The function of the feelings, as the cause of association—The law of affective association, conceived as general, and as local—I. Function of unconscious feeling: ancestral or hereditary unconsciousness; personal unconsciousness arising from cœnæsthesia; personal unconsciousness arising from the events of our life—Law of transference by contiguity, by resemblance: wide or narrow—II. Function of the conscious feelings: accidental cases, permanent cases, exceptional or rare cases.

In this chapter we have still to deal with the relation between the feelings and the memory, but under quite another form, seeing that we have to study the feelings as a cause. Instead of establishing, as we have hitherto done, that there is such a thing as a real memory of the feelings, our present aim is to determine the function of states of feeling in the recalling of recollections and the association of ideas. Their importance as a hidden factor of revivability has been recognised by several contemporary writers,^[110] some even having a tendency to exaggerate it.

We know that the theory of the association of ideas has been reduced to two fundamental laws—that of contiguity and that of resemblance. I may remark, without insisting on the fact, that they are not of the same nature; the first, being purely mechanical, the result of experience, while the second supposes, in addition to this, a certain degree of mental labour, for a complete correspondence between two states is rarely met with, and can only be grasped in consequence of a dissociation or abstraction operating on the raw materials. These two laws are purely intellectual; they are regulative principles deduced from facts, nothing more. They are rather descriptive than explanatory. They reveal the mechanism, but not the motive force. They suppose something beyond, unless we admit that ideas are psychic atoms endowed with some mysterious attraction or affinity. With regard to the determining reasons, they are dumb. Now it cannot be doubted that in many cases (not all) the cause of the association is to be found in a permanent or momentary state of the feelings.

The writers who have pointed out this influence (often efficacious though latent) have conceived this superior law, which might be called the Law of Feeling, in two different ways, some as absolute and universal, others as partial and local. I take my stand among the latter.

1. Fouillée (as also, it seems, Horwicz) has maintained the former thesis. “The association of ideas presupposes that of the emotions, and, with the latter, that of the impulses. The dominant impulse awakens, by association, the secondary impulses tending in the same direction. The tie which unites them is the unity of an aim in relation to which the impulses are medium, the unity of an effect in relation to which they are co-operating forces.... The laws of association and contrast are what dominate the association of the feelings” (loc. cit., p. 221). I shall not be suspected of hostility to the essential spirit of this thesis, since the present work is only one long vindication of the primordial nature of tendencies. But unless we are led astray by the mirage of unity at any price, it is impossible to admit that every association supposes an emotional factor as a determining reason. Not to speak of the numerous cases resulting from contiguity, in which the part played by the feelings is very doubtful, I find an important category of purely intellectual associations, where the intervention of the feelings appears to me impossible to verify. Is it likely that the mathematician and the metaphysician who connect together a long series of abstractions have an emotional state as the support and vehicle of their thought, whether the latter be discursive or constructive? I do not see, in theory or in fact, any reason for admitting this, unless we wish to involve the love of truth; and in any case this would only be a primum movens, not the direct and immediate cause of the associations.

2. The influence of emotional states must be stated as a principal, but not an exclusive cause. It is summed up in what Shadworth Hodgson has called the “Law of Interest.” In a past event, everything is not equally interesting; in its revival, all the elements are not equally active, the most emotional bringing the others with them. “Two processes are constantly going on in redintegrations. The one a process of corrosion, melting, decay; the other a process of renewing, arising, becoming.... Those parts of the object, however, which possess an interest, resist this tendency to gradual decay of the whole object.”^[111] Coleridge rightly says that “The true practical general law of association is this: that whatever makes certain parts of a total impression more vivid or distinct than the rest will determine the mind to recall these, in preference to others equally linked together by the common condition of contemporaneity or continuity. But the will itself, by confining and intensifying the attention, may arbitrarily give vividness or distinctness to any object whatsoever.”^[112] The power attributed by Coleridge to the attention and the will finally resolves itself into an emotional state as ultimate cause, and from it alone can an increase of intensity be derived.

I shall insist no further on these generalities, as it will be more instructive to determine by means of a few details the influence of the emotional life on the memory. To this end I shall divide our study into two parts, the function of unconscious feeling, and that of the conscious feelings.

I.

It is not always easy to determine positively the degree in which unconscious feeling influences the memory in order to awaken it, or to connect ideas with one another. I have purposely employed the vague term “unconscious feeling” as prejudging nothing with regard to its nature. We may form any conception we like of it, either considering it as purely physiological or assigning to it a psychological character—that of a consciousness diminishing to infinity. Both these opinions have their partisans, but this does not matter as regards the following considerations. In this unconscious feeling I distinguish three strata, passing from the deepest upwards, from the more obscure to the less obscure.

1. Hereditary or ancestral unconsciousness. I mention this merely for the sake of completeness. It would consist in the influence of certain modes of feeling, inherited and fixed in a race, which might, without our knowing it, exercise some sway over our associations. Under this form, at least, it appears to me extremely hypothetical. Laycock^[113] (1844), one of the founders of the physiology of the unconscious, attempts to explain by this means certain national and individual tastes; the Hungarians are supposed to like plains because these appeal to the ancestral recollection of the Mongolian steppes, their primæval home. Herbert Spencer, who, however, has not occupied himself much with the influence of sentiments on the association of ideas, says incidentally that, in the impression produced by a landscape, “along with the immediate sensations, there are partially excited the myriads of sensations that have been in time past received from objects such as those presented; further, there are also excited certain deeper, but now vague combinations of states which were organised in the race during barbarous times, when its pleasurable activities were chiefly among the woods and waters.”^[114] Schneider assumes this ancestral revivification in every æsthetic perception. We shall return to this subject in Part II. The predatory tastes of primitive man would explain certain agreeable associations (e.g., the pleasure of constructing a bloodthirsty drama) which contrast with the habits of civilised man.

These facts seem to me reducible to a single explanation. There are in every man latent tendencies, which may remain latent throughout his life, but may also be awakened and revealed by some accidental occurrence. They might be called hereditary, since they are found in an inherited organism; but it would be quite as correct to call them innate. In any case, it is very difficult to prove that they are a survival, and above all a resurrection of once existing tendencies.

2. Personal unconsciousness arising from cœnæsthesia, i.e., from the internal sensations collectively. This imperceptibly brings us down to consciousness, from the moment when the affective state can be verified without induction. A certain disposition, a certain manner of feeling, is the direct and immediate cause of association. It is permanent or transitory. If permanent, it answers to the temperament or disposition. As the subject is cheerful, melancholy, erotic, or ambitious, an unconscious selection is exercised on the ideas arising in consciousness; an artist and a practical man, in face of the same object, have two totally distinct modes of association. If transitory, it corresponds, in the same individual, to states of health and sickness, to changes of age; each one of these distinct states produces a distinct selection. The unity of certain dreams, in spite of the apparent difference of associations, has its easily discovered cause in an organic or affective disposition—fatigue, depression, oppression, circulatory or digestive troubles, sexual excitement. The simplicity and frequency of these facts will permit us to dispense with insisting upon them.

3. Personal unconsciousness, a residuum of affective states connected with anterior perceptions, or with events of our life. This emotional residuum, although latent, is no less active, and can be recovered by analysis. This case, one of the most important connected with our subject, has recently been studied by Lehmann^[115] under the name of displacement (Verschiebung) of the sentiments, and by Sully under the name of transference of feelings; this second denomination seems to me the clearer and more accurate of the two.

Under its most general form—for its mechanism is not always the same—the law of transference consists in directly attributing a sentiment to an object which does not itself cause it. There is no transference in the sense that the feeling is detached from the primary event in order to be connected with another; but there is a moment of generalisation or extension of the sentiment, which spreads like a drop of oil. This transference can be symbolically represented. Let us represent an intellectual state by A, and by s the affective state which accompanies it; A by association excites B, C, D, E, etc., while s is successively transferred to B, C, D, E, etc. Thus we have, first, A
s, B, C, D, E, etc., then A,B,C,D,E, etc.
s, so that C, D, or E can directly produce s quite as A can, and even without the assistance of A. “The feeling is excited without the mediacy of the particular presentative element of which it was originally a concomitant” (Sully).^[116] This law of transfer is of sufficient importance to delay us a little, because it plays a somewhat important part in the formation of complex emotions, and we shall need to recall it more than once. Besides, it does not always operate in the same manner. I distinguish two principal cases, according as the transfer is the result of contiguity, or of resemblance.

Transference by Contiguity.—When intellectual states have co-existed and formed a complex by contiguity, and one of them has been accompanied by a special sentiment, any one of these states has a tendency to excite the same sentiment.

We can find numerous and simple examples in common life. The lover transfers the sentiment at first called forth by the person of his mistress to her clothes, her furniture, her house. For the same reason, hatred and jealousy vent their rage on inanimate objects belonging to the enemy. In absolute monarchies the reverence in which the king’s person is held is transferred to the throne, to the emblems of his power, to everything directly or indirectly connected with his person. The following charming passage from Herbert Spencer relates to a less simple case of the same nature: “The cawing of rooks is not in itself an agreeable sound; musically considered, it is very much the contrary. Yet the cawing of rooks usually produces pleasurable feelings—feelings which many suppose to result from the quality of the sound itself. Only the few who are given to self-analysis are aware that the cawing of rooks is agreeable to them because it has been connected with countless of their greatest gratifications—with the gathering of wild flowers in childhood; with Saturday afternoon excursions in schoolboy days; with midsummer holidays in the country, when books were thrown aside and lessons were replaced by games and adventures in the field; with fresh, sunny mornings, in after years, when a walking excursion was an immense relief from toil. As it is, this sound, though not causally related to all these multitudinous and varied past delights, but only often associated with them, rouses a dim consciousness of these delights; just as the voice of an old friend, unexpectedly coming into the house, suddenly raises a wave of that feeling which has resulted from the pleasures of past companionship.” We must remark that in the transfer by contiguity, which, by its very nature is automatic, the intellectual states act as causes, since the extension of the sentiment is subordinated to them.

Transference by Resemblance.—When an intellectual state has been accompanied by a vivid sentiment, every similar or analogous state tends to excite the same feeling.

In this psychological fact lies the secret of the emotion of love, tenderness, antipathy, respect, which we feel towards a person at first sight, without apparent reason, and which we are apt to put down to the account of instinct. But those who devote themselves to the analysis of their own consciousness will discover, in many cases, a more or less close resemblance to a person who inspires, or has inspired, us with love, tenderness, antipathy, or respect. A mother may feel a sudden sympathy for a young man who is like her dead son, or even merely of the same age. The explanation of many of these cases lies in an unconscious state which is not easy to seize, but which, if it returns to consciousness (a process in which the will is only very indistinctly concerned), elucidates everything. There are also so-called instinctive fears, without conscious motives, which, by going a little below the surface, can be referred to the same explanation.^[117]

This transfer can take place in two ways, one narrow, the other broad. The narrow method rests on resemblance only: B resembles A, the perception or representation of whom is or was accompanied by a certain feeling; the transfer goes no further. The broader method rests on analogy, and has a much wider scope; it passes from one individual to several—to a class or classes. “A friend of mine,” says Lehmann, “hated dogs; circumstances forced him to keep one; he attached himself to this animal, and gradually his feeling of sympathy spread to the whole canine race” (loc. cit.). This possibility of a limited transfer has been a social and moral factor of the first importance; it has allowed of the extension of the sympathetic sentiments from the small exclusive clan to more and more distant groups—the tribe, the nation, the human race. The wider transfer has been the great agent of the transition from particularism to universalism.^[118]

II.

From the unconscious states to the affective states, of which the subject is fully conscious, the transition is made gradually and through doubtful forms; but whether obscure, semi-obscure, or clear, their influence remains the same. Among the numerous cases in which the association of ideas depends on a conscious affective disposition we may distinguish three groups:—

1. Individual, accidental, ephemeral cases. These can be reduced to a single formula: when two or more states of consciousness have been accompanied by the same emotional state, they tend to be associated with one another. Emotional resemblance unites and intertwines disparate impressions. It is a case of association by resemblance, but not intellectual; impressions are associated because they resemble one another in a common emotional colouring, not qua impressions. Examples of this are abundant. L. Ferri (in his Psychologie de l’Association, where, by-the-bye, he does not note this emotional law) tells us that one day, being stung by a fly, he suddenly remembered a child seen by him, long ago, when himself very young, on its death-bed. Whence this sudden vision? “In the first place, I was lying on my bed, then I had been stung by a fly, and lastly, the sight of the corpse had caused in me a deep sadness, while, at this same moment, I also happened to be very sad.” Association through emotional identity or resemblance is of frequent occurrence in dreams, as has been already said. I remember, among many others, a dream whose unity, in spite of the apparent incoherence of the association, was due to a general sense of fatigue. A road without milestones stretched before me, of which I was about to complete the last stage; steep mountains kept rising one behind another; my eyes were wearied with trying to catch sight of the longed-for town on the horizon; and every time I wished to inquire the way I had to speak a foreign language which I understand but imperfectly, and in which it is very difficult for me to express myself. I awoke, feeling a general aching and heaviness of all the limbs. Sully relates a dream whose unity consisted in a sense of anxiety and vexation. He was suddenly called upon to give a lecture on Herder; he began by stammering out some generalities; then he was addressed by one of his audience, who suggested difficulties to him; then the entire assembly broke up tumultuously. One of his children, who had seen, for the first time, the great clock at Strasburg, and, after an interval of two days, the Swiss glaciers, dreamed on the following night that the figures of the clock were walking about on the snow. In this case the groundwork of the dream is a feeling of admiration or surprise.

2. Permanent and stable cases; to be met with everywhere, because involved in the structure of the human mind. They are fixed in language. When dealing with the expression of the emotions (Chap. IX.) we met with “the principle of association of analogous sensations,” formulated by Wundt. Adapting it to our present subject, we may say that sensations imbued with a similar emotional colouring are easily associated, and strengthen each other. Nothing can differ more in nature than our external sensations (except smell and taste), and the qualities which they make known to us; the data of sight and hearing have no resemblance to one another as cognitions of the external world, yet we speak of sombre voices, clear voices, screaming colours, coloured music. We associate sight with thermal sensations, as when we speak of warm or cold colours. Taste also has its share—bitter reproaches, subacid criticism. Finally, touch, as Sully-Prudhomme has remarked, is perhaps the most abundant source of associations between the idea of the physical sensation and an emotional state; compare the terms touching, hard, tender, heavy, firm, solid, harsh, penetrating, poignant, piquant, etc. At the bottom of all these associations there is a common emotional colouring which both causes and supports them. Perhaps it would be more accurate to class them among the cases of semi-conscious emotional influence; but we have already said that our division into conscious and unconscious factors is superficial and of no great importance.

3. Exceptional and rare cases. Flournoy, in his important work on “coloured hearing,” rightly explains this anomaly by “emotional association.” We know that several hypotheses on the origin and cause of this phenomenon have been constructed. On the embryological one, it would be the result of an incomplete differentiation between the sense of sight and that of hearing; a survival, we are told, from a primitive epoch when this state was the rule. On the anatomical theory, we suppose anastomoses between the cerebral centres of the visual and auditory sensations. Besides these we have the physiological theory, or that of nervous irradiation, and the psychological, or that of association. I do not inquire if all cases may be reduced to a single explanation; certainly most seem reducible to association. We are not, however, dealing with any and every form of association—it must be a psychological one, as Flournoy was the first to remark. “By emotional association, I mean that which establishes itself between two impressions, not on account of a qualitative resemblance (for the two may be as disparate as sound and colour), nor in virtue of their regular and frequent concurrence in the consciousness, but in consequence of the analogy between their emotional characteristics. Each sensation or perception possesses, in fact, along with its objective quality or its intellectual content, a sort of subjective coefficient, springing from the roots which it sends down into our being, and from the peculiar way in which it impresses, pleases or displeases, excites or subdues us, in a word, makes our whole nature vibrate. We can conceive how two absolutely heterogeneous sensations, incommensurable as far as their objective content is concerned, such as a colour and the sound i, may be comparable with one another and resemble each other more or less, by virtue of vibrations produced by them in the organism; and by the same process of thought it is conceivable that this emotional factor might become a link between the two, an associative bond by means of which one awakens the other.”^[119]

Let us add that we meet, though much more rarely, with cases of coloured smell and taste, and even, it appears, of coloured pain.^[120] This abnormal association between determinate colours and determinate tastes, odours, pains, may be explained in the same manner.

Shall we attribute to the same cause a fact, ascertained (exceptionally, however) in the case of certain hysterical subjects in the hypnotic state, which may be described as follows? The excitation of certain circumscribed regions of the body immediately causes to arise in the mind either ideas or feelings which are imperiously imposed on the consciousness and last as long as the excitement which provoked them. Pitres, who has made an extended study of these “zones idéogènes,”^[121] has discovered about twenty scattered over various parts of the body in the same subject. The effect of excitation (by friction or compression) is always the same in the same individual, but varies from one individual to another, which excludes the hypothesis of a previously existing mechanism. Among the feelings aroused by this procedure I note sadness, cheerfulness, anger, fear, eroticism, piety, ecstasy.

Most writers have limited themselves to the statement of the fact, without attempting to explain it. Pitres alone proposes the hypothesis of auto-suggestion, which is not far from an association of ideas. Must we admit an original fortuitous coincidence between a local bodily modification and a certain emotional state (or idea), whence an association through contiguity fixed and strengthened by repetition, so as to become indissoluble? Or can it be that friction and compression produce in certain subjects peculiar organic reactions, capable of exciting a special emotional state? We can only hazard conjectures.

In conclusion: the influence of emotional dispositions on the memory is great, and continually active; it contributes to the revival and association of ideas. Now, the emotional states are not entities, but modes of consciousness, the psychical equivalents of certain organic reactions—visual, vaso-motor, or muscular; so that the emotional influence reduces itself to all this. And is all this to be reduced to movements? A marked tendency towards this opinion is visible in several of our contemporaries. Fouillée, as we saw a little while ago, refers all association to that of impulses; Horwicz does the same under another form (loc. cit.). He places in the feelings the basis of all conservative memory, and the basis of all feelings in motion. “We recall our emotional state in proportion as we can reproduce the movements implied in it.” By a different road—that of experiment—Münsterberg has attempted to show that so-called successive association is reducible to a rapid simultaneity, and that, if we suppress all movements during the reception of impressions, memory is much diminished and reproduction difficult.^[122] It is true that his experiments were limited to articulatory movements.

I merely indicate in passing this general hypothesis. Whether admitted or not, the relation between the feelings and the association of ideas, though often misunderstood, has been indubitably proved by a mass of facts which, in spite of their heterogeneous character, all point to the same conclusion.