The process of abstraction, an experimental study

Front matter

Book /Li:

SMITHSONIAN DEPOSIT.

IN

Contents

PAGE

Introduction 74

I. Literature of the Problem 76

Introduction

The decade that is just now drawing to a close has witnessed a notable extension of the field of psychological research. In the beginnings of modern psychology the field of experiment was seldom extended beyond the domain of sensation. Progress in physics and physiology made it possible to subject our sensations to experiment, and for some time the sensory processes received the chief share of the attention of psychologists. It was not long, however, before the emotions began to receive their due amount of consideration and the invention of the plethysmograph opened the way to a new line of research. But only within the last ten years has the experimental study of such higher processes of thought, as abstraction, commenced to develop. The impetus to this new development has come mainly from Professor Oswald Kiilpe at the University of "Wiirzburg. The present research, although its origins are not to be traced to the school of Wiirzburg, belongs to the field which Professor Kiilpe and his students have so admirably developed. Our problem was to study the mental processes involved in the formation of our abstract ideas. It is indeed true that the very existence of such ideas has been called in question. Still we may at present assume, for the purpose of stating our problem, that it is possible for the mind to perceive a series of objects, and to recognize some one quality or group of qualities as recurring constantly in every member of the series. The botanist examining a set of specimens will classify them according to certain characteristics which mark off the genera and species. The group of characteristics constitutes what may be termed his concept of the genus or species that he has segregated. Of each species he has a more or less definite "concept," by which he can represent to himself a number of specimens, no two of which are precisely the same. Such "concepts," whatever may be their real nature, are facts of conscious experience; we form them and use them incessantly. But what after all is the "concept"? What is the process of its formation? This is the problem of the present research.

The history of the problem dates back to the days of the Greek philosophy; but only within the last few years has it been subjected to an experimental investigation. The more recent literature is of immediate interest for our present problem. The metaphysical discussions, valuable as they are within their own sphere, bear only indirectly on the experimental question. Consequently, only the experimental literature bearing in some manner on the process of abstraction has been analyzed. Not every allusion in the extensive psychological literature of the day could be picked out, but a general account of the important pieces of experimental work from Galton to the present time has been given. The individual studies have been analyzed with some completeness because the history of the literature is an integral part of the evidence on one point of the present study, viz.: Is there or is there not a distinction between thought and imagery ; and if so, in what sense is thought to be interpreted?

LITERATURE OF THE PROBLEM.

The first experiments which in any way approached the domain of our abstract ideas were made by Francis Galton in 1878. In the Proceedings of the Royal Institution of Great Britain for 1879 1 was published his memoir on Generic Images. In this article he refers to an earlier one in the Journal of the Anthropological Institute for 1878. 2 The bearing of these experiments on abstraction is suggested rather than direct. But they have become the basis of the now famous composite-image theory of ideas, and are therefore deserving of mention. Galton described in this article the composite photographs which he had just succeeded in obtaining. These he compared to "our general impressions." Just what he meant by "our general impressions" is not clear; but he congratulates himself that his explanation coincides with that of Professor Huxley in his work on Hume. "I am rejoiced," he says, "to find that from a strictly physiological side this explanation is considered to be the true one by so high an authority, and that he has, quite independently of myself, adopted a view which I also entertained, and had hinted at in my first description of composite portraiture, though there was no occasion at that time to write more explicitly about it." 3

Huxley's meaning is clearer, and to him we may turn for an outline of the theory. In the above-mentioned work on Hume, the following quotation gives a clear idea of the generic image theory of general concepts.

2 This article is mainly concerned with the technique of composite photographs.

3 Proceedings of the Royal Institution of Great Britain, 9, 1879-1881, p. 166.

"Now when several complex impressions which are more or less different from one another — let us say that out of ten impressions in each, six are the same in all, and four are different from all the rest — are successively presented to the mind, it is easy to see what must be the nature of the result. The repetition of the six similar impressions will strengthen the six corresponding elements of the complex idea, which will therefore acquire greater vividness; while the four differing impressions of each will not only acquire no greater strength than they had at first, but in accordance with the law of association, they will appear at once, and will thus neutralize one another.

"This mental operation may be rendered comprehensible by considering what takes place in the formation of compound photographs — where the images of the faces of six sitters, for example, are each received on the same photographic plate for a sixth of the time requisite to take one portrait. The final result is that all those points in which the six faces agree are brought out strongly, while all those in which they differ are left vague; and thus what may be termed a generic portrait of the six, in contradiction to a specific portrait of any one, is produced. . . .

"The generic ideas which are formed from several similar, but not identical, complex experiences are what are commonly called abstract or general ideas; and Berkeley endeavored to prove that all general ideas are nothing but particular ideas annexed to a certain term which gives them a more extensive signification, and makes them recall, upon occasion, other individuals which are similar to them. Hume says that he regards this as 'one of the greatest and the most valuable discoveries that has been made of late years in the republic of letters,' and endeavors to confirm it in such a manner that it shall be 'put beyond all doubt and controversy.'

' ' I may venture to express a doubt whether he has succeeded in his object; but the subject is an abstruse one; and I must content myself with the remark, that though Berkeley's view appears to be largely applicable to such general ideas as are formed after language has been acquired, and to all the more abstract sort of conceptions, yet that general ideas of sensible objects may nevertheless be produced in the way indicated, and may exist independently of language." 4

It would thus seem that Huxley's theory — and probably Galton's also — is that only our abstract ideas of sensible objects are to be compared with the composite photographs. Galton points out that the mind in forming its generic images is much less perfect in its mechanism than the camera. "Our mental generic composites are rarely defined; they have that blur in excess which photographic composites have in a small degree and their background is crowded with faint and incongruous imagery. The exceptional effects are not overmastered, as they are in the photographic composites, by the large bulk of ordinary effects." 5

The experiments on composite photographs were not experiments on abstract ideas — they merely suggested a theory of general concepts. Nor did Galton's later experiments on mental imagery 6 approach very much nearer the problem. They called general attention to mental imagery and perhaps stimulated the next investigation of any importance 7 which was made by Ribot.

In October, 1891, M. Ribot published in the Revue Philosophique 8 his "Enquete sur les idees generates. " This he afterwards amplified in the fourth chapter of his book, L'E 'volution des idees generales. 9 His problem was this : At the moment of thinking or reading or hearing a general term, what is there in consciousness — immediately and without reflection? On the basis of the imagery which his subjects reported he classified

4 Huxley, David Hume. New York, 1879, pp. 92-94.

s Loc. cit., p. 169.

e Cf. Inquiries into Human Faculty, 1883, Section on Mental Imagery.

7 The article entitled ' ' Observations on General Terms, " by S. E. Wiltse, in the American Journal of Psychology, 3, 1890, pp. 144-148, was only tentative and contained no definite results.

them into: (1) The concrete type (visual or muscular imagery of an object). (2) Typographic visual type (visual image of the printed word). (3) Auditory type. A great many of his subjects said that they had nothing in mind. For example, fifty per cent, of the answers to the imagery of the word 'cause' said that the subjects had represented to themselves nothing at all. M. Ribot then asked himself the question, "What is this "nothing"? The word alone? No. Otherwise there would be no difference between a general term and a word of a language that one did not understand. The word is a sign of some object. We have learned the mental habit of designating many objects that have some point of agreement by this symbol. The objects designated lie hidden and are unconsciously represented by this general term. "General ideas are habits in the intellectual order. ' ' Our higher concepts consist of two elements — one clear and conscious, and this is always the word which may at times be accompanied by some shred of imagery. The other element is obscure and unconscious. M. Ribot refrains from saying precisely what this obscure and unconscious element is. From the context it would seem that he means the unconscious trace left by the habitual use of the word to designate various objects. A word of criticism may be said in passing. M. Ribot 's interpretation of this "nothing," which accompanies the perception of a general term, is purely theoretical and is not based on any published data given by his subjects. Furthermore, he has not followed out his sign theory to its logical consequences. For every sign, we have on the one hand the object signified and on the other, the signification. Smoke has on the one hand fire, of which it is a sign, and on the other a signification in the mind of the observer. If, then, general terms are signs, they have on the one hand the objects that they signify, and on the other a signification. This signification, as M. Ribot admits, is not the image and not the word itself. It certainly is not the unconscious factor he speaks of — for he would scarcely maintain that his subjects were not conscious of the meaning of the word. It is therefore a clearly conscious mental process distinct from both the image and the word.

M. Ribot's work comes closer to being an experimental study of the problem than Galton's experiments on composite photographs. Had he examined the state of mind of his subjects when they lacked imagery and not trusted to theory on that point, he would have carried his investigation into the heart of the problem of abstract ideas.

After Galton and Ribot the study of mental imagery took up no small portion of the time and labors of psychologists. But the extensive literature on imagery is not directly concerned with the problem of abstraction.

An excellent piece of work in this field of research is that of William Chandler Bagley of Cornell University. 10 He undertook to study the effect of imperfectly formed words on the perception of spoken sentences, and parts of sentences. The words and sentences were first recorded by a phonograph, the initial, middle, or final consonants being unpronounced. The subject listened to the phonograph, and was called upon to repeat what he heard, and analyze the mental processes he experienced. The section of the work which bears upon our problem is that entitled "The Conscious Process Involved in the Apperception of Spoken Symbols." In comparison with the later German experiments, Dr. Bagley 's are remarkable for the very frequent occurrence of imagery of one kind or another. His subjects in perceiving the meaning of sentences report with surprising frequence the presence of visual, auditory or kinesthetic imagery. This might be due to the fact that Dr. Bagley 's subjects were capable of sharper introspection than the German psychologists. Still this can hardly be the case. The German psychologists, among whom are such men as Professor Kiilpe, cannot be supposed to be lacking in the power of introspection. Another possible explanation is that Bagley laid special stress upon the report about imagery and in that way developed in his subjects a "task" to associate definite images with the given sentences.

10 "The Apperception of the Spoken Sentence: a Study in the Psychology of Language." American Journal of Psychology, 12, 1900-01, pp. 80- 134.

Though his subjects generally experienced imagery in the perception of the meaning of a sentence, still he finds cases in which this imagery is lacking. One factor which has to do with the lack of imagery is that "familiarity with the sentence sometimes militates against a clear and direct reference on the part of the observer." In this his results agree with those of Dr. Taylor reported below. 11 Bagley's general conclusion is that,

"The consciousness concomitant with the apperception of auditory symbols is made up of sensational and affective elements — some peripherally, some centrally aroused — in connections which vary in character with different individuals and under different conditions. These connections are arranged in patterns which change rapidly into one another, and are in general transitory and fleeting. When the attention is directed to the peripherally excited elements exclusively — when the external stimuli occupy the burning point of apperception — the meaning which they as symbols should convey is not clearly apperceived. When the attention is directed upon the centrally aroused ideas which the symbols suggest, the 'meaning' is apperceived, but errors and lapses in the stimuli are apt to pass unnoticed." (p. 125.)

He thinks that Stout goes too far in suggesting the existence of representative mental contents different from "visual, auditory, tactual, and other experiences." He thinks that his experiments lead him to no such conclusion. "From the series of observations which were made in the course of our experiments, no conscious 'stuff' was found which could not be classed as sensation or affection, when reduced to its ultimates by a rigid analysis. Neither do our experiments show that there is in the apperception of spoken sentences such a thing as 'imageless apprehension.' " (p. 126.)

Still Dr. Bagley finds something which he does not feel justified in putting down as either imagery or feeling. To this something which is not imagery or feeling and still has to

ii See p. 87.

do with the understanding of the sentence, he applies the name 'mood.' "We may say with Stout that the new is referred to a mental 'system/ in so far as such a system is a mood, an attitude, a tendency, an adaptation. The mind adjusts itself uniformly to uniform conditions: this seems to be the essence of the apperceptive 'mood.' When C, in the sentence "The play was bad," interpreted 'play' as a drama, her mind adapted itself in a degree to the drama environment. This was not necessarily a focal reference to a given play, but the mind was in the dramatic 'mood.' Should particular parts of a typical play-environment have been ideally reproduced, the situation would only have been reinforced. Should certain verbal ideas such as 'drama,' 'theaters/ 'Shakespeare,' etc., have been reproduced in consciousness, either visually, auditorily or kinesthetically, these ideas would have been constituents of the dramatic 'mood,' but not necessarily the fundamental constituents. The fundamental constituents may and do vary from time to time. Only very seldom can they be called constant, and the 'constant supplements' which we have noticed are instances of such occasions. The fact that the focal constituents of the apperceptive consciousness are not necessarily consistent with the situation represented bears testimony to this point of view. "There was not room for a stove in the corner"; with this sentence one observer imaged distinctly a stove in the corner of a small, otherwise bare room. His own surprise at the inconsistency of this imagery was shown by his exclamation upon recording the introspection: "But there was a stove there!" (p. 127.)

A 'mood/ therefore, is something that has to do with the past experience of the subject in regard to the words of the sentence that is understood. Just what it is, as a present psychical state, Dr. Bagley does not say. It is the revival of past experience. It is not mental imagery, although mental imagery enters in as a partial element in the complex termed 'mood.' If this is so, what is that present psychical state in the mood, which is neither imagery nor feeling ? It is not past experience, for the past is not present. It is not revived imagery and feeling, for Dr. Bagley admits that the mood contains something besides imagery and feeling. It therefore seems that Dr. Bagley has found something more than he is willing fully to recognize. His experiments, like those of the later German writers, reveal the existence of an imageless mental content. Just what we call this is not of prime importance. But its existence should be recognized.

In 1901 appeared Marbe's 12 experimental study of judgment. His main problem is not that of the present work, but it is an early attempt to apply the experimental method to supra-sensuous mental processes. He also mentions, as a side issue, his attempt to develop the experiments of Ribot. The work was done in Professor Kiilpe's laboratory at the University of Wurzburg, where Dr. Marbe was at the time Privatdozent of Philosophy. The method is essentially the same as that of the later Wurzburg experiments which are described below. His final conclusion in regard to the judgment is that any special mental process — a word or gesture or image — may become a judgment. Taken literally, this conclusion is not borne out by the experiments. What they seem to prove is that a judgment may be signified by a variety of different processes. And this may be the author's meaning (cf. Chapter III). The perception of a judgment, however, is not a sensation or an image or a feeling, or anything that can be pointed out in consciousness. The perception of a judgment is a knowing — a "Wissen" (cf. p. 17).

In his concluding remarks on experimentation in the domain of logic Dr. Marbe mentions Ribot 's work, L' Evolution des idees generates, and refers to some similar experiments of his own on ideas and imagery. His meaning is not clear to me, so I quote entire the brief account of his work in this field (pp. 99- 101).

"Seit den Zeiten des Sokrates hat man angenommen, da8 den Begriffen im Bewusstsein ausser den zugehorigen Worten irgend etwas direkt entspreche, d. h., daB es neben diesen Worten

12 K. Marbe, Experimentell-psychologische Untersuclmngen iiber das Urteil. Leipzig, 1901.

84 University of California Publications in Psychology. [ Vo1 - 1

psychische Gebilde gabe, welche der Gesamtheit der Gegenstande, auf welche sich die Worte beziehen, korrespondieren sollen. Diese Gebilde wurden urspriinglich, wie gelegentlich noch heute im Gegensatz zu den Worten, die nur Zeiehen ihrer Bedeutungen sind, als Abbilder derselben aufgefasst, indem sie die gemeinschaftlichen Merkmale ihrer Gegenstande im Bilde enthalten sollten. Solche psychischen Gebilde hat man spater je nach dem Grade der Abstraktheit, den man ihnen zuschrieb, bald als Gemeinbilder, bald als allgemeine Vorstellungen, bald als Begriffsvorstellungen bezeichnet. Obgleich, wie bekannt, ihre Existenz schon im Altertum nnd Mittelalter bestritten und in der Neuzeit hauptsachlich dureh Berkeley bekampft wurde, so halt man doch auch heute noch vielfach in der einen oder anderen Form an derselben fest. Auch die Frage, ob es solche psychologische Aquivalente der Begriffe giebt, ist eine rein psychologische, und ihre Behandlung sollte nicht, wie es in der Regel geschieht, mit logischen Untersuchungen vermiseht werden. Die Aussagen unserer Versuchspersonen liber die Bewusstseinsvorgange, welche sie nach dem Erleben von Urteilsworten und Urteilssatzen zu Protokoll gaben, enthalten iibrigens nichts von solchen Parallelerscheinungen der Begriffe, ebensowenig, wie die wertvollen Untersuchungen von Ribot, 13 in welchen dieser Forscher einer Reihe von Beobachtern Substantiva zurief, um sich dann von ihnen sagen zu lassen, was die gehorten Worte fur Erlebnisse auslosten. Ich selbst habe mehreren Beobachtern ca. 20 Substantiva zugerufen und mir dann von ihnen berichten lassen, was fur Erlebnisse die zugerufenen Worte erzeugten. Dann gab ich denselben Beobachtern der Reihe nach verschiedene Karten in die Hand, auf welchen jeweils ein Substantivum aufgedruckt war, wahrend sie nach einigen Augenblicken die Erlebnisse zu Protokoll geben mussten, die dureh den Anblick der gedruckten Worte in ihnen ausgelost wurden. Endlich stellte ich ihnen die Auf gabe, die Begriffe: Baum, Volk, Gesellschaft, Zeit u. a. zu denken und mir dann die Resultate ihrer Bemiihungen mitzuteilen. In alien diesen Fallen zeigten die Proto-

13 L'Evolution des idees generates. Paris, 1897, p. 127ff.

kolle nichts von Begriffsvorstellungen u. dergl. Die Erlebnisse der Beobachter bestanden vielmehr aussehliesslich in Wahrnehmungen, Vorstellungen und Bewusstseinslagen, die teilweise gefiihlsbetont, teilweise o-hne jeden Gefiihlston verliefen. Man wird also wohl sagen diirfen, daB es keine psychologischen Aquivalente der Begriffe im Sinne der Begriffsvorstellungen giebt. Jedenfalls aber sehen wir leicht ein, daB auch diese Frage experimentell behandelt werden kann und muB und daB sie die Logik weiter nicht tangiert."

The obscurity arises from the fact that it is not perfectly clear whether Marbe merely denies the existence of the general images which Bishop Berkeley 14 termed abstract ideas, or that he claims that there are neither general images nor universal concepts.

It would seem, however, that Marbe found no evidence for the existence of a general image in the understanding of the words given to his subjects. He did find, however, Wahrnehmungen, Vorstellungen, and Bewusstseinslagen. This latter is a term introduced by Mayer and Orth 15 to represent certain "states of mind" which are more or less refractory toward all

14 <■ ' Whether others have this wonderful faculty of abstracting their ideas, they can best tell; for myself I dare be confident I have it not. I find indeed I have indeed a faculty of imagining, or representing to myself, the ideas of those particular things I have perceived, and of variously compounding and dividing them. I can imagine a man with two heads, or the upper parts of a man joined to the body of a horse. I can consider the hand, the eye, the nose, each by itself or separated from the rest of the body. But then whatever hand or eye I imagine, it must have some particular shape and colour. Likewise, the idea of man that I frame to myself must be either of a white, or a black, or a tawny, a straight or a crooked, a tall or a low, or a middle-sized man. I can not by any effort of thought conceive the abstract idea described [in his previous account of the abstract ideas of the traditional logic]. And it is equally impossible for me to form the abstract idea of motion distinct from the moving, and which is neither swift nor slow, curvilinear nor rectilinear; and the like may be said of all other abstract general ideas whatsoever." A Treatise concerning the Principles of Human Knowledge. Introduction, 10, pp. 141-142, vol. I of Fraser's Oxford (1871) Edition of his Works.

It is evident from the context that Bishop Berkeley does not distinguish between the mental image and the abstract concept — between what is termed by the later Wiirzburg School the Vorstellung and the Begriff.

is < ' Zur qualitativen Untersuchung der Association. ' ' Zeitschrift fur Psychologie und Physiologie, 26, 1901, p. 6.

analysis — and in which images (Vorstellungen) are not to be found. In a later study 10 Dr. Orth attempted to show that the Bewusstseinslage was not a state of feeling.

That part of Dr. Narziss Ach's work Uber die Willenstdtig- Jceit und das Denken, 17 which refers to thought has a direct relation to this line of work, which we may consider as originating in Ribot's "Enquete sur les idees generales. " The Bewusstseinslage of Marbe appears under the name of Bewtisstheit. This author too has recognized the existence of mental states in which there "could not be detected any such phenomenal elements as visual, auditory, or kinesthetic sensations or memory pictures of such sensations which qualitatively determined the mental content reported as knowledge" (p. 210). There are often present along with such states of consciousness words or fragments of words. "Such a presence of kinesthetic or auditory kinesthetic images may well be the cause of the widely disseminated hypothesis that our thought continually takes place in an inner speech or adequate visual, acoustic, or similar kinds of memory images. Against such a view one must point to the fact that there are very complex contents in which, as already mentioned, the partial contents are consciously represented in their manifold opposing relations and still these individual contents are not expressed by any adequate vocal designations and the like — and indeed, it is absolutely impossible that they should be" (p. 215).

The question then arises, what are these imageless states of consciousness ? This Dr. Ach explains by an example : ' ' Every idea which is given in consciousness, for example, the word 'bell' puts, as is well known, a number of ideas in readiness, with which it stands in associative connections. This putting of ideas in readiness, or stimulation of tendencies to reproduction, suffices for the conscious representation of what we call is Dr. Johannes Orth, Gefuhl und Bewusstseinslage. Sammlung von Abhandlungen aus dem Gebiete der Padagogischen Psychologie und Physiologie, edited by Ziegler and Ziehen. Vol. 6, No. 4. Berlin, 1903, cf. especially pp. 69-75.

it Gottingen, 1905.

'meaning' without its being necessary that the ideas should actually become conscious" (p. 217). A nonsense syllable or a word in an unknown language does not place in readiness any such set of tendencies to reproduction and consequently has no meaning. Every signification and every idea is an associative abstraction because it picks out some of a vast number of possible associations. And no signification is identical with any other but only more or less analogous. 18

The problem of the understanding of words and sentences was taken up by Dr. Clifton 0. Taylor in 1906. 19 His first experiment was based upon a similar one made by Marbe in the "Philosophischen Gesellschaf t " during the winter semester of 1904, and may be considered as a continuation of Marbe 's line of work. He read to his subjects the following sentence : "Imagine that in a rectangular space a plane is laid passing through the upper and lower edges of two opposing sides. The plane then must stretch obliquely through the space. How many such planes can you imagine in this space?" Then followed seven subordinate tasks based upon this fundamental problem.

From the protocol obtained from his subjects it was evident, that for the understanding of sentences expressed in concrete terms the development of mental images can be useful, but that they are not indispensable. These auxiliary images become less frequent the more familiar the subject is with the text.

A second experiment was carried out in which the subject read a text from Gegenbauer's "Anatomic" He had to take care that he understood the text perfectly, and while reading marked the places where he experienced any mental imagery. Visual imagery aided materially in understanding the text. But on writing out the text and then rereading the written copy, the imagery was reduced from fourteen pictures to but one in the third reading. On the other hand, from experiments with is For a criticism of Dr. Ach's view see below, p. 181 ff.

19 ' - Ueber das Verstehen von Worten und Satzen. ' ' Zeitschrift fiir Psychologie, 40, 1906, pp. 225-251.

sentences composed of abstract terms, it seemed that the appearance of images hindered rather than helped the understanding of the text. With Taylor, as with other members of this school, the Beivusstseinslage comes into prominence, and it is found that these attitudes of consciousness are the more frequent as the subject is less familiar with his text.

"We must now go back a few years to an independent line of research. Two years after the appearance of Marbe's experimental study of judgment, .Binet published his brilliant L' Etude experimental de V intelligence.- His experiments were commenced before the appearance of Marbe's work, for we read on page 76 of a series made in November, 1900. Binet may therefore be considered as a real pioneer in this field. He must too have exercised no little influence on later German authors, for he showed how the method of controlled introspection could give very valuable assistance in the study of our higher mental processes. He used in his experiments a variety of subjects, but most of all his two daughters, Marguerite fourteen years old, and Armande thirteen. His first experiment was to give them the task of writing twenty words — any that they might wish. By questioning he then found out in what sense the words had been used — and how they came to be thought of. By classification of the words used in repeated experiments M. Binet found very characteristic differences in the vocabularies of his two children. Their environment having always been the same, this difference, he concluded, was due to their temperaments. Temperament therefore has its influence on our choice of words. From the fact that the words written formed well-defined groups, M. Binet concluded that association alone does not entirely account for our train of thoughts. Association accounts for word after word in any group — but it does not account for the origin of a new group.