of science, has become itself the object of a positive science, and
that, between the ideology of the introspectionists and biological naturalism, a psychological
naturalism has been founded, the legitimacy of which the present article will, perhaps,
help to demonstrate.
1
From this it can be seen how difficult it is to define social facts as phenomena produced
in but also by the society. The expression is not exact, for there are social facts, and not among
the least, which are produced, not by the society but by already formed social products. It is as
though one were to define as mental facts those which are the product of the combined action of
the cerebral cells or a certain number of them. Such a definition will not serve to determine and
circumscribe the object of sociology. The relation of these derivatives can only be established as
the science advances. At the beginning of research one does not know the causes of the phenomena
that are being studied, and indeed one can only hope to know little. We must then limit the
field of investigation by another criterion if we wish to know clearly what we are concerned with.
The process by which these social products of the second degree are formed, if analogous with that
observed in the individual mind, is also an individual phenomenon. The combinations from which
myths, theogonies and popular cosmogonies result are not identical with the association of ideas in
the individual mind, even though each throws some light on the other. A special branch of sociology,
which does not yet exist, should be devoted to research into the laws of collective ideation.
 
15
A similar transformation should take place in sociology, and it is towards this goal that
all our efforts are directed. If there are no longer many thinkers who dare explicitly to
put social facts beyond the realm of nature, many still think that it is sufficient in order to
explain them to go to the individual mind; certain others even wish to reduce them to the
general properties of organic matter. For all of them, consequently, society is nothing in
itself; it is only an epiphenomenon of individual life (organic or mental, it makes no difference)
just as, according to Maudsley and his disciples, individual representation is only
an epiphenomenon of physical life. The first would have no other reality than that which
it received from the individual, just as the second would have no other existence than
that which it takes from the neural cell, and sociology would become applied psycholo-
gy.1
But even the example of psychology shows that this conception of science should be
discarded. Beyond the ideology of the psycho-sociologist and the materialistic naturalism
of the socio-anthropologist there is room for a sociological naturalism which would
see in social phenomena specific facts, and which would undertake to explain them while
preserving a religious respect for their specificity. Nothing is wider of the mark than the
mistaken accusation of materialism which has been levelled against us. Quite the contrary:
from the point of view of our position, if one is to call the distinctive property of
the individual representational life spirituality, one should say that social life is defined by
its hyperspirituality. By this we mean that all the constituent attributes of mental life are
found in it, but elevated to a very much higher power and in such a manner as to constitute
something entirely new. Despite its metaphysical appearance, this word designates nothing
more than a body of natural facts which are explained by natural causes. It does, however,
warn us that the new world thus opened to science surpasses all others in complexity; it is
not merely a lower field of study conceived in more ambitious terms, but one in which as
yet unsuspected forces are at work, and of which the laws may not be discovered by the
methods of interior analysis alone.
1
When we use the word ‘psychology’ by itself we mean individual psychology, and for the sake of
clarity in discussion it is convenient to limit the word to this. Collective psychology is sociology,
quite simply—why not employ the latter term exclusively? Inversely the word ‘psychology’ has
always designated the science of the individual mentality—why not reserve this meaning to it? Thus
we should avoid much ambiguity.
Individual and Collective Representations