v. H.Pf.Schu.St.Questioner.IIIIIIIIIIIIv. Osten9153417-—827Schillings-—1917616-—Pfungst613——-—9—
We have omitted from this table several tests in which the observer of the questioner noticed no head jerks whatever, and therefore could not arrest his stop-watch, although the horse responded correctly. Four tests of this kind were made by Mr. von Hornbostel, two by Mr. Pfungst, two by Mr. Schumann and five by Mr. Stumpf. In the case of Mr. Pfungst the horse gave the unusually high number of fifty taps. The attention of the observer had been taxed too long and had failed him (two seconds is the most favorable time). The head-jerk of Mr. von Osten evidently occurred during a lapse in Mr. Pfungst's attention and therefore remained unnoticed.
v. H.Pf.Schu.St.Questioner.IIIIIIIIIIIIR.44%60%62%88%———0%48%V. Osten.W.56%20%12%0%———100%22%R.100%92%——————100%—Pfungst.W.0%0%——————0%—R.———74%100%83%100%———Schillings.W.———5%0%17%0%———
The results of the experiments are given in the second table. The general arrangement corresponds to that of the first table. Even though the absolute number of tests was small, yet for the sake of giving a better general view, all values are given in percentages. The tests in which the movement of the questioner had preceded that of the horse—as had been anticipated—are recorded under "R" (right); under "W" (wrong), we have recorded those cases in which the testimony of the stop-watches—contrary to our expectation—indicated that the reverse order prevailed. Finally, those cases which would complete the 100%, i. e. those in which the watches indicate simultaneity of the movements in question, are not recorded.
From this table we may note the following: The time-measurements for Mr. Schillings and Mr. Pfungst are quite in agreement and go to show that the order in time of the head movement of the questioner and the back-step of the horse was exactly what had been expected. The few contradictory cases which occur in Series I of the observations upon Mr. Schillings are to be accounted for by the fact that he was here for the first time the subject of observation, whereas the recorded time-measurements in the case of Mr. Pfungst had been preceded by a number of practice tests. The results of the measurements taken in the case of Mr. von Osten were far less satisfactory. Even if one were to allow a series containing barely more than 50% of "right" cases as sufficient proof of the correctness of our expectation regarding the order of the movements of the questioner and the horse, only three of the six series obtained with Mr. von Osten as subject, would satisfy this expectation. However, since four of the six series show a greater number of cases of simultaneity (their percentage may be easily deduced by referring to the per cent of "right" and "wrong" cases), the proposed method would give a distorted view, and therefore it appears that the more correct method would be to consider simply the numerical ratio of the "right" and "wrong" cases. Since, furthermore, Series II shows, in every case, a decided change which is similar for all observers (note especially Pfungst), there can be no doubt but that practice is here involved, and that Series II is to be regarded as the true standard. Throughout this series we find a preponderance of "right" cases. Therefore, the table unmistakably confirms the expected order in time. That there were more "wrong" cases with Mr. von Osten as subject than with the other questioners is to be explained by the fact that the decisive movements were far less easily observed in this case, than in that of the other questioners. (Seepage 49.) We expect that Series III would show thesame results, or approximately the same results in the case of Mr. von Osten that it did for Mr. Pfungst and Mr. Schillings, but unfortunately he declined to act as subject. In the meantime, however, new and decisive proof presented itself which destroyed all possible doubt.
Before adverting to it, let us consider in a few words the reaction-time of the horse,—the time elapsing between the final sign of the questioner and the reaction of the horse (i. e., the back-step). Unfortunately this time cannot be directly determined. All that can be ascertained from our time-measurements, is the time intervening between the moment of the head-jerk and the moment in which the reaction of the horse is noted. (Seepage 51). This time averaged, for the 127 measurements, .45 seconds. If we stated the unavoidable error, (obtained on the basis of extended supplementary measurements which it is not necessary to consider here) as .15 seconds, and apply it to the value found above, we obtain .3 seconds as the probable reaction-time of the horse.[K]
That the tapping—as well as all other movements of the horse—was nothing other than a reaction upon certain visual stimuli, was proved beyond a doubt by the fact that the voluntary execution of the head-jerk and of other movements—which we will describe in more detail later on,—brought about all the proper responses on the part of the horse. Thus, artificial synthesis became the test of the correctness of analytical observation.
To elucidate; if the questioner retained the erect position he elicited no response from the horse, say what hewould. If, however, he stooped over slightly, Hans would immediately begin to tap, whether or not he had been asked a question. It seems almost ridiculous that this should never have been noticed before, but it is easily understood, for as soon as the questioner gave the problem he bent forward—be it ever so slightly—in order to observe the horse's foot the more closely, for the foot was the horse's organ of speech. Hans would invariably begin to tap when I stooped to jot down some note I wished to make. Even to lower the head a little was sufficient to elicit a response, even though the body itself might remain completely erect. Of thirty tests made in this position, twenty-nine were successful. Hans would continue to tap until the questioner again resumed a completely erect posture. If, for instance, I stooped forward after having told the horse to tap 13, and if I purposely remained in this position until I had counted 20, he would, without any hesitation, tap 20. If I asked him to add 3 and 4, but did not move until 14 was reached, he would tap 14. Twenty-six such tests gave similar results.
The reaction of the horse upon such a signal for stopping showed slight modifications according to thetime which elapsed between the last tap and the signal for stopping. These modifications, which had hitherto been paraded as expressions of the horse's psychical power may be illustrated by the following schematic figures (Figures1-4). In all of them the dotted linec-drepresents the ground level;dshows where the horse's right forefoot was located before he began tapping;aandc, respectively, indicate the place to which the foot is lowered during the process of tapping. The unbroken line gives the direction of the back-step.
If Hans, having raised his foot fromatob—preparatory to tapping,—receives the signal at or just before the moment he lowers the foot, he immediately swings it in a wide circle fromcback to its original position atd, (Fig. 1). As a matter of factaandccoincide, but are juxtaposed in the diagram for the sake of schematicutility. This was the usual form of the back-step.
Fig. 1.
Fig. 1.
If the signal for stopping is given a little after the last tap (Fig. 2), i. e., at the time that the foot is already being raised for another tap, then the back-step occurs asa-b-d. The horse thus gives, at the moment it receives the signal for stopping, a changed impulse to the movingfoot. The curve, therefore, has a kink atb, and the back-step occurs with seeming hesitancy,—Hans appears not quite certain of his result.
Fig. 2.
Fig. 2.
If the signal be given somewhat later still (Fig. 3), i. e., when the foot is being lowered to complete a tap, Hans is still able to put on the brakes—as it were—and draw back his foot before it reaches the ground. The whole process gives the impression that the horse was just about to make a "mistake" of one unit, but at the last moment had bethought himself of the correct answer.
Fig. 3.
Fig. 3.
Finally, if the signal be deferred still longer, it becomes impossible to prevent the extra tap. The back-step again has the same form as infigure 1; Hans has made a "mistake" in his answer by one unit too many.
Conversely, if the head-jerk of the questioner occurs too soon; i. e., at the moment the horse has raised his foot for the final tap to the heightb, (Fig. 4), then the tap is not completed,—but the foot, without touching the ground, makes the curveb c2d, back to its original position. Hans has again made a "mistake" in his answer,—this time by one unit too few.
Fig. 4.
Fig. 4.
All these variations go to show one thing: Hans neverknows in advance which tap is to be the final one. These variations in his reactions occurred often without having been intended by the questioner. But to bring them about at will required skill, on account of the shortness of the time involved in the reaction.
Whenever the signal for stopping—which we have just discussed—was followed by the complete erection of the head and trunk, Hans would definitely cease tapping. If, however, the questioner failed to assume a completely erect position, or if he stooped forward ever so slightly, the horse would follow the back-step of the right foot with an extra tap of the left foot. Besides occurring in tests in which Mr. von Osten assumed the rôle of questioner, this fact was also noted when the Count zu Castell and Mr. Schillings acted as subjects. Since the extra tap just mentioned was not given like the others with the right foot forward, but with the left foot upon the spot, it was possible for the horse to execute it with a greater show of energy. This simulated a high degree of mental certainty on the part of the horse, as if he wished to indicate that this was the correct solution of the problem and it would have to stand. In spite of all this, manyerrors would creep in. It was possible to prolong this extra tap and thus make it appear more dilatory. We need hardly add that henceforth it was within the power of the experimenter to have the tapping executed entirely with the right foot or with the final extra tap of the left foot. Hitherto the view had been current that this lay solely within the pleasure of the horse.
If the questioner still inclined forward, still remained in the bent posture after Hans had given the final tap with his left foot, the horse would immediately begin to tap once more with his right foot, which had, in the meantime, become ready for further action. If the head jerk was then made, Hans would bring his right foot back, give the extra tap with his left foot, then resume tapping with the right and thus continue until the questioner once more resumed the erect posture. Thus the horse on one occasion when I wished him to tap 100, gave—contrary to my desire—the following response; 39 with the right foot, 1 with the left, 24 with the right, 1 with the left, 35 with the right, and 1 with the left. Later it became possible for me to cause him to tap 1 right, 1 left, 1 right, 1 left, etc. I could even get him to tap exclusively with the left foot by standing at his left rather than at his right as had been customary with his questioners. These taps with the left foot were executed in a far less elegant fashion than those with the right foot, and with a great waste of energy. Hans had become a right-handed individual—as it were—as a result of long habit.
With regard to the distance at which the experimenter directed the horse, the following may be said: The usual distance was one-quarter to one-half meter. This holds for all tests hitherto described. Seventy tests which were made for the purpose of discovering the influence ofchange in distance showed that the reaction of the horse upon the customary signal of the head-jerk was accurate up to a distance of three and one-half meters. At a distance of three and one-half to four meters there suddenly occurred a fall of 60-70% in the number of correct responses. At a distance of four to four and one-half meters only one-third of the responses were correct, and at a distance beyond four and one-half meters there were no correct responses. The greater number of these tests were made in our presence by Mr. von Osten, who was under the impression that we were testing the accuracy of the horse's hearing, whereas we were really testing the accuracy of his perception of movements.
With regard to the different positions which the experimenter might assume with reference to the horse, the following may be noted: The normal position was to the right of the horse. If the experimenter stood immediately in front of Hans, the latter's reaction would be just as accurate, though he would always turn his head and make desperate efforts to see the questioner, even though he was held in short by the reins. When a position immediately behind the horse was taken—a somewhat dangerous proceeding, since Hans would at once begin to kick—no response could be obtained until he succeeded in turning far enough around to get the questioner within view. If he was restrained from turning completely around, he would at least turn his head,—and always to the right. One might even turn his back upon Hans during the tests, for the signal for stopping was not obtained from the face of the questioner, but from a movement of the head. The following incident will show to what extent the horse had become accustomed to seeing the questioner in a certain definite position. For a longtime I had been in the habit—without exception—of standing close to the horse's shoulder. Mr. von Osten, on the other hand, would stand farther back. When, on a certain day, I assumed the latter position, the horse would not suffer it, but would move backward until he had his accustomed view of me.
Finally we sought to discover by what movements the horse could be made to cease tapping. We discovered that upward movements served as signals for stopping. The raising of the head was the most effective, though the raising of the eyebrows, or the dilation of the nostrils—as in a sneer—seemed also to be efficacious. However, it was impossible for me to discover whether or not these latter movements were accompanied by some slight, involuntary upward movement of the head. The upward movement of the head was ineffective only when it did not occur as a jerk, but was executed in a circuitous form,—first upward and then back again. Such a movement was occasionally observed in the case of Mr. von Osten. The elevation of the arms or of the elbow nearest the horse, or the elevation of the entire body was also effective. Even if a placard, with which the experimenter tried to cover his face, were raised at a given moment, the horse would make the back-step. On the other hand, head movements to the right and to the left or forward and back, in fine, all horizontal movements, remained ineffective. We also found that all hand movements, including the "wonderfully effective thrust of the hand into the pocket filled with carrots", brought no response. I might also change my position and walk forward and then backward some distance behind the horse, but the back-step would only occur in response to the characteristic stimulus. After what has been said it is easy to understand how vain were Mr.Schillings' attempts to disturb the horse and how naturally he might conclude that Hans was not influenced by visual signs. Mr. Schillings simply did not know which signs were effective.
While the horse could thus be interrupted in the process of tapping by movements which were executed at the level of the questioner's head, yet movements below this level had the opposite effect. If Hans showed that he was about to cease tapping before it was desired, it was possible to cause him to continue by simply bending forward a trifle more. The greater angle at which the questioner's trunk was now inclined caused the horse to increase the rate of tapping. The rule may be stated thus: The greater the angle at which the body inclined forward, the greater the horse's rate of tapping, andvice versa. It was noticeable that whenever Mr. von Osten asked for a relatively large number—in which case he always bent farther forward than in the case of smaller numbers—Hans would immediately begin to tap very swiftly. Not being entirely satisfied with these observations, the following more exact measurements were taken. I asked the horse to tap 20. From 1 to 10 I held my body at a certain constant angle, at 10 I suddenly bent farther forward and retained this posture until 20 had been reached. If there existed a relationship between the angle of inclination and the rate of tapping, then the time for the last ten taps ought to be less than for the first ten. Of 34 such tests 31 were sucessful. The following are two specimen series.
The first series consisted of ten tests of 15 taps each. In all cases my head was bent at an angle of 30° to the axis of the trunk, but I constantly changed the angle of inclination of the trunk. It was not possible to measurethis angle accurately on account of the rapidity with which the whole test had to be made. I was able, however, to differentiate between them with enough accuracy to designate the smallest angle (about 20°) as belonging to Grade I, and the greatest angle (about 100°) as belonging to Grade VII. By fixing certain points in the environment, it was possible to get approximately the same angle repeatedly. The time from the third to the thirteenth tap was, in all cases, taken by Prof. Stumpf by means of a stop-watch. The tests were taken in the following order:
Grade of inclination:IVIIIIIIVVVIVIITime for 10 taps:5.24.65.05.04.84.84.64.4sec
From this series it will be seen that in the case of the same angle of inclination (II and VI were repeated and III was omitted) the same rate obtained in the tapping. In two other tests I constantly increased the angle of inclination during the 15 taps, and Hans gradually increased the rate of tapping accordingly.
In a second series I had the horse tap 14, five times. I myself took the time of the taps up to 7 by means of the stop-watch, while Prof. Stumpf took the time of the taps from 8 to 13. At 8 I suddenly bent forward a little more and retained this position until tap 13. The results were as follows:
Taps2 to 7(Pf.):3.22.2-2.42.42.2-2.42.4seconds"8 to 13(St.):2.62.02.02.22.2seconds
Such good results, however, were possible only after a number of preliminary practice tests had been made. The experiment was especially difficult because the horse was often on the point of stopping in the midst of a test. This was probably due to some unintentional movementon my part. In such cases I could induce him to continue tapping only by bending forward still more, but this effected also, as we have seen, an increase in his rate of tapping. Such tests, of course, could not give unambiguous results.
The rate of tapping was quite independent of my rate of counting. Thus, if I counted aloud rapidly, but bent forward only very slightly, the horse's tapping was slow and lagged behind my count. If I counted slowly but bent far forward, Hans would tap rapidly and advance beyond my count. Thus we see that his rate of tapping was in accordance with the degree of inclination of my body and never in accordance with the rate of my counting, i. e., it was quite independent of every sort of auditory stimulation.
Direct observation and a comparison of the records of the time Hans required in giving to his master responses involving small, medium and large numbers, with the records of the time which he required to respond to my questions when I bent only slightly, moderately or very far forward, proved that the increased rapidity in tapping in the case of large numbers, which many regarded as an evidence of high intelligence, (seepage 20), was, as a matter of fact, brought about in the way described. The two series (in each of which the time measured was for 10 taps) are quite in accord. The horse did not tap faster because he had been given a large number by Mr. von Osten, but because the latter had bent farther forward.
From all this it readily appears why it was possible to cause Hans to increase his rate of tapping but not to decrease it. To do the latter would involve a decrease in the angle of inclination of the body. This would necessitate the erection of the body. As we have seen, this was the signal to which Hans reacted by ceasing to tap. And as a matter of fact we never knew the horse to decrease his rate of tapping in the course of any single test, except in the case of very large numbers, and then it was probably due to fatigue. Mr. von Osten insisted that Hans often slowed down toward the end of a test, "in order to obviate mistakes", but all the tests in which he tried to demonstrate this to us, were unsuccessful. In spite of all exhortation, Hans would tap either uniformly or somewhat more rapidly as soon as his master—in all probability unconsciously—bent somewhat lower. Only once was such a test successful. Mr. von Osten—upon our request—asked the horse to give a certain large number. In this instance the decrease in the rate of tapping was due to fatigue and had nothing whatever to do with the desire on the part of the horse to avoid error.Furthermore, Mr. Hahn, who had visited Hans twenty times and had made careful notes of his observations, corroborated my statement when he said that he himself never noted the decrease in rate mentioned. Contrary statements may perhaps be due to the fact that the tense state of expectancy on the part of the observer made the interval between the last taps appear subjectively somewhat longer.
So much for the technique of the tapping. Now a word about the numbers which Hans tapped. (I refer only to the results obtained in series which involved no volitional control). The number 1 was very difficult to get. Hans usually tapped 2 instead. Thus even in the case of Mr. von Osten he responded five times with 2, and only in the sixth test did he react correctly. As far as other questioners were concerned, 1 was seldom everobtained, except in the case of Mr. Schillings and myself. The numbers 2, 3 and 4, on the other hand, were very easily obtained and, above all, 3 seldom failed. 3 seemed to be the horse's favorite number and was very frequently given instead of other numbers. Thus, one-sixth of all the horse's incorrect responses which were given to me were in terms of the number 3. The numbers 5 and 6 were a little more difficult to obtain and above 10 the difficulty increased rapidly. Indeed, I never saw Hans respond with a number exceeding 20 to any questioner, Mr. Schillings and Mr. von Osten excepted. I saw the nine vain attempts of Count zu Castell to get the number 15, and Count Matuschka's eight unsuccessful attempts to obtain the number 16 as a response. But even with Mr. von Osten and Mr. Schillings such failures were not infrequent. Thus, Mr. von Osten tried five consecutive times to obtain the number 24. I myself did not fare any better at first. But the following table shows what practice can do. If we compare the percentage of correct responses (involving the numbers 1 to 7—for which alone I have sufficient material, viz., 80 to 100 cases), obtained in the first half of our tests, with that of the second half, we get the following:
For Number:1234567In first half of tests:49,92,89,86,74,62,53%" second " " :92,95,92,98,97,86,96%
From this we see how hard it was at first to get the number 1 and that failure was as frequent as success, and how much easier it was on the other hand to get the numbers 2 and 3 (and which, therefore, do not show any great improvement in the second half of the tests). Beyond the 3 the percentage of correct responses decreased and the number 7 stood at the same level as thenumber 1. In the second half of the tests, all these differences disappeared and errors were infrequent and seldom exceeded +1 or -1. These results of practice are not to be accredited to the horse, but to the experimenter, who was at first quite unskilled. This difference in results does not appear in the case of Mr. von Osten, for his initial practice had been had many years previous. The values obtained in his case were very constant throughout our experimentation and generally showed something like 90% of correct responses. To be sure, in his case also, the number 1 was somewhat unfavorable, (79% were correct responses). But the percentages obtained in his case showed no improvement whatever throughout our experimentation. We need scarcely add that with the voluntary control of the giving of the signs, in the case at least of such small numbers as are here discussed, no errors, whatever, occurred.
We have discussed the influence of the experimenter, i. e., the one who asked the horse to tap; now let us consider the influence of others present upon the horse.
As a general rule, other persons had no effect upon the horse's responses. This appears from the failure of nearly all tests in which all of those present—with the exception of the questioner himself—knew the number which the horse was to tap. Even when the others concentrated their whole attention upon the number, it profited little as a close analysis of the 136 cases, which belong under this head in our records, go to prove. Thus, in the presence of a group of twenty interested persons—during the absence of Mr. von Osten—twenty-one problems were given to the horse, the solutions of which were known to everyone but myself, the questioner. Result: only two correct responses. Only when there was among thespectators someone to whom the horse was accustomed to respond or one from whom he regularly received his food, would such an influence be effective.[L]But such caseswere few. The most important were the following: I at one time whispered a number to Hans (on the occasion of the tests mentioned onpage 37), and Mr. von Osten asked for it the moment I stepped aside. Hans answered incorrectly even though I stood close beside Mr. von Osten; I did not, however, think intently of the number. As soon as I concentrated my attention upon the number he promptly responded correctly. Further cases are those mentioned onpage 38, in which the keeper of the horse unintentionally aided in giving four dates which were unknown to all others present, including the questioner. This single instance shows the necessity of the rule that during tests in which the method is that of procedure without knowledge the solutions should be known to no one of those present. Finally the tests made by the September-Commission and reported inSupplement III (page 255)may possibly belong under this head. Since they were not followed out any further, I am unable to render a definite judgment upon them. In most of these tests the question itself, as put by Mr. von Osten, was not adequately answered, but curiously enough, however, the number which had been given to Hans in von Osten's absence and which formed the initial number of some mathematical operation, was tapped correctly. This may possibly be explained by the assumption that this initial number had been retained in the memory of some of those present, (seepage 149, on the "perseverative tendency"), and that the horse, since he had been working with some of them, responded to one of those present. Chance may have played some part also.
If the questioner knew the number of taps desired, (which was not the case with the tests hitherto discussed), then the environment had still less influence upon thehorse—except that it caused occasional interruption. The horse's responses, therefore, did not tend to become more successful just because a number of persons were simultaneously concentrating upon the result desired. This was proven by the experiments which we repeatedly made for this purpose. Only one person at a time had any influence upon Hans. If two questioners tried to influence the horse at the same time,—other conditions being the same,—success would be for the one who had the greater control over the animal when working alone with him. Prof. Stumpf and I made the following experiment. Both of us stood to the right of the horse, each thinking of a number. In ten such tests Hans always tapped my number. When Stumpf concentrated upon 5 and I upon 8, the horse responded with 8, i. e., the larger number. When Stumpf had 7 in mind, and I had 4, the response would be 4, i. e., the smaller number. When Stumpf thought of number 6, and I had fixed upon none, Hans tapped 35. He was evidently awaiting my signal. When I went away Stumpf again demanded the number 6, and the horse responded properly. When I returned, Stumpf's attempts again failed. On another occasion Count Matuschka put a number of questions, while Mr. von Osten stood behind him. All of the horse's responses were correct, even the one answering the question: "How much is 7 times 7?", which was difficult on account of the great number of taps required. I was able to note from the direction of the horse's eyes that he was attending only to his master and not to the Count. On still another occasion Mr. Grabow sang two tones—the second being the fourth of the first—and asked Hans: "How many intervals lie between?" I was standing erect before the horse, and was thinking intently of thenumber 2, but without giving any voluntary sign of any sort. Hans tapped 2, whereupon Mr. Grabow put a number of similar questions; but I no longer thought of the answers, and all of Hans's responses went wrong.
Although Hans was not influenced by others so long as a suitable experimenter was present, yet he might be disturbed and under certain conditions might be led to make the back-step in response to certain movements in his environment. The person to whom he responded would have to be close to the experimenter and would necessarily have to execute a movement greater in extent than the experimenter's. In such instances the raising of the head, arm or trunk, was a sufficient stimulus. Thus we made the following two series of tests. Mr. Stumpf stood with trunk bent forward before the horse, and at a moment decided upon beforehand, assumed an erect position. I myself stood beside Hans and asked him to tap. When I stood at the horse's neck, then Mr. Stumpfs interruption was effective. When I stood at the horse's flank, the interruption effected only a seeming hesitation, and when I moved still farther back, the horse continued to tap despite any attempted disturbance. In the second series the questioner remained constantly at the right shoulder of the horse, while the one who attempted to distract him, changed positions. When the latter stood to the right immediately in front of or beside the questioner, the disturbance was effective in 10 out of 13 cases. But when he stood back of, and to the right of, the questioner, the attempts at disturbance were seldom successful. If he chose a place before and to the left of the horse, there was hardly any distraction (in 4 cases only, out of 13), and if he stood to the left and behind the animal, he exerted no influence whatever. Hans manifestly turnedhis attention, almost exclusively, to the side at which the questioner stood.
That knowledge of thismodus operandimade it possible for those persons to get responses from the horse, who hitherto had been unsuccessful, is shown in the case of Mr. Stumpf when he began to control his movements voluntarily on the basis of observations which had been made.
We are here concerned with the horse's head movements upward, downward, to the right and to the left, and also with nodding and shaking of the head to signify "yes" and "no". We soon discovered that these experiments, also, were successful without an oral statement of the problem,—in other words, the auditory stimulus was quite superfluous. The tests with the blinders showed that Hans was lost as soon as his questioner was out of his view, but responded adequately the moment the questioner was in sight. Hans, therefore, had established no idea of any sort in connection with the terms "up", "down", etc., but in these cases, likewise, he reacted in response to certain visual stimuli. The nature of these stimuli I discovered at first in my observations of Mr. von Osten and also of myself, when working with the horse.
Above all things it was necessary that the questioner, during these tests, should stand perfectly erect. If he stooped ever so slightly, the test was unsuccessful. If he carefully refrained from any movement whatsoever, and looking straight before him asked the horse, "Whichdirection is right?" or "Which way is upward?", Hans would execute all sorts of head movements without rhyme or reason. It was evident that he noted that a head movement of some kind was expected of him, but did not know the particular one that was wanted. But if the questioner now raised his head, Hans would begin to nod and would continue doing so until the questioner lowered his head. This reaction was interpreted as signifying "yes". Mr. von Osten had always asked Hans before each of the more difficult tests whether he had comprehended the meaning of the problem, and was reassured only upon seeing the horse's affirmative response. But contrary to Mr. von Osten's expectation, Hans also responded in this manner after a pair of ear-caps had been drawn over his ears. In the case of the tests described at the beginning of the chapter, in which the method was that of "procedure without knowledge", Mr. von Osten had always insisted that we await Hans's nod of comprehension before proceeding. We complied; Hans nodded and—regularly disgraced himself!
When the questioner raised his head somewhat higher than normal, Hans would throw his own upward, which was supposed to signify "upward". A lowering of the head on the part of the questioner was followed by a lowering on the part of Hans, which was his form of response for "down". For some time I was in a quandary as to the difference between the questioner's signal for this latter response and the one which was the signal for the horse to begin tapping, although I had often given both kinds unwittingly. Further experiments showed that Hans responded with a nod of the head whenever the questioner, while bending forward, chanced to stand in front of, or to the side of the horse's head, but thathe would begin to tap in response to the same signal, as soon as the experimenter stood farther back. The difference in the two signals, therefore, was very slight, and I repeatedly noted that instead of tapping, as he had been requested, Hans would respond to the Count zu Castell's and Mr. Schillings' questions by a nod of the head.
If, while standing in the customary position to the right of and facing the horse, the questioner would turn his head a little to the right—a movement which, when seen from the horse's position, would appear to be to the left,—Hans would turn his head to his left. But if on the other hand the questioner would turn slightly to the left,—i. e. seen from the horse's position, to the right,—then Hans would turn his head to his right. And finally, whenever the questioner turned his head first to the right, then to the left, Hans would respond by turning first to his left, then to his right. This, according to Mr. von Osten, signified "zero" or "no". Since this movement could not be executed by the experimenter while in a stooping position, it can now readily be seen why it was that Hans, instead of shaking his head, always began to tap whenever a placard with "O" upon it, was shown to him in the course of the experiments in which the method was procedure without knowledge on the part of the questioner. The latter expected the horse to tap, and therefore bent forward. Like all of the horse's other forms of response, this, too, was always unsuccessful whenever the questioner stepped behind the animal. Although Hans had always responded to Mr. von Osten and Mr. Schillings, and at first also to me, by means of the stereotyped movement of the head to the right and then to the left to signify "zero" or "no", I later succeededin controlling my signals so as to get the inverted order in the horse's response. In the case of Mr. Schillings and of Mr. von Osten all of the movements just described were very minute, and long after the movements, which were effective stimuli for releasing the process of tapping, were recognized, it was still exceedingly difficult to discover them in these two gentlemen. The signal for "zero" and "no" was relatively the most pronounced of the group in the case of Mr. von Osten, while with Mr. Schillings it was the least pronounced, in comparison with his very strong "jerk". Yet in both cases Hans respondedwith absolute certainty.
It is now readily conceivable how it was possible to make the horse respond to all sorts of foolish questions, both by involuntary signs—i. e., expressions following upon the bare imaging of the response expected,—as well as by means of controlled signs. One could thus obtain consecutively the answers "yes" and "no" to the same question. Or one might ask: "Hans, where is your head?", and Hans would bend to the earth. "And where are your legs?" He would look at the skies. Etc.
Let us examine for a moment the directives which the horse required for the various positions. If one called him, while he was running about the courtyard, he paid no attention whatever, but if one beckoned to him, he came immediately. A raising of the hand brought him to a standstill. If one now stepped forward or pointed one's hand in that direction, he would step forward, orvice versa, he would step backward. By means of minimal movements of the head, of the arm nearest the horse, or of the whole body, Hans could be induced to assume the position one desired, without touching him or speaking a word. I noticed this quite early in the courseof the investigation. Once, when intending to ask the horse to step backward to the right, I inadvertently said "Step backward to the left!", whereupon he stepped backward to the right. In spite of my verbal error, I had involuntarily given him the proper directives.
Finally we may note that Mr. von Osten had occasionally asked the horse to jump or to rear. The command in this case was: "Jump", or the question was: "What do the horses do in the circus?". Since these tests were just as effective when the command was given silently, it was an indication that these, too, depended upon visual stimuli. What was necessary to cause the horse to step backward and then jump forward was to step backward oneself, or make a slight movement of the hand in that direction. If one wished to make him rear, it might be effected by throwing the arm or head slightly upward.
The method pursued in these tests was the following: From five to eight pieces of colored cloth ½ × ¼ meters in size were arranged in changing series upon the ground, the interval between them being equal to the width of one piece, or else they were hung upon a string a man's height above the ground. This method was also employed when placards of like size with written symbols were used. The horse stood ten paces away and opposite the middle of the series, while Mr. von Osten stood at his right. Hans was asked to go and point out the cloth of a certain color or the placard with a certain word upon it. If the cloth lay upon the ground, Hans picked it up withhis mouth and carried it to the questioner. If the cloth, like the placards, hung from the cord, he approached, pointed it out with his nose and then backed up to his original position. Before approaching the objects, Hans was required to indicate, by tapping, the number of the place in the series (counting from left to right), which the cloth or placard occupied. Mr. von Osten never omitted this requirement. Then the command "Go!" was given, and Hans obeyed. (As a matter of fact, a slight directive movement of the head or hand was just as effective as the spoken command).
The following cases, chosen in a haphazard fashion, show that the horse's indication of the object's place in the series, by means of tapping, was by no means a guarantee that he would point it out correctly. Five placards hung from the cord. Mr. von Osten asked: "What is the position, counting from left to right, of the placard which has the word 'aber' inscribed upon it?". Hans answered: 3. (It was indeed the middle placard.) Then he was commanded: "Go!". Thereupon Hans went straight to the fourth placard. On another occasion Hans happened to drop a brown cloth upon a black one. His master asked him: "In which place are there two cloths?". Hans responded correctly, "In the second place". To the question "Which of the two is the black one?" he also answered rightly: "The lower one". Upon being asked to get it, he brought the white cloth.
The large number and the irregularities of the errors showed that there was no manner of intelligence involved in the pointing out process. Thus during the two months of our experimentation Hans was asked twenty-five times by Mr. von Osten to bring the green cloth. Only six times did he succeed in the first attempt, while in fiveinstances he selected an orange-colored cloth, four times a blue, three times a white one.
The fact that the errors were equally distributed over the tests with the colored cloths and those with the placards is strong evidence that the horse's response involved no intellectual process, for if that were the case, then the responses in the tests with the placards would have been very much more difficult, for they would have involved the ability to read, whereas the tests with the colored cloths demanded only that a few names be remembered. Nevertheless, the horse was as unsuccessful in tests of one kind as he was in those of the other,—even when Mr. von Osten acted as questioner. (50% failures in 78 placard tests; 46% failures in 103 color tests.)
The fact that commands which were purposely enunciated poorly, or else not spoken at all, were executed with just as much accuracy as those given aloud, strengthened us in our supposition. On one occasion I placed a blank placard with the others. When I ordered him to approachtabula rasa, he invariably went to the right one. The following illustrates how he fulfilled quite nonsensical commands. A series of blue and green cloths lay upon the ground. Being asked where the black, the orange, and the yellow cloths lay, Hans shook his head energetically, i. e. they were not there. And yet, upon being asked to bring them in the order named, he regularly brought one of the blue ones.
All this goes to show that Hans did not know the names of the colors (to say nothing of the symbols on the placards). It was plain that here also, as in all the other cases, he was controlled by signs made by the questioner, the nature of which I soon discovered. Standing erect, Mr. von Osten always turned head and trunk in the direction of the cloth or placard desired. Hans, keeping his eye on his master, would proceed in that direction. Even after he had already started out, thanks to his large visual field one could control his direction by turning slightly more to the right or to the left. If, however, he had already arrived at the row of placards or cloths, this method ceased to be effective, for then he could no longer see the experimenter. It made no difference whether the cloths lay on the ground, or were suspended, like the placards.
The following fact justifies the conclusion that the bodily attitude of the questioner was the effective signal. The more numerous the cloths, or the nearer they were placed together, the more difficult one would expect it to be for the horse to select the one indicated by the experimenter. Such was indeed the case, for the number of errors increased with the number of cloths presented.
But no matter how many cloths there might be, or how closely they might be placed, it was always possible to indicate either end of the row, for in that case one had merely to turn to the extreme left or the extreme right, and might even turn beyond the row. Hans seldom failed in these cases, whereas he made many errors when cloths or placards within the series were wanted.
To turn from the nature and number of Hans's errors, to their distribution,—observation proved the hypothesis that the nearer two cloths lay together, the greater was the chance of their being mistaken one for the other. If we designate as "error 1" all those cases in which Hans went to cloth II instead of to cloth I, cloth III instead of cloth II, to V instead of IV, etc., and as "error 2" when he mistook III for I, IV for II, in fine, whenever he went two places too far to the right or left, and as "error 3" whenever he went three places too far to either side of the cloth desired, we find the following grouping of errors: