The evolution of general ideas

Before we embark on the study of the principal concepts, it is incumbent upon us (in order to determine for each of these, separately, the conditions of their genesis and development—as was shown for abstraction in general) to throw as much light as possible upon the very vexed question of the psychological nature of the concepts of pure symbolism, where the word appears as the sole element that exists in consciousness. Is it true that we can think effectually and usefully with words and nothing but words, as has been sustained to satiety? Is not this assertion founded upon the misapprehension of a factor which, although it does not enter into consciousness, is none the less in active existence? The investigation of this point is the prime object of the following chapter.

It is unnecessary to enter in detail into the researches of the last thirty years, as to the seat and the nature of images. Yet since these have been the point of departure of the following inquiry, the results may be briefly summarised.

It is generally admitted that the image occupies the same seat as the percept of which it is a weak and incomplete residuum, i. e., in order to produce itself in consciousness it demands the putting into activity of certain definite portions of the cerebral centres. The energy of the representative faculty does not merely vary from individual to individual in a general manner: there are particular forms of imagination, constituted by the very marked predominance of a certain group of representations, visual, auditory, muscular, olfactory, gustatory.

Normal observation, and still more pathological documents, have thus determined certain types. We may also (though this is mere hypothesis and difficult to verify) admit a “mixed” or “indifferent” type, in which the different species of sensations are represented by corresponding images of equal clearness and vigor, without marked predominance of any one group, whilst still maintaining their relative importance: e. g., it is clear that in man the visual and olfactory images cannot be equivalent in absolute importance. Excluding this indifferent type, we have three principal “pure” types: visual, auditory, muscular or motor, signifying a tendency to represent things in terms borrowed from vision, from sound, or from movement. If we push the investigation further, we find that these types again imply variations or sub-types. Thus there may be a lively faculty for representation of complex visual forms (faces, landscapes, monuments) along with a weak sense for graphic signs (printed or written words) and so on.

The numerous works devoted to this subject, and too well known to be insisted on here, lead us to this conclusion: that there is no general faculty of imagination. This is a vague term which designates very different individual variations; these last alone have any psychological reality, and are alone important in cognising the mechanism of the intellect.

May it not be the same for the faculty of conception? May not the word “general idea” or “concept” be in its kind the equivalent of the word image, namely a vague formula,—its psychological reality lying in types or variations as yet undetermined? I am exposing for ideas, the problem that has already been set forth for images, while recognising its much greater obscurity. The psycho-physiological conditions of the existence of concepts are practically unknown: this is a terra incognita wherein the new psychology has hardly adventured itself, and where it would indeed have been chimerical to tread before the preliminary study of the image.

I.

The question I have set myself to elucidate is very modest, very limited and circumscribed, representing only part of the problem indicated above. It may, however, teach us something of the ultimate nature of concepts. It is as follows:

When we think, hear, or read a general term, what arises as sign in consciousness, directly and without reflexion?

I have purposely italicised these words in order to emphasise my principal aim, which was to discover the instantaneous operations (conscious or unconscious) that occur in such a case, in persons whose habits of mind are widely different. I endeavored as much as possible to eliminate reflexion and to seize the mental state. With time and effort, minds that are least apt in abstraction will arrive at a more or less successful translation of general terms, or at the substitution for them of some mangled and halting definition. I set myself as far as possible to suppress this secondary phase of the mental process, and to arrest it at the first, in order to determine what the word evokes immediately[80] and in what degree this differs with the individual.

In order to make the answers more exactly comparable, I interrogated only the adults of both sexes, excluding all children. It was indispensable to my investigation that it should comprise people of very different degrees of culture, habits of mind, and profession. The principal classes were mathematicians, physicists, doctors, scientists, philosophers, painters, musicians, architects, men of the world, women, novelists, poets, artisans. The last class made such confused replies that I must regard their documents as worthiest. Too much is left for individual interpretation. The total number of persons interrogated amounted to one hundred and three.

The method was invariably the same. We said to the subject: “I am going to pronounce certain words; will you tell me directly, without reflexion, whether this word calls up anything or nothing in your mind? If anything, what is suggested to you?” The reply was noted down at once; if delayed beyond five to seven seconds, it was held to be null, or doubtful. In the case of naïve subjects, I employed certain preliminaries: before pronouncing abstract words, concrete terms (designating a monument, or person) such as would evoke a simple image, were heard; then the impulse being given, I proceeded to the enumeration of general terms.

The words which served as material for the inquiry were fourteen in number, proceeding from the concrete to the highest abstractions. They were enunciated in an indifferent order and were as follows: dog, animal, color, form, justice, goodness, virtue, law,[81] number, force, time, relation, cause, infinity.

The inquiry was invariably oral, never in writing, the greatest care being taken to prevent the person from knowing the end in view, unless afterwards: which led in certain cases to interesting explanations. The very nature of my method prevented me from extending it as widely as I could have wished. I could not, as was done in England, distribute printed questions among the public, because it was necessary to note the spontaneous answer immediately before it was corrected by later reflection. Moreover, I needed unsophisticated subjects, ignorant of my purpose, and therefore eliminated all whom I suspected of being even indirectly acquainted with it.

The majority were interrogated on the fourteen terms cited above, others on a few only; so that the total number of responses was over nine hundred. It would be beside the mark to publish them here. They are nothing more than data which have to be interpreted. Three principal or pure types appear to stand out from them, besides the failures or mixed cases. These may be termed the concrete type, the visual typographic type, and the auditory type. Each of these corresponds with a particular mode of representing the general idea. We will examine them separately.

1. Concrete Type. Here the abstract word nearly always evokes an image, vague or precise; usually visual, sometimes muscular. It is not a simple sign, it does not represent the total substitution, it is not dry, and finally reduced. It is immediately and spontaneously transformed into a concrete. In fact, the persons of this type think only in images. Words are for them no more than a kind of vehicle, a social instrument of mutual comprehension. When a sequence of general or abstract terms passes through their minds, what really passes is a succession of concretes, save for the very abstract words which “evoke nothing.” This is an answer I have often received, and which, in virtue of its importance, will be considered apart, at the end of this chapter.

The concrete type appears to be the most widely distributed; it obtains almost to exclusion among women, artists, and all who have not the habit of scientific abstraction. I have selected a few examples from among the many observations belonging to this type.

A painter.—Cause: nothing. Relation: relations of terms; recital, written report. Law: judges in red robes. Number: vague. Color: contrast between green of plant, and red of drapery. Form: a round block, a woman’s shoulder. Sound: a murmur. Dog: ears of a dog running. Animal: vague collection, as in certain Dutch pictures. Force: hits out with his fists. Goodness: his young mother, seen vaguely. Time: Saturn with his scythe. Infinity: a black hole.

A woman.—Cause: I had been the cause of her son’s success. Law: the government is bad. Color: sees an impressionist picture by her son. Form: names a beautiful person. Goodness and Virtue: names two people who each have this quality. Force: sees men fighting. Relation: social relations between husband and wife. Justice: sees an audience-hall and judges. Dog: sees a dog that bit one of her parents. Infinity: nothing. Time: a metronome.

These two interrogatories are complete. I might proceed by another method: that of taking each general term (law, cause, number, etc.) and quoting all the answers received, among which many would be identical. Such an enumeration would be long and superfluous: we cannot, however, neglect a few of the particulars. For the word cause, several persons (women, artists, people in society) replied “cause célèbre,” “procès célèbre,” for the most part mentioning one only, and that some recent trial. At first this reply annoyed me, and appeared to be useless for my inquiry. Later, on the other hand, I felt it to be instructive, because it characterises better than any description the type which I have denoted as concrete, and the particular turn of this kind of mind, in which the abstract sense does not present itself, at any rate at the beginning.

I may also note two answers given me immediately by a celebrated painter:—Number: I see many brilliant points. Law: I see parallel lines. (Is this the unconscious idea of levelling by the law?)

The terms goodness and virtue suggested answers which are easily summarised: they fall into two categories. (1) Nothing; this answer does not belong to the concrete type; (2) a definite person, who was always named and who thus becomes the incarnation, the concrete representation.

Nearly all the images evoked belong to the visual sense; the word force, however, most frequently called up pure muscular images, or the same accompanied by a vague visual representation. Example—Seeing somebody lifting a weight; I vaguely see something pulling; a weight suspended by a ring; a string drawing on a nail; pressure of my fist in a fluid; the Marshal of Saxony breaking an écu of six pounds, etc.

I have been describing the ordinary and principal form of the concrete type. It consists in the immediate and spontaneous substitution of a particular case (fact or individual) for the general term. In certain observations a slightly different variation may be detected; I have encountered it among several historians and learned men. In the ordinary type, the whole (general) is thought by means of the part (concrete); in the variation, the thinking is by analogy, and the mechanism seems to be reduced to pure association. A few examples will explain the distinction. The replies in duplicate were given by different persons. Number: the “Language of Calculation,” Pythagoras. Cause: Hume’s theory of causality; Kant’s theory. Law: the “Tables of Malaga,” Montesquieu’s definition. Color: the chemistry of the spectrum. Justice: Littré’s definition. Animal: the πὲρι ψυχῆς of Aristotle. Time: a vague metaphysical theory. Relation: discussion of Ampère and Tracy on this subject. Infinity: books on mathematics. Color: treatises of photography, etc.

It might be objected that there is a certain association in ordinary cases as in these; but the distinction will readily be perceived. The former proceed from that which contains, to the content—from the class to the fact: they think the whole by means of the part; there is an internal association. The latter form associations beside and from without. Apparently these do not reach to the concrete, they stop half way; for a complete generality they substitute a semi-generality. Further than this, my data are neither sufficiently numerous, nor clear enough, for the point to be insisted on.

2. Visual Typographic Type. Nothing is easier to define. In its pure form it consists in seeing printed words and nothing more; in three cases words were seen written. Among some the vision of the printed words was accompanied by a concrete image as in the first type, but only for semi-concrete concepts (dog, animal, color); for the higher abstracts (time, cause, infinity, etc.) the typographical vision alone exists.[82] This mode of representation is widely distributed among those who have read much; but there are many exceptions.

No doubt many of my readers will discover from self-observation that they belong to this type. I have further noticed that all who have this mode of representation regard it as normal, and necessary, in anyone who knows how to read. This is a fallacy. I do not possess it myself in the faintest degree, and have met many others who resemble me.

Thus I was little prepared to discover this type; and had even reached my thirtieth observation without suspecting it, when I encountered such a clear case as to put me on the track. I was interrogating a well-known physiologist. To every word except Law and Form, he replied “I see them in printed characters” and was able to describe these accurately.

Even the words dog[83], animal, color, were unaccompanied by any image. He volunteered further information which may be reduced to the statement, “I see everything typographically.” The same holds good for concrete objects. If he hears the names of his intimate friends whom he meets every day, he sees the names printed; it is only by an effort of thought that he sees the image. The word “water” appears to him as if printed, and he has no vision of a liquid. If he thinks of carbonic acid, or nitrogen, he sees indifferently either the words printed or the symbols CO₂, N. He does not see the complex formulæ of organic chemistry, but the words only.

Surprised (from the reasons above indicated) at this observation—of the sincerity and precision of which there could be no doubt—I continued my investigation, and discovered this mode of thinking in general terms to be sufficiently common. Several cases indeed were as pure and as detailed as the one just cited. Thenceforward I adopted the habit of invariably asking at the close of my interrogatory “Did you see the words printed?”

Several people remarked that they had read a great deal, and corrected many proofs, and that this would account for their belonging to the typographical visual type. The influence of habit is certainly enormous, but is no adequate explanation here, since there are many exceptions. I have myself read and corrected many proofs, but no word ever appeared in my consciousness as printed, unless after considerable effort, and then vaguely. Hence this mode must be due in great part to natural disposition.

Among the compositors questioned I found: (1) That they saw my fourteen words printed in some special type, which they occasionally specified; (2) they had a concomitant image for semi-concrete terms; (3) for abstract terms no image accompanied the typographical vision. Here we have the superposition of two types: the one natural, and of primitive formation (concrete type), the other acquired, and of secondary formation (typographical visual type).

In short,—in many minds the existence of the concept is associated with a clear vision of the printed word and nothing beyond it.

3. Auditory Type.—In its pure form this seems to be rare. It consists in having in mind nothing but signs (auditory images) unaccompanied either by the vision of printed words or by concrete images. Possibly it may preponderate among orators and preachers; of this I have no documentary evidence. Musicians do not appear to belong to this type.

One very clear and complete case of the kind I have, however, encountered. This was a polyglot physician known as the author of several works, who for many years had lived among books and manuscripts. He has no trace of typographical vision, but all words “sound in his ear.” He can neither read nor compose without articulating; as the interest of his book or work grows upon him he speaks aloud—“He must hear himself.” In his dreams there are few or no visual images; he hears his voice and that of his interlocutors: “His dreams are auditory.” None of my words, even when semi-concrete, evoked visual images.

In most cases the auditory type is not clear. For very general terms the heard word alone exists, but in proportion as the concrete is approached, the sound is accompanied by an image; thus returning upon our former type.

It is worth while to note that the term flatus vocis “nomina,” first employed in the Middle Ages and which has since become the formula of Nominalism, seems by its nature to indicate that it was originally invented by persons who belonged to the auditory type, and I may even hazard an hypothesis. The typographical visual type did not exist (printing not being invented); it is true that a substitute might have existed in the graphic visual type (reading of manuscripts). But considering that in the Middle Ages instruction was essentially oral, that learning came rather through listening than by reading, that the oratorical jousts and arguments were daily and interminable, it is undeniable that the conditions for developing the auditory type were highly favorable here.

I need hardly say that the three types described above are rarely met with in the pure and complete form. As a rule a mixed type prevails: a concrete image for certain words, and typographical vision, or auditory images, for others. To sum up: all cases seem to be capable of reduction to the following: (1) The word heard; beyond this, nil (we shall subsequently have to examine this “nothing”); (2) typographical vision alone; (3) the same, accompanied by a concrete image; (4) the word heard, accompanied invariably by a concrete image.

4. Prior to the commencement of this inquiry I felt much hesitation on one point: should one in questioning use general words or general propositions? I decided in favor of words because these are brief, simple, isolated, and undisguised, and have the advantage of being understood directly, while they in no way suggest to the subject what line he is to follow.

I still however felt scruples in the matter. Was not the investigation as conducted on these lines a little artificial? In point of fact, general terms most frequently occur as members of a phrase, co-operating with others, and connected with them by certain relations. I therefore recommenced my inquiry, using the same method, but replacing words by phrases. The general propositions employed are purposely trite, to avoid contradiction, and to ascertain the immediate mental state. They were as follows:

Cause invariably precedes effect.—Infinity has several meanings.—Is Space infinite?—Has Time any limits?—Law is a necessary relation.—I need not enlarge upon the results: they are precisely the same as for words. In every case, and for each person, there is one predominating word which absorbs all the content of the phrase, and is a substitute for it. On this the instantaneous mental operation is concentrated.

If of the concrete type, the subject sees images. In the second phrase, e. g., everything converges on the word infinity. Replies: Sensation of obscurity and depth, vague luminous circles, a sort of cupola, a never-receding horizon, etc. If a typographic visualist, the printed sentence is seen less clearly than the simple words: “in minute characters; no capitals”; some persons glimpse it rapidly: others see only “the principal word printed.”

For the pure auditory type, the answer is always very simple. “I hear the sentence, I see absolutely nothing.”

The new method therefore simply confirms the previous observations, with no variations. This identity of result seems to me to militate against a distinction admitted by many authors. In the classical treatises a distinction is made between “necessary ideas” and “necessary truths” (I use their terms uncritically), i. e., general concepts and general propositions. Example: cause, principle of causality. In my opinion there is merely a difference of form between the two positions, the one psychological, the other logical. A concept is a judgment in a state of envelopment, or of result. The proposition is a word in the state of development. The difference is not material, but formal; it is the passage from synthesis to analysis.

I thought that after an interval of two years it might be interesting to repeat the same inquiry on the same people; but the results were not encouraging in this direction. Some, remembering the previous investigation, declared that “they felt themselves influenced beforehand.” Others, who had a more vague recollection (perhaps because they did not understand the object of the inquiry) gave answers analogous to their former replies. In short, notwithstanding the lapse of time, and change of circumstances, each seemed to be consistent with his former self.

I must admit that in the preceding research the psychological nature of the concepts was studied under a particular aspect. This objection was made at the London Psychological Congress[84] by the President, Professor Sidgwick, whose remarks may be summarised as follows:

First, Professor Sidgwick believes that the act of suddenly calling attention to a word, in a person not accustomed to introspective observation, evokes a response which does not exactly correspond to the state ordinarily aroused by such words. In his own particular case he has found that the images evoked (usually visual) were extremely feeble, but that when he dwelt upon them they were enlivened. Secondly, the images vary a great deal according to the terms employed; for example, when he is occupied with mathematical and logical trains of thought, he sees only the printed words. If he is engaged upon the subject of political economy, the general terms sometimes have for their concomitants extremely fantastic images: like value, for instance, which is accompanied by the indistinct and fragmentary image of a man placing something upon the pan of a balance. Thirdly, when for such words as infinity, relation, etc., the subject answers nothing, the only conclusion justified is that the subject is incapable of describing the confused elements which exist in his consciousness. Fourthly, Professor Sidgwick’s own experience points to the conclusion that my types may succeed each other in the same person.

On this last point—the co-existence of several modes of conception in the same person—I am quite in agreement with Professor Sidgwick, and my own data, drawn up from personal observations, would provide me with sufficient evidence. At the same time the object of my investigation was not to determine the manner in which each individual conceives, but the forms under which men as a whole think of concepts. Nor did I profess to follow the work of the mind when it resolves its general ideas into concretes, when it makes coin out of its bank-notes, but only to seize the subjacent labor that accompanies the current and facile use of general terms, in speaking, listening, reading or writing. No doubt it would be advisable to treat the subject in another manner by studying—no longer the momentary state that corresponds with the presence of the concept in consciousness—but the stable organised turn of mind due to a long habit of dealing with concepts. To this end it would be desirable more especially to question mathematicians and metaphysicians. My data are neither numerous nor clear enough to permit of my hazarding any dictum on this subject. Some mathematicians have told me that they invariably require a figured representation, a construction, and that even when these are considered as purely fictitious their support is indispensable to the train of reasoning. Contra those who think geometrically, there are others who think algebraically, eliminating all configuration, or construction, and proceeding by simple analysis with the aid of signs: which (with the necessary corrections and descriptions) would bring the first under the concrete, and the second under the audito-motor type. Among metaphysicians the typographical visual type seems largely to predominate. One (who is well known) belongs to the pure auditory type. All this, however, is inadequate; the investigation would have to be followed out, by and upon others.

A young Russian doctor, M. Adam Wizel, who was interested in the subject, put the same questions (following the method indicated above) to persons in the hypnotic state. Admitting the unconscious mental activities to preponderate in this state he asked whether by this procedure it would not be possible to penetrate farther into the unknown substrate of consciousness. His experiments were undertaken at the Salpêtrière, in Charcot’s clinique, upon six women—hysterics of the first order. The subjects were first put into a state of somnambulism, then after a preliminary explanation were questioned, as above. After getting the answers Wizel ordered the subjects to forget all that had happened, and then woke them. He now began again in the waking state, asking the same questions, so that he was able to compare the answers given successively in the two cases. They are nearly always clearer and more explicit during somnambulism than during the waking state, as may be judged by the following example (taken from the third observation):

Where the replies are concrete in the two cases I note a tolerable analogy between them. M. Wizel (who eliminated all doubtful cases, and any accompanied by crises) never encountered the typographical visual type, nor the pure auditory type, in his experiments. His six hysterics belong to the concrete type, with the predominance of visual images—much more rarely of motor images, provoked by the word “force.” The answer “nothing” is very frequent; less so, however, during somnambulism than during the waking state.

II.

We now reach the most obscure and difficult part of our subject. Among the nine hundred and odd replies collected, the one most frequently met with is “nothing.” There is no observation in which it does not occur at least once: in the majority of cases it is found one, two, three, four, or more times. If I take the word cause, the formula “I have no representation,” forms fifty-three per cent. of the total of answers collected; the rest saw the printed word or some concrete image; e. g., a stone falling, horses drawing, or other simulacra, of which several have already been enumerated. It is the same with all the other highly abstract terms (time, infinity, etc.). So that to return to the question which was to be the exclusive object of our investigation,—“Is the general idea when thought, read, or heard, accompanied by anything in consciousness?”—we may reply, an image, a typographical vision, or nothing. We must now inquire, what this nothing is, for it must be something.

We are face to face with the problem which the pure Nominalists attacked, when they took this nil in its proper sense. Were there indeed any who really pretended that we could have in mind words, and words only—nothing besides? This is a historical problem into which it is useless for us to enter. It is possible that some may have pushed their reaction against the extravagances of realism even to this point, but the thesis is totally insupportable; for at that rate there would be no difference between a general term, and a word of any unknown language: the latter is the pure flatus vocis, a sound that evokes nothing. If, on the other hand, by word we mean sign, everything changes, since the sign implies and envelops something. Such appears to me to be the true interpretation.[85]

So that for the cases which alone concern us for the moment, i. e., those in which the reply was “Nothing,” there are two elements, the one existing in consciousness (word heard or auditory image), the other subconscious, but not therefore invalid and inaccurate. Hence we must penetrate into the obscure region of the unconscious, in order to apprehend the something which gives to the word its signification, its life, its power of substitution.

Leibnitz wrote: “Most frequently, e. g., in any prolonged analysis, we have no simultaneous intuition of all the characteristics or attributes of a thing: in their place we use signs. In actually thinking, we are accustomed to omit the explanation of these signs by reference to what they signify—knowing or believing that we have this explanation in our power: but we do not judge the application, or explanation, of the words to be positively necessary.... I have termed this method of thinking, blind or symbolic. We employ it in algebra, in arithmetic, and in fact universally:” which is equivalent to saying that potential knowledge is stored up beneath the general or abstract terms; nor should we be surprised at finding this doctrine in the man who first introduced the idea of the unconscious into philosophy.

To determine the rôle of this inevitably active, albeit silent, factor is a difficult enterprise, and one that is necessarily inaccurate,—since it amounts to the translation of obscure and enveloped states into the clear and analytical language of consciousness. The simplest procedure is to examine how we arrive at the comprehension of general terms.[86] Set a page of a philosophical work before the eyes of a novice or of a man wholly ignorant of the subject. He understands nothing. The only method that will render it intelligible is to take the general or abstract terms one after the other, and translate them into concrete events, into facts of current experience. This labor demands an hour or two. In proportion to the progress of the novice, the translation is effected more quickly; it becomes superfluous for certain terms; and finally but a few minutes are required for the comprehension of a page. Untrained minds are often surprised, on reading a sentence consisting of abstract terms, at understanding each word, and yet not knowing what the whole means. This signifies that they have not beneath each word potential knowledge sufficient to establish a connexion or relation between all the terms, and giving meaning to them. Apart from those who are familiar with abstraction by natural gift or by habit, it is incontestable that to the vast majority the reading of an abstract page is a slow and painful and very fatiguing process. This is because each word exacts an act of attention, an effort, which corresponds to labor in the unconscious or subconscious regions. When this labor becomes useless, and we think, or appear to think, by signs alone all goes rapidly and easily.

In short, we learn to understand a concept as we learn to walk, dance, fence, or play a musical instrument: it is a habit, i. e., an organised memory. General terms cover an organised, latent knowledge which is the hidden capital without which we should be in a state of bankruptcy, manipulating false money or paper of no value. General ideas are habits in the intellectual order. Suppression of effort corresponds with perfected habit; as again with perfect comprehension.

What occurs each time we have in consciousness merely the general term, is only a particular case of a very common psychological fact: as follows:—The useful work is carried on below consciousness, and above its surface only results, indications, or signs appear. The facts enumerated above are all taken from motor activity. Their equivalents abound in the domain of feeling. The “causeless” states of joy or sorrow, which are frequent in the sound man, and still more in the invalid, are only the translation into consciousness of ignored organic dispositions operating in obscurity. What gives intensity and duration to our passions is not the consciousness we have of them, but the depth of the roots by which they plunge into our being, and are organised in our viscera, and subsequently in our brain. They are no more than the expression of our constitution, permanent, or momentary. We might run over the whole province of psychology, with variations on the same theme. I do not propose to do so here, but would simply recall that every state of consciousness whatsoever (percept, image, idea, feeling, passion, volition) has its substrate; that the concept reduced to the bare word is but another case of the same kind, and in no wise peculiar: that to believe that there is nothing more than the word, because it alone exists in consciousness, is to seize only the superficial and visible part of the event,—perhaps, all things considered, the least part. This unconscious substratum, this organised and potential knowledge, gives not merely value, but an actual denotation to the word,—like harmonics superadded to the fundamental note.

To conclude: we think not with words in the strict sense (flatus vocis) but with signs. Symbolic thought, which is a purely verbal operation, is sustained, co-ordinated, vivified, by potential knowledge and unconscious travail. To this it must be added that potential knowledge is a genus, of which the concept is only a species. All memory can be reduced to latent knowledge, organised, susceptible of revival, but all memory is not material for concepts. The man who knows many languages even when not speaking them, the naturalist capable of identifying millions of specimens while not classifying them, have a very extended potential knowledge, but it is all concrete. The potential knowledge which underlies concepts consists in a sum of characters, qualities, extracts, which are the less numerous in proportion as the concept approximates to pure symbolism: in other terms what underlies the concept is an abstract memory, a memory for abstracts.

In my opinion, a large measure of the obscurities and dissensions which prevail as to the nature of concepts, arises from the fact that the rôle of unconscious activity has for centuries been misunderstood or forgotten,—psychology having confined itself exclusively to consciousness: and while its action is universally admitted to-day for all other manifestations of mental life—instincts, percepts, feelings, volition, etc.—it is still excluded from the domain of concepts. The whole of the foregoing discussion is an essay towards its reintegration.

Need we add that the opinion adopted as to the nature of the unconscious matters little in the present connexion? On this point there are, we know, two principal hypotheses. According to the one it is a purely physiological event, and can be reduced to unconscious cerebration. According to the other, the unconscious is still a psychical fact; whether it be an affective rather than a representative state, or a complex of little, scattered consciousnesses, isolated, evanescent, with no linkage to the self, or, again, an organisation or sequence of states, which forms another current coexistent with that of clear consciousness. These theories, and others which I omit, have nothing to do with our purpose. It is sufficient to recognise unconscious activity as a fact, without any explanation, and this position would seem to be incontestable.

We have seen that abstraction, in proportion as it ascends and strengthens, separates itself more and more from the image, until finally, at the moment of pure symbolism, the separation becomes antagonism. This is because there is fundamentally, and from the outset, an opposition of nature and procedure between the two. The ideal of the image is an ever-growing complexity, the ideal of abstraction an ever-growing simplification, since the one is formed by addition, the other by subtraction.

To the man who is gifted with a rich internal vision, the shape of people, of monuments, of landscapes, surges up clearly and well defined: under the influence of attention, and with time, details are added,—the representation completes itself, and approaches more and more completely to the reality. So too with internal audition: witness the musician who hears ideally every detail of a symphony.

The contrary holds for abstraction. “There is,” says Cournot, “an analysis which separates objects, and an analysis which distinguishes without isolating them. The use of the refracting prism is an instance of the analysis which separates or isolates. If, instead of isolating the rays so as to cause them to describe different trajectories, they are made to traverse certain media which have the property of extinguishing a definite color, we are able to distinguish without isolating.”[87]

Abstraction belongs to this last type, with intervention of the process described by Cournot. Attention brings a feature into relief; inattention, and voluntary inhibition, act as extinguishers to the other characteristics.

Let us pass from theory to practice. This antagonism is of current observation, almost a banality, whenever men of imagination are confronted with abstract thinkers. We must of course exclude those who by a rare gift of nature (Goethe), or by the artifice of education, are capable of handling the image and concept alternately.

Let us take the artists as type of the imaginative thinker: the novelist, poet, sculptor, painter, musician, etc. Each dreams of an organic, living work, a complex. Some with words, others with forms, others with sounds; realists with the aid of minute detail, classics by means of general sketches; all make for the same end. Music again, which from its nature seems a thing apart, is an architecture of sounds of amazing complexity, often exciting contradictory states of mind.

Among abstract thinkers (theorists, scientists) the tendency is always towards unity, law, generalities—towards simplification: by what is fundamental and essential, if the man be genuine; by shifting and accidental features, if he is a charlatan. The mathematician and the pure metaphysician have usually a distaste for facts, for multiplicity of detail. A writer whose name has escaped me says: “Every scientist smells of the cadaver.” This is our thesis, under the form of an image. Abstract thought is a cadaver. It would be more just, though less picturesque, to say a skeleton; for scientific abstraction is the bony framework of phenomena.

The antagonism between the image and the idea is thus fundamentally that of the whole and the part. It is impossible to be at the same time an abstract thinker and an imaginative thinker, because one cannot simultaneously think the whole and the part, the group and the fraction; and these two habits of mind while not absolutely exclusive are antagonistic.

In conclusion, have we general ideas, or merely general terms? It must first be remarked that the expressions, “general ideas or notions,” “concepts,” are equivocal or rather multivocal. We have seen that concepts differ widely in their psychological nature according to their degree, having but one characteristic in common—that of being extracts. It is therefore chimerical to attempt to include them all under a single definition. To take the highest only, as most frequently debated, some say, “There are no general ideas but only general terms.” To others the general idea is only an indefinite series of particular ideas, or “a particular idea that the mind proposes as the first stake in a forward march.”[88] To others it is a system of tendencies accompanied or not by the possibility of images.[89] For my own part I prefer the formula of Höffding[90]: “General ideas exist therefore in the sense that we are able to concentrate the attention on certain elements of the individual idea, so that a weaker light falls on the other elements.”

This is the sole mode of existence that can be legitimately conceded to them.

In regard to the higher concepts we have endeavored to show that they have their distinctive psychological nature: on the one hand a clear and conscious element which is always the word, and sometimes in addition the fragmentary image; on the other an obscure and unconscious factor,—without which, nevertheless, symbolic thought is only a mechanism turning in the air, and producing naught but phantoms.