THE ASSOCIATIONS OF NORMAL SUBJECTS

by C. G. Jung and Franz Riklin

1 For some time past, attention has been paid in this clinic to the process of association. In order to produce scientifically useful material for this, my director, Professor Bleuler, has compiled a list of 156 stimulus-words and experimented with them on all types of psychosis. In these experiments a very considerable difficulty soon presented itself. There existed no means of precisely and quantitatively separating association in abnormal subjects from that in normal ones. No work had been done giving any facts on the range of normal subjects and formulating the apparently chaotic coincidences of association into rules. In order to fill this gap to some extent and thereby to pave the way for experiments on pathological associations, I decided to collect more material on association in normal people and at the same time to study the principal conditions involved. I carried out this plan with my colleague, Dr. Riklin.

2 The main experimental methods are as follows: Initially we collected associations from a large number of normal people, with the intention, first, of examining the reactions to see whether they are at all subject to any law; and, next, of discovering whether individual patterns occur, i.e., whether any definite reaction-types are to be found. We combined with this a second experiment of a general psychological nature.

3 The mechanism of association is an extraordinarily fleeting and variable psychic process; it is subject to countless psychic events, which cannot be objectively established. Among the psychic factors that exert the main influence on the mechanism of association, attention is of cardinal importance. It is the factor that in the first place directs and modifies the process of association; it is also both the psychic factor that can most easily be subjected to experiment and the delicate affective apparatus that reacts first in abnormal physical and mental conditions and thereby modifies the associative performance.

4 Attention is that infinitely complicated mechanism which by countless threads links the associative process with all other phenomena of the psychic and physical domain in consciousness. If we know the effects of attention on the process of association, then we also know, at least in general, the corresponding effects of every psychic event that attention is capable of affecting.

5 These considerations led us to investigate the effects of attention on the process of association, hoping to clarify as precisely as possible the following questions:

1. What are the laws governing the range of association in normal subjects?

2. What are the direct effects of attention on the association process? In particular, does the valency of the association decrease with the distance from the focus of consciousness?

6 Our experiments have revealed a series of facts that not only encourage us to follow the paths on which we have set out into psychological regions but also, as we believe, fit us to do so.

C. G. JUNG

Part One

I. GENERAL EXPERIMENTAL PROCEDURE

7 The experiments were carried out alternately by the two authors so that each one in turn undertook a series of experiments on the subjects concerned. Altogether thirty-eight people took part: nine educated men, fourteen educated women, seven uneducated men, and eight uneducated women; the age-bracket was 20–50 years. Care was taken to use, as far as possible, normal subjects for the experiment. This, however, led to unexpected difficulties, particularly with the educated subjects, as precisely on this level the concept of normality must be very elastic. Nevertheless we hope we have not deviated too far from the norm in our selection of subjects for experiment. We give the numbers of the subjects in detail and in many cases combine with this a short description of the personality, which will facilitate the understanding of possible anomalies. Naturally the two authors have also carried out the experiment on each other.

8 In noting associations we have entirely limited ourselves to those produced by calling out stimulus-words. We used altogether four hundred different stimulus-words. These, grammatically classified, are as follows:

nouns 231
adjectives 69
verbs 82
adverbs and numerals 18

9 The number of syllables was not taken into account (the stimulus-words have one, two, or three syllables). Nor were the stimulus-words arranged in definite categories as Sommer, for instance, has arranged them. On the contrary, as much care as possible was taken to see that stimulus-words of similar forms or meaning should not follow each other, so as to avoid the subject adapting to a particular topic after one or two reactions. Through an unfortunate coincidence it happened that among the first hundred stimulus-words there were about thirty that can easily be associated according to temporal or spatial co-existence; in the second hundred there are only about twenty of these, which caused a notable difference of the co-existence association in the first and second hundred. The shortage of stimulus-words of this kind is made up by verbs. It was considered important completely to exclude difficult and rare words, in order to prevent mistakes or lengthened reaction-time due to lack of knowledge on the part of the subjects. The stimulus-words were therefore taken as far as possible from everyday life.

10 This consideration was all the more essential for us, as with most of our subjects we had to work under somewhat abnormal linguistic conditions. In German-speaking Switzerland the vernacular consists, as is well known, of the Swiss-German dialect or dialects, which not only deviate considerably from standard German but also show significant phonetic differences among themselves. In the schools children learn standard German as if it were a foreign language. In later life educated people gain a fairly complete knowledge of and facility in the German language. The uneducated man, however, unless he has spent a considerable time in Germany, retains at best those German phrases that he has learned at school and later learns little or no more. Nevertheless, literary German is familiar to him in printed or written form and he also understands it as a spoken language without being able to speak fluent, correct standard German himself. We tried therefore in many cases to call out the stimulus-words in the dialect form, but we soon noticed that the uneducated subjects did not understand dialect words as well as standard German. They reacted to the dialect words more laboriously and tried to react in standard German. This somewhat paradoxical phenomenon can be explained by the fact that Swiss-German is a purely acoustic-motor language, which is very rarely read or written.

11 Everything printed or written is in standard German. The Swiss is therefore not used to experiencing words individually but knows them only in acoustic-motor connection with others. If he has to say a single word without an article he will usually choose the standard German form. We therefore avoided dialect words completely in our experiments. In most cases a correct standard German reaction was given, but any reactions that were in dialect were fully accepted. The reactions were, of course, written down in the form in which they were given. To subjects who had never taken part in such experiments, their significance was first explained and practical examples of how they had to react were demonstrated to them. Not a few of the uneducated subjects thought that it was a kind of question-and-answer game, the point of which was to find an appropriate word connection to stimulus-words, e.g., house / housecat, wild / wild cat. The experiments were never started till it was certain that the subjects understood the experiment. We stress that a case of not understanding never occurred and that general lack of intelligence was much less disturbing than affects, particularly a fairly frequent emotional obtuseness. It is of some significance that many of the uneducated came with a certain "schoolroom" attitude and a certain correct and stiff demeanour.

12 We organized our experiments as follows: The first two hundred reactions were noted without further conditions. The reaction-time was measured with a 1/5-second stop-watch, which we started on the accented syllable of the stimulus-word and stopped on the uttering of the reaction. We do not, of course, presume to have in any way measured complicated psychological times by this simple procedure. We were merely concerned with establishing a general idea of a roughly average reaction-time which is in many cases not without importance, being very often of value in the classification of reactions.

13 After two hundred reactions, these were as far as possible classified, with the help of the subjects. With educated subjects this was always done; with uneducated subjects, who only rarely have any capacity for introspection, it was of course impossible. We had to limit ourselves to having the connection explained in particularly striking associations. The results of the experiment were divided into a first and second hundred and these were written down separately. During the experiment the psychic state of the subject was as far as possible established, both objectively and subjectively. If for any reason physiological fatigue occurred, we waited till the next day before doing the second experimental series. With the educated subjects fatigue almost never occurred during the first experiment, so that we could continue at once with the second series in nearly every case,

14 The second series of experiments consisted of one hundred reactions which were recorded under the condition of internal distraction. The subject was asked to concentrate his attention as much as possible on the so-called "?-phenomenon" (Cordes) and at the same time to react as quickly as possible, i.e., with the same promptness as in the first experiment. By the "A-phenomenon" we understand, with Cordes, the sum of those psychological phenomena that are directly stimulated by the perception of acoustic stimulus. To establish whether the subject had observed the A-phenomenon he had occasionally to describe it after the reaction, and this was noted. On completion of this experiment new classifications were again made. Of course, for this experiment only educated people could be used and of these unfortunately only a selection, because it takes a certain psychological training to be able to observe attentively one's own psychic phenomena.

15 The third experimental series was sometimes not carried out till the second day. It consisted of one hundred reactions and was based on the condition of external distraction. The distraction in this experiment was brought about in the following way: The subject had to make pencil marks of about one centimetre, in time with a metronome. The beat for the first fifty reactions was 60 per minute and for the second fifty reactions 100 per minute. The classification results of the first fifty reactions and the second fifty were recorded separately and for ease of calculation brought to one hundred. With a very few subjects the metronome was speeded up at every twenty-fifth reaction in order to exclude an all too quick habituation. The beat was in these cases increased from 60 to 72 and from 100 to 108 per minute.

16 The factor of habituation, in any case, unfortunately plays a large part in these experiments, as one would expect. Many people very quickly get used to a purely mechanical activity in which, in the second phase of the experiment, only the beat changes. It is difficult to introduce other disturbing stimuli of equal continuity and variability without adducing word-images, particularly when one does not wish to make too great demands on the intelligence and will-power of uneducated subjects.

17 In trying to find a suitable disturbing stimulus we were above all intent on excluding that which might have had an excitatory effect on verbal imagery. We think we did exclude such effects by our experimental procedure.

18 From these experiments three hundred to four hundred associations, on an average, were obtained from every subject. We also tried to supplement our material in other directions, in order to obtain a certain connection with Aschaffenburg's results, and for this purpose we took associations from some of our subjects in a condition of obvious fatigue. We were able to obtain such reactions from six subjects. Associations were also taken from one subject in a state of morning sleepiness after a night of undisturbed sleep, in which the factor of fatigue was completely excluded. With one subject associations were taken when he was in a state of acute moodiness (irritability) without fatigue.

19 In this way we obtained about 12,400 associations.

II. CLASSIFICATION

I. General

20 Anyone with practical experience of work on association has been confronted with the difficult and unrewarding task of classifying the results of the experiments. On the whole we agree with Cordes when he says that in earlier association experiments the false assumption prevailed that the fundamental psychological phenomenon corresponds to the stimulus-word and that the connection between stimulus-word and reaction is an "association." This somewhat too simple interpretation is at the same time too pretentious, for it maintains that in the connection between the two linguistic signs there is also a psychological connection (the association). We do not, of course, share this point of view but see in the stimulus-word merely the stimulus in the strict sense of the word and in the reaction merely a symptom of psychological processes, the nature of which we cannot judge. We do not, therefore, claim that the reactions we describe are associations in the strictest sense; we even wonder if it would not be altogether better to drop the word "association" and talk instead of linguistic reaction, for the external connection between stimulus-word and reaction is far too crude to give an absolutely exact picture of those extraordinarily complicated processes, the associations proper. Reactions represent the psychological connection only in a remote and imperfect way. Thus, when describing and classifying linguistically expressed connections, we are not then classifying the actual associations but merely their objective symptoms, from which psychological connections can be reconstructed only with caution. Only in psychologically educated subjects is the reaction what it really should be—namely, the reproduction of the next idea; in all others a distinct tendency to construct something is mixed with the reaction so that in many cases it is the product of deliberation, a whole series of associations. In our association experiments we stimulate the language apparatus. The more one-sided this stimulus is, the greater the number of linguistic connections that will appear in the reaction. As we shall see, this is mainly the case with educated subjects, from whom a finer differentiation of psychological mechanisms, and therefore a greater ability for isolated application, can a priori be expected. One must therefore guard against the fallacious assumption that the educated subject has in any way more external associations of ideas than the uneducated. The difference will be a psychological one, as in uneducated subjects other psychological factors insinuate themselves. In the second part of this paper we shall refer to this difference.