Association Of Ideas

Earlier Theory.—In the long and erudite Note D**, appended by Sir W. Hamilton to his edition of Reid’s Works, many anticipations of modern statements on association are cited from the works of ancient or medieval thinkers; and for Aristotle, in particular, the glory is claimed of having at once originated the doctrine and practically brought it to perfection.1 As translated by Hamilton, but without his interpolations, the classical passage from the De Memoria et Reminiscentia runs as follows:—

“When, therefore, we accomplish an act of reminiscence, we pass through a certain series of precursive movements, until we arrive at a movement on which the one we are in quest of is habitually consequent. Hence, too, it is that we hunt through the mental train, excogitating from the present or some other, and from similar or contrary or coadjacent. Through this process reminiscence takes place. For the movements are, in these cases, sometimes at the same time, sometimes parts of the same whole, so that the subsequent movement is already more than half accomplished.”

The passage is obscure, but it does at all events indicate the various principles commonly termed contiguity, similarity and contrast. Similar principles are stated by Zeno the Stoic, by Epicurus (see Diog. Laert. vii. § 52, x. § 32), and by St Augustine (Confessions, x. e. 19). Aristotle’s doctrine received a more or less intelligent expansion and illustration from the ancient commentators and the schoolmen, and in the still later period of transition from the age of scholasticism to the time of modern philosophy, prolonged in the works of some writers far into the 17th century, Hamilton adduced not a few philosophical authorities who gave prominence to the general fact of mental association—the Spaniard Ludovicus Vives (1492-1540) especially being most exhaustive in his account of memory.

In Hobbes’s psychology much importance is assigned to what he called, variously, the succession, sequence, series, consequence, coherence, train of imaginations or thoughts in mental discourse. But not before Hume is there express question as to what are the distinct principles of association. John Locke had, meanwhile, introduced the phrase “Association of Ideas” as the title of a supplementary chapter incorporated with the fourth edition of his Essay, meaning it, however, only as the name of a principle accounting for the mental peculiarities of individuals, with little or no suggestion of its general psychological import. Of this last Hume had the strongest impression; he reduced the principles of association to three—Resemblance, Contiguity in time and place, Cause and (or) Effect. Dugald Stewart put forward Resemblance, Contrariety, and Vicinity in time and place, though he added, as another obvious principle, accidental coincidence in the sounds of words, and further noted three other cases of relation, namely, Cause and Effect, Means and End, Premisses and Conclusion, as holding among the trains of thought under circumstances of special attention. Reid, preceding Stewart, was rather disposed to make light of the subject of association, vaguely remarking that it seems to require no other original quality of mind but the power of habit to explain the spontaneous recurrence of trains of thinking, when become familiar by frequent repetition (Intellectual Powers, p. 387).

Hamilton’s own theory of mental reproduction, suggestion or association is a development, greatly modified, of the doctrine expounded in his Lectures on Metaphysics (vol. ii. p. 223, seq.), which reduced the principles of association first to two—Simultaneity and Affinity, and these further to one supreme principle of Redintegration or Totality. In the ultimate scheme he posits no less than four general laws of mental succession concerned in reproduction: (1) Associability or possible co-suggestion (all thoughts of the same mental subject are associable or capable of suggesting each other); (2) Repetition or direct remembrance (thoughts coidentical in modification, but differing in time, tend to suggest each other); (3) Redintegration, direct remembrance or reminiscence (thoughts once coidentical in time, are, however, different as mental modes, again suggestive of each other, and that in the mutual order which they originally held); (4) Preference (thoughts are suggested not merely by force of the general subjective relation subsisting between themselves, they are also suggested in proportion to the relation of interest, from whatever source, in which they stand to the individual mind). Upon these follow, as special laws:—A, Primary—modes of the laws of Repetition and Redintegration—(1) law of Similars (Analogy, Affinity); (2) law of Contrast; (3) law of Coadjacency (Cause and Effect, &c.); B, Secondary—modes of the law of Preference, under the law of Possibility—(1) laws of Immediacy and Homogeneity; (2) law of Facility.

The Associationist School.—This name is given to the English psychologists who aimed at explaining all mental acquisitions, and the more complex mental processes generally under laws not other than those which have just been set out as determining simple reproduction. Hamilton, though professing to deal with reproduction only, formulates a number of still more general laws of mental succession—law of Succession, law of Variation, law of Dependence, law of Relativity or Integration (involving law of Conditioned), and, finally, law of Intrinsic or Objective Relativity—as the highest to which human consciousness is subject; but it is in a sense quite different that the psychologists of the so-called Associationist School intend their appropriation of the principle or principles commonly signalized. As far as can be judged from imperfect records, they were anticipated to some extent by the experientialists of ancient times, both Stoic and Epicurean (cf. Diogenes Laertius, as above). In the modern period, Hobbes is the first thinker of permanent note to whom this doctrine may be traced. Though, in point of fact, he took anything but an exhaustive view of the phenomena of mental succession, yet, after dealing with trains of imagination, or what he called mental discourse, he sought in the higher departments of intellect to explain reasoning as a discourse in words, dependent upon an arbitrary system of marks, each associated with, or standing for, a variety of imaginations; and, save for a general assertion that reasoning is a reckoning—otherwise, a compounding and resolving—he had no other account of knowledge to give. The whole emotional side of mind, or, in his language, the passions, he, in like manner, resolved into an expectation of consequences, based on past experience of pleasures and pains of sense. Thus, though he made no serious attempt to justify his analysis in detail, he is undoubtedly to be classed with the associationists of the next century. They, however, were wont to trace their psychological theory no further back than to Locke’s Essay. Bishop Berkeley was driven to posit expressly a principle of suggestion or association in these terms:—“That one idea may suggest another to the mind, it will suffice that they have been observed to go together, without any demonstration of the necessity of their coexistence, or so much as knowing what it is that makes them so to coexist” (New Theory of Vision, § 25); and to support the obvious application of the principle to the case of the sensations of sight and touch before him, he constantly urged that association of sound and sense of language which the later school has always put in the foreground, whether as illustrating the principle in general or in explanation of the supreme importance of language for knowledge. It was natural, then, that Hume, coming after Berkeley, and assuming Berkeley’s results, though he reverted to the larger inquiry of Locke, should be more explicit in his reference to association; but he was original also, when he spoke of it as a “kind of attraction which in the mental world will be found to have as extraordinary effects as in the natural, and to show itself in as many and as various forms” (Human Nature, i. 1, § 4). Other inquirers about the same time conceived of association with this breadth of view, and set themselves to track, as psychologists, its effects in detail.

David Hartley in his Observations on Man, published in 1749 (eleven years after the Human Nature, and one year after the better-known Inquiry, of Hume), opened the path for all the investigations of like nature that have been so characteristic of English psychology. A physician by profession, he sought to combine with an elaborate theory of mental association a minutely detailed hypothesis as to the corresponding action of the nervous system, based upon the suggestion of a vibratory motion within the nerves thrown out by Newton in the last paragraph of the Principia. So far, however, from promoting the acceptance of the psychological theory, this physical hypothesis proved to have rather the opposite effect, and it began to be dropped by Hartley’s followers (as F. Priestley, in his abridged edition of the Observations, 1775) before it was seriously impugned from without. When it is studied in the original, and not taken upon the report of hostile critics, who would not, or could not understand it, no little importance must still be accorded to the first attempt, not seldom a curiously felicitous one, to carry through that parallelism of the physical and psychical, which since then has come to count for more and more in the science of mind. Nor should it be forgotten that Hartley himself, for all his paternal interest in the doctrine of vibrations, was careful to keep separate from its fortunes the cause of his other doctrine of mental association. Of this the point lay in no mere restatement, with new precision, of a principle of coherence among “ideas,” but in its being taken as a clue by which to follow the progressive development of the mind’s powers. Holding that mental states could be scientifically understood only as they were analysed, Hartley sought for a principle of synthesis to explain the complexity exhibited not only in trains of representative images, but alike in the most involved combinations of reasonings and (as Berkeley had seen) in the apparently simple phenomena of objective perception, as well as in the varied play of the emotions, or, again, in the manifold conscious adjustments of the motor system. One principle appeared to him sufficient for all, running, as enunciated for the simplest case, thus: “Any sensations A, B, C, &c., by being associated with one another a sufficient number of times, get such a power over the corresponding ideas (called by Hartley also vestiges, types, images) a, b, c, &c., that any one of the sensations A, when impressed alone, shall be able to excite in the mind b, c, &c., the ideas of the rest.” To render the principle applicable in the cases where the associated elements are neither sensations nor simple ideas of sensations, Hartley’s first care was to determine the conditions under which states other than these simplest ones have their rise in the mind, becoming the matter of ever higher and higher combinations. The principle itself supplied the key to the difficulty, when coupled with the notion, already implied in Berkeley’s investigations, of a coalescence of simple ideas of sensation into one complex idea, which may cease to bear any obvious relation to its constituents. So far from being content, like Hobbes, to make a rough generalization to all mind from the phenomena of developed memory, as if these might be straightway assumed, Hartley made a point of referring them, in a subordinate place of their own, to his universal principle of mental synthesis. He expressly put forward the law of association, endued with such scope, as supplying what was wanting to Locke’s doctrine in its more strictly psychological aspect, and thus marks by his work a distinct advance on the line of development of the experiential philosophy.

The new doctrine received warm support from some, as Law and Priestley, who both, like Hume and Hartley himself, took the principle of association as having the like import for the science of mind that gravitation had acquired for the science of matter. The principle began also, if not always with direct reference to Hartley, yet, doubtless, owing to his impressive advocacy of it, to be applied systematically in special directions, as by Abraham Tucker (1768) to morals, and by Archibald Alison (1790) to aesthetics. Thomas Brown (d. 1820) subjected anew to discussion the question of theory. Hardly less unjust to Hartley than Reid or Stewart had been, and forward to proclaim all that was different in his own position, Brown must yet be ranked with the associationists before and after him for the prominence he assigned to the associative principle in sense-perception (what he called external affections of mind), and for his reference of all other mental states (internal affections) to the two generic capacities or susceptibilities of Simple and Relative Suggestion. He preferred the word Suggestion to Association, which seemed to him to imply some prior connecting process, whereof there was no evidence in many of the most important cases of suggestion, nor even, strictly speaking, in the case of contiguity in time where the term seemed least inapplicable. According to him, all that could be assumed was a general constitutional tendency of the mind to exist successively in states that have certain relations to each other, of itself only, and without any external cause or any influence previous to that operating at the moment of the suggestion. Brown’s chief contribution to the general doctrine of mental association, besides what he did for the theory of perception, was, perhaps, his analysis of voluntary reminiscence and constructive imagination—faculties that appear at first sight to lie altogether beyond the explanatory range of the principle. In James Mill’s Analysis of the Phenomena of the Human Mind (1829), the principle, much as Hartley had conceived it, was carried out, with characteristic consequence, over the psychological field. With a much enlarged and more varied conception of association, Alexander Bain re-executed the general psychological task, while Herbert Spencer revised the doctrine from the new point of view of the evolution-hypothesis. John Stuart Mill made only occasional excursions into the region of psychology proper, but sought, in his System of Logic (1843), to determine the conditions of objective truth from the point of view of the associationist theory, and, thus or otherwise being drawn into general philosophical discussion, spread wider than any one before him its repute.

The Associationist School has been composed chiefly of British thinkers, but in France also it has had distinguished representatives. Of these it will suffice to mention Condillac, who professed to explain all knowledge from the single principle of association (liaison) of ideas, operating through a previous association with signs, verbal or other. In Germany, before the time of Kant, mental association was generally treated in the traditional manner, as by Wolff. Kant’s inquiry into the foundations of knowledge, agreeing in its general purport with Locke’s, however it differed in its critical procedure, brought him face to face with the newer doctrine that had been grafted on Locke’s philosophy; and to account for the fact of synthesis in cognition, in express opposition to associationism, as represented by Hume, was, in truth, his prime object, starting, as he did, from the assumption that there was that in knowledge which no mere association of experiences could explain. To the extent, therefore, that his influence prevailed, all inquiries made by the English associationists were discounted in Germany. Notwithstanding, under the very shadow of his authority a corresponding, if not related, movement was initiated by J.F. Herbart. Peculiar, and widely different from anything conceived by the associationists, as Herbart’s metaphysical opinions were, he was at one with them, and at variance with Kant, in assigning fundamental importance to the psychological investigation of the development of consciousness, nor was his conception of the laws determining the interaction and flow of mental presentations and representations, when taken in its bare psychological import, essentially different from theirs. In F.E. Beneke’s psychology also, and in more recent inquiries conducted mainly by physiologists, mental association has been understood in its wider scope, as a general principle of explanation.

The associationists differ not a little among themselves in the statement of their principle, or, when they adduce several principles, in their conception of the relative importance of these. Hartley took account only of Contiguity, or the repetition of impressions synchronous or immediately successive; the like is true of James Mill, though, incidentally, he made an express attempt to resolve the received principle of Similarity, and through this the other principle of Contrast, into his fundamental law—law of Frequency, as he sometimes called it, because upon frequency, in conjunction with vividness of impressions, the strength of association, in his view, depended. In a sense of his own, Brown also, while accepting the common Aristotelian enumeration of principles, inclined to the opinion that “all suggestion may be found to depend on prior coexistence, or at least on such proximity as is itself very probably a modification of coexistence,” provided account be taken of “the influence of emotions and other feelings that are very different from ideas, as when an analogous object suggests an analogous object by the influence of an emotion which each separately may have produced before, and which is, therefore, common to both.” To the contrary effect, Spencer maintained that the fundamental law of all mental association is that presentations aggregate or cohere with their like in past experience, and that, besides this law, there is in strictness no other, all further phenomena of association being incidental. Thus in particular, he would have explained association by Contiguity as due to the circumstance of imperfect assimilation of the present to the past in consciousness. A. Bain regarded Contiguity and Similarity logically, as perfectly distinct principles, though in actual psychological occurrence blending intimately with each other, contiguous trains being started by a first (it may be, implicit) representation through Similarity, while the express assimilation of present to past in consciousness is always, or tends to be, followed by the revival of what was presented in contiguity with that past.

The highest, philosophical interest, as distinguished from that which is more strictly psychological, attaches to the mode of mental association called Inseparable. The coalescence of mental states noted by Hartley, as it had been assumed by Berkeley, was farther formulated by James Mill in these terms:—

“Some ideas are by frequency and strength of association so closely combined that they cannot be separated; if one exists, the other exists along with it in spite of whatever effort we make to disjoin them.”—(Analysis of the Human Mind, 2nd ed. vol. i. p. 93.)

J.S. Mill’s statement is more guarded and particular:—

“When two phenomena have been very often experienced in conjunction, and have not, in any single instance, occurred separately either in experience or in thought, there is produced between them what has been called inseparable, or, less correctly, indissoluble, association; by which is not meant that the association must inevitably last to the end of life—that no subsequent experience or process of thought can possibly avail to dissolve it; but only that as long as no such experience or process of thought has taken place, the association is irresistible; it is impossible for us to think the one thing disjoined from the other.”—(Examination of Hamilton’s Philosophy, 2nd ed. p. 191.)

It is chiefly by J.S. Mill that the philosophical application of the principle has been made. The first and most obvious application is to so-called necessary truths—such, namely, as are not merely analytic judgments but involve a synthesis of distinct notions. Again, the same thinker sought to prove Inseparable Association the ground of belief in an external objective world. The former application, especially, is facilitated, when the experience through which the association is supposed to be constituted is understood as cumulative in the race, and transmissible as original endowment to individuals—endowment that may be expressed either, subjectively, as latent intelligence, or, objectively, as fixed nervous connexions. Spencer, as before suggested, is the author of this extended view of mental association.

Modern Criticism.—Of recent years the associationist theory has been subjected to searching criticism, and it has been maintained by many writers that the laws are both unsatisfactorily expressed and insufficient to explain the facts. Among the most vigorous and comprehensive of these investigations is that of F.H. Bradley in his Principles of Logic (1883). Having admitted the psychological fact of mental association, he attacks the theories of Mill and Bain primarily on the ground that they purport to give an account of mental life as a whole, a metaphysical doctrine of existence. According to this doctrine, mental activity is ultimately reducible to particular feelings, impressions, ideas, which are disparate and unconnected, until chance Association brings them together. On this assumption the laws of Association naturally emerge in the following form:—(1) The law of Contiguity.—“Actions, sensations and states of feeling, occurring together or in close connexion, tend to grow together, or cohere, in such a way that, when any one of them is afterwards presented to the mind, the others are apt to be brought up in idea” (A. Bain, Senses and Intellect, p. 327). (2) The law of Similarity.—“Present actions, sensation, thoughts or emotions tend to revive their like among previous impressions or states” (A. Bain, ibid. 457. Compare J.S. Mill, Logic, ii. p. 440, 9th ed.). The fundamental objection to (1) is that ideas and impressions once experienced do not recur; they are particular existences, and, as such, do not persevere to recur or be presented. So Mill is wrong in speaking of two impressions being “frequently experienced.” Bradley claims thus to reduce the law to “When we have experienced (or even thought of) several pairs of impressions (simultaneous or successive), which pairs are like one another; then whenever an idea occurs which is like all the impressions on one side of these pairs, it tends to excite an idea which is like all the impressions on the other side.” This statement is destructive of the title of the law, because it appears that what were contiguous (the impressions) are not associated, and what are associated (the ideas) were not contiguous; in other words, the association is not due to contiguity at all.

Proceeding to the law of Similarity (which in Mill’s view is at the back of association by contiguity), and having made a similar criticism of its phrasing, Bradley maintains that it involves an even greater absurdity; if two ideas are to be recognized as similar, they must both be present in the mind; if one is to call up the other, one must be absent. To the obvious reply that the similarity is recognized ex post facto, and not while the former idea is being called up, Bradley replies simply that such a view reduces the law to the mere statement of a phenomenon and deprives it of any explanatory value, though he hardly makes it clear in what sense this necessarily invalidates the law from a psychological point of view. He further points out with greater force that in point of fact mere similarity is not the basis of ordinary cases of mental reproduction, inasmuch as in any given instance there is more difference than similarity between the ideas associated.

Bradley himself bases association on identity plus contiguity:—“Any part of a single state of mind tends, if reproduced, to re-instate the remainder,” or “any element tends to reproduce those elements with which it has formed one state of mind.” This law he calls by the name “redintegration,” understood, of course, in a sense different from that in which Hamilton used it. The radical difference between this law and those of Mill and Bain is that it deals not with particular units of thoughts but with universals or identity between individuals. In any example of such reproduction the universal appears in a particular form which is more or less different from that in which it originally existed.

Psychophysical Researches.—Bradley’s discussion deals with the subject purely from the metaphysical side, and the total result practically is that association occurs only between universals. From the point of view of empirical psychologists Bradley’s results are open to the charge which he made against those who impugned his view of the law of similarity, namely that they are merely a statement—not in any real sense an explanation. The relation between the mental and the physical phenomena of association has occupied the attention of all the leading psychologists (see Psychology). William James holds that association is of “objects” not of “ideas,” is between “things thought of”—so far as the word stands for an effect. “So far as it stands for a cause it is between processes in the brain.” Dealing with the law of Contiguity he says that the “most natural way of accounting for it is to conceive it as a result of the laws of habit in the nervous system; in other words to ascribe it to a physiological cause.” Association is thus due to the fact that when a nerve current has once passed by a given way, it will pass more easily by that way in future; and this fact is a physical fact. He further seeks to maintain the important deduction that the only primary or ultimate law of association is that of neural habit.

The objections to the associationist theory are summed up by G.F. Stout (Analytic Psychol., vol. ii. pp. 47 seq.) under three heads. Of these the first is that the theory as stated, e.g. by Bain, lays far too much stress on the mere connexion of elements hitherto entirely separate; whereas, in fact, every new mental state or synthesis consists in the development or modification of a pre-existing state or psychic whole. Secondly, it is quite false to regard an association as merely an aggregate of disparate units; in fact, the form of the new idea is quite as important as the elements which it comprises. Thirdly, the phraseology used by the associationists seems to assume that the parts that go to form the whole retain their identity unimpaired; in fact, each part or element is ipso facto modified by the very fact of its entering into such combination.

The experimental methods now in vogue have to a large extent removed the discussion of the whole subject of association of ideas, depending in the case of the older writers on introspection, into a new sphere. In such a work as E.B. Titchener’s Experimental Psychology (1905), association is treated as a branch of the study of mental reactions, of which association reactions are one division.

Bibliography.—See Psychology; and the works of Bradley, Stout, and James, above quoted, and general works on psychology: articles in Mind (passim); A. Bain, Senses and Intellect (4th ed., 1894), and in Mind, xii. (1887) pp. 237-249; John Watson, An Outline of Philosophy (1898); H. Höffding, Hist. of Mod. Philos. (Eng. trans., Lond., 1900), Psychologie in Umrissen auf Grundlage der Erfahrung (2nd ed., Leipzig, 1893); Jas. Sully, The Human Mind (1892), and Outlines of Psych. (Lond., 1892); E.B. Titchener, Outline of Psych. (New York, 1896), and in his trans. of O. Külpe’s Outlines of Psych. (New York, 1895,); Jas. Ward in Mind, viii. (1883), xii. (1887), new series ii. (1893), iii. (1894); G.T. Ladd, Psychology, Descriptive and Explanatory (Lond., 1894); C.L.C. Morgan, Introd. to Comparative Psych. (Lond., 1894); W. Wundt, Princip. of Physiol. Psych. (Eng. trans., 1904), Human and Animal Psych. (Eng. trans., 1894), pp. 282-307; Outlines of Psych. (Eng. trans., 1897); E. Claparède, L’Association des idées (1903). For associationism in Greek philosophy see J.I. Beare, Greek Theories of Elementary Cognition (Oxford, 1906), part iii. §§ 14, 43 seq.