Chapter V: Explanation of the Observations (2)

[Footnote Z: Since no opportunity was given us to examine Hans's
eyes we do not know what their condition is in this respect. Though
it would have been interesting to know, it would hardly make any
difference in the views presented. If Hans should prove to be either
far or near-sighted, then, if we are to make any supposition at all,
it would be that the defect could not be very great, since near
sightedness exceeding 2 or 3 diopters and far-sightedness exceeding
one diopter is seldom found in the case of the horse. According to
Mr. von Osten, Hans at one time manifested a tendency to shy easily.
Be this as it may, for little could be concluded from it, since in
many extremely shy horses, no kind of visual imperfection can be
discovered.]

But in the case of some of the horses in whom Berlin had seen the phenomenon for which we sought in vain, he himself tells us, the deflection was very slight. In that case, it would appear, no great advantage would be gained along the lines indicated. But even assuming the degree of deflection to be very great, his theory goes to pieces on the very point it was supposed to explain. A concrete example will make this clear. If Mr. von Osten, standing two feet away from the horse, raised his head 1/5 millimeter (which figure by no means represents the extreme values that were obtained), then in the horse's retinal image every point of the man's head would move through a distance of 0.0025 millimeter--assuming the horse's eye to be free from astigmatism and assuming its focal distance to be 25.5 millimeters. If, however, other conditions remaining the same, we presuppose an extreme form of astigmatism, one in which the path of the retinal image is not a straight line, but is deflected into a semicircle, then each point would pass through a distance of nearly 0.004 millimeter. If the sensitive retinal elements have a diameter of 0.002 millimeter (as Berlin, somewhat inexactly, states), then from two to four elements would be stimulated in case there were no astigmatic deflection. But in case the deflection did take place, it would not necessarily involve more elements, as can be seen by making a simple graph; indeed we can imagine cases in which the circuitous path would involve even fewer elements than the straight one. And finally, when the movement which the horse is to perceive, does not occur in a straight line but in the form of a curve, (which will generally be the rule), then the astigmatism will tend in many cases to decrease the curvature of the image's path on the retina, and sometimes even obviate it entirely. In all these cases, on Berlin's own theory, the perception of the movements would be hindered rather than aided.[AA]

[Footnote AA: For the benefit of specialists I would say the
following in addition to the more general remarks just made. For the
most part, the determinations of refraction made on the eye of the
horse are still rather unreliable. In sciascopy there is a dispute
among investigators concerning ambiguous shadows, and in the use of
the refraction-ophthalmoscope no definite region of the eye's
background has been adhered to by the various investigators. It
appears that Riegel, whose diligent researches mentioned on page 164
were published in 1904, knew nothing concerning the round area in
the horse's eye, discovered by I. Zuern[42] in 1902. Also, if so
great a degree of astigmatism is really the rule as is emphasized
especially by Hirschberg[43] and Berlin,[44] then the simple
refractive index usually given--sometimes within a half
diopter--would be meaningless. Berlin[45] and Bayer[46] believe the
vagueness of the retinal image resulting from the astigmatism, is
offset by this: that the oval pupil functions as a stenopaic slit.
In view of the width of the horse's pupil this appears to me to be
rather hypothetical.

Concerning Berlin's theory of deflecting astigmatism I would say the
following: Of the two ophthalmoscopic signs mentioned as being
characteristic of this form of astigmatism,--the concentric circles
and the arcuate deflection of the pathway of the fixated
points,--when there is a movement of the eye of the observer (or of
the eye observed), according to Berlin the former is not so constant
as the latter. So far as I know, the concentric ring formation is
mentioned only by Bayer[47] and Riegel,[48] and is said to occur
principally in horses with myopic vision--and hence, relatively, in
a minority of cases. Judging from the particulars, we are inclined
to believe that a case of "Butzenscheiben"-lens reported by
Schwendimann[48_a_] is in reality a case of senile sclerosis.
Berlin repeatedly warns us against mistaking the one for the
other.[48_b_] The arcuate deflection, on the other hand, has not
been mentioned elsewhere as a personal observation. In Berlin's
calculation[49] of the increase in the extent of the retinal pathway
an ambiguity has crept in. He says that "in the astigmatic eye there
are stimulated 207 times as many nervous elements as would be
stimulated in the ideally normal eye." It ought to read "207 more"
instead of "207 times as many." And this number holds only for the
one case computed by Berlin, and under the specific assumption that
exactly [Greek: pi]/2 times the normal number of elements were
stimulated (571 instead of 364). Therefore the general statement
which Bayer[50] makes in his text-book, that according to Berlin's
evaluation "207 times more nervous elements" are stimulated in the
astigmatic eye than in the non-astigmatic one, does not hold true.

Closing this note, a few remarks concerning the experiments made by
Dr. Simon and myself. All of the nine horses were tested for the
vertical image by means of the ophthalmoscope. In most cases Wolff's
electric speculum was used. Atropine was not employed.--For the
laboratory tests the adipose and the muscular tissues were removed
from the eye-ball and the rear part of the bulb cut away. The front
part, containing the cornea and the lens, was fastened over one
opening of a metal cylinder which was closed at the other end by
means of a disc of ground glass. The whole, approximately as long as
a horse's eye, was filled with a normal salt solution whose
refractive index (1.336) corresponds quite closely with that of the
vitreous humor of the horse's eye. The pressure from within was
regulated so that on the one hand it was not dimmed and yet on the
other there were no wrinkles in the cornea. The source of light--the
filament of a Nernst lamp--was moved about in a plane 120 cm.
distant from the eye and perpendicular to the optic axis. It was
moved through the point of intersection as well as at various
distances from it. Movement in horizontal and vertical directions
was in each case along lines 150 centimeters in length, which would
correspond to an angle of vision of not less than 64 deg.. The pathway
of the imaged point was controlled by means of the cross-hairs of
the telescope. If in the same way we observe through the sclerotic
of an intact eye-bulb a point of light falling upon the retina and
shining through the sclerotic and choroid (which is not difficult
when we use an intense light), then to the observer its pathway
will, of course, appear to be deflected convexly toward the
periphery,--and the deflection will appear the greater, the farther
the point of light is removed from the optic axis.]

But to come now to the most pertinent objection. We saw that Berlin's whole train of thought rested upon the assertion that it made no difference whether we regarded by means of the speculum the seeming movement of a fixed retinal point, or whether the image of an external moving object is passing over the horse's retina. As a matter of fact, however, these two processes are very different from one another. In moving the mirror, with its small opening we are looking through ever changing portions of the horse's lens,--testing it out, as it were. The horse, on the other hand, sees with all parts of the lens simultaneously, in so far as the lens is not covered by the iris. The arcuate deflection, which is nothing but a registration of the difference in the indices of refraction of the different parts of the lens used consecutively, might thus be formed for the observer using the mirror, but never for the horse. For these reasons we cannot conclude that the kind of astigmatism described can really increase the horse's acuity in the perception of movements.

Since the light-refracting apparatus of the horse's eye does not offer a satisfactory explanation for the extraordinary keenness of visual perception possessed by the Osten horse, we must go a step further and ask whether it may not perhaps be found in the part immediately sensitive to light, the retina. That portion really would seem to be adapted to the perception of movements of minimal extent, and for this reason: it is more than three times as great in extent as the human retina, and the horse's retinal images are likewise larger owing to the position of the nodal point. The cells of the retina that are sensitive to light, the rods and cones, might therefore be correspondingly larger than those of the human eye, without thereby making the whole organ less efficient than the human eye. But the most recent measurements[51] have shown that the rods and cones of the horse's eye are more minute than ours. Assuming that, in the case of the horse, as is presumably the case in human vision, the transition of a stimulus from one retinal cell to the next already in itself induces a sensation of movement, then the horse ought indeed be extraordinarily keen in the perception of moving objects (provided that the horse's more minute cells are packed just as closely as in the human retina). And besides, there are two specially adapted areas within the retina of the horse. The "band" ("streifenfoermige Area") which was discovered fifteen years ago by Chievitz,[52] is a strip of 1 to 1-1/2 millimeters in width, traversing the entire retina horizontally, and is noteworthy on account of its structure and probably, too, on account of its greater efficiency. It may have something to do with the accomplishments of the Osten horse; but in how far it would be hard to say. The other noteworthy portion of the horse's retina is the "round area" discovered some four years ago, located at the rear outer end of the "band", and it is the best-equipped part of the horse's retina and corresponds to the area of clearest vision, the yellow spot, in the human eye. But this round area need not come in for consideration by us, for its location would indicate that it is used in binocular vision, that is, seeing with both eyes.[53] But in all our experiments the Osten horse observed only with one eye. That does not mean, however, that under other circumstances the round area may not be of very great importance.

In the present state of our knowledge, all attempts at explanation are, of course, of the nature of hypotheses. If further investigations should disclose this explanation to be untenable, then we would either have to suppose some unknown power in the eye of the horse,[AB] or else seek a cause in the animal's brain. Further experiments on other horses would be necessary in order to discover whether the species as a whole possesses this ability or whether only certain ones are thus endowed. The former is of course more probable. In this particular case conditions were unusually favorable for the development of this ability. We must bear in mind that in all probability Mr. von Osten's movements very gradually became as minute as they are now, and that therefore Hans at first learned to react to such as were relatively coarse. Furthermore, his practice extended throughout four years and during this time it was his sole occupation. Without specific predisposition, however, all this practice would have been utterly futile. We can also readily appreciate how indispensable in the struggle for existence a well-developed power of perceiving moving objects must be to horses (and most other animals) living in their natural condition and habitat, in order to be aware of the approach of enemies, or, in the case of carnivora, the presence of prey. In view of all these considerations we can readily see how it was possible that the horse, perhaps in spite of rather defective vision, could react with precision to movement-stimuli which escaped observation by human eyes.

[Footnote AB: Koenigshoefer, who as we have already said, seconds the
explanation given by the ophthalmologist Berlin (and who confounds
"Butzenscheiben" astigmatism with the common, so-called regular
form), believes[54] that not only astigmatism but also the shape of
the blind-spot of the eye must be taken into consideration. This
portion of the retina, where the fibres of the optic nerve enter the
eye (and called "blind-spot" because there are no cells there that
are sensitive to light) is very nearly circular in man, but differs
in shape in the different species of animals. Koenigshoefer thought he
had discovered that a relatively elongated blind spot was favorable
to keenness of vision. If we place the mammalia in series on the
basis of their relative keenness of vision, he says, we would find
that this series is identical with the one in which they are grouped
with reference to the form of the blind-spot from the circular up to
the most elongated. (In such a series the marmot takes the place of
honor.)

This exposition is not very satisfactory, however. We cannot be sure
what he means by "keenness of vision" ("scharfaeugigkeit"). Is it
visual acuity in the usual sense of the term (as is said in one of
his passages), or keenness in the perception of the movements of
objects, (this would appear to be his real meaning), or both at the
same time. But whatever the significance he may put into the term,
any such attempt at grouping the lower forms must prove
unsatisfactory from the very start on account of the scant data
which we possess on visual perception in animals. The experiences of
the hunt upon which Koenigshoefer partly bases his view, are entirely
inadequate for such a purpose. This much is certain, that the Osten
horse, in spite of a blind-spot which, though somewhat oval, is by
no means very elongated, possesses an extraordinary acuity in the
perception of movements. Even if the parallelism mentioned by
Koenigshoefer were really shown to exist, it would not explain the
matter until it were also shown in what way keenness of vision is
dependent upon the shape of the blind-spot,--a portion of the eye
which is not immediately operative in the visual sensation at all.]

We can understand also the horse's never-flagging attentiveness when we recall that self-preservation prompts eternal vigilance over against all that is going on in the animal's environment. (In the case of Hans, hunger was at first the motive; later, habit did the work.) Furthermore, the lower form is not hindered in giving itself over to its sense-impressions by the play of abstract thought which tends so strongly to direct inward our psychic energy,--at least, in the case of the cultured.

Nevertheless, Hans still remains a phenomenon not only in excelling all his critics in the power of observation, but also in that he is the first of his species, in fact the first animal, in which this extraordinary perceptual power has been proven experimentally to be present. It has long been known[55] that horses could be trained to respond to cues in the form of slight movements, which remained unnoticed by the layman, and this fact has been made use of by circus trainers to its fullest extent. But such signs, I have discovered, are without exception, of a far coarser sort than those we have here described, and they can be instantly detected by the practised observer. Nor was it known to professional trainers that it was possible for the master to direct a horse to any point of the compass simply by means of the quiet posture of the body. For this reason it was believed that no signs could possibly be involved in the color-selecting-tests (cf. Supplement III, page 255). In this we have the support of some of our experts, as is witnessed by the following extract from a letter of his Excellency Count G. Lehndorff, one of our best hippological authorities, who at one time carefully examined the Osten horse. (The letter was addressed to Mr. Schillings, and I have permission of both gentlemen to use it). In it he says: "If the author's statements, in which you also have concurred, are correct, and if, as a matter of fact, the horse really does react to such minute movements as are absolutely imperceptible to the human observer, then we have indeed something quite new, for hitherto no one would have believed that horses can perceive movements which man cannot. But I am even more surprised by the explanation of the color-selecting feats.--This too, is something absolutely new. One would not have deemed it possible that a horse could do anything of the kind simply by using the posture of a man's body as a cue to which it could react with such precision."

And yet, even though both facts were new concerning the horse and had not hitherto been proven experimentally regarding any other species, nevertheless something of this sort has been known concerning the dog for some time. His ability to single out an object upon which his master had intently fixed his gaze, was made the basis of a special form of training, called "eye-training,"[56] nearly one hundred years ago. The dog was taught to focus constantly upon his master's eyes and then upon command to select the object which he, the master, had been fixating. Such a dog has been described by the naturalists A. and K. Mueller.[57] But the master of the dog, unlike Mr. von Osten, would not permit anyone else to work with the animal, and the two brothers, recognizing the trick, were justified in adding that "the whole affair aimed at deceiving the public, and the dog's reputation was but a means of making money". The success of such exhibitions appeared furthermore, to depend upon the close proximity of the trainer and the dog, whereas the direction of the head (and even of the body) could very probably be perceived at greater distances also. At least we learn from a reputable source that in the hunt, dogs can perceive from the mere posture of their master, what direction he intends to take.[58]

But a still more curious fact is this, that dogs, too, learn--evidently spontaneously--to react to the minimal involuntary expressive movements of their master. The first example mentioned in the literature on the subject is that of an English bull-dog called Kepler, belonging to the English astrophysicist, Sir William Huggins.[59] We are told that this dog seemingly could solve the most difficult problems, such as extracting square roots and the like. The numbers were indicated by barking,--thus one bark was for one, two barks for two, etc. Every correct solution was rewarded with a piece of cake. Huggins states explicitly that he gave no signals voluntarily, but that he was convinced that the dog could see from the questioner's face, when he must cease barking, for he would never for an instant divert his gaze during the process. Huggins was unable, however, to discover the nature of the effective signs. This satisfactory, though still unproven, explanation has been accepted by specialists, among them Sir John Lubbock.[60] I, too, regard this dog as a predecessor of our Hans.

A similar case is reported by Mr. Hugo Kretschmer, a writer of Breslau, in the "Schlesische Zeitung" of August 21, 1904. To him I am beholden for a detailed written statement, which he has kindly permitted me to use in this connection. The gentleman named, first trained his dog to ring the table-bell, and this, by pressing the dog's paw upon the bell-button. When the dog had learned to do this independently, his master tried to teach him the rudiments of numbers, in such a way that the animal was to give one ring of the bell for the number 1, two for 2, etc. But these attempts failed utterly and had to be abandoned. But Mr. Kretschmer had noticed that he was able to get the dog to ring any number which he, Mr. Kretchmer, might decide upon. (Success was always rewarded by a bit of bread and butter.) At first Mr. Kretschmer tried to imagine vividly only the final number, but failed thereby to elicit correct responses from the dog. But he did succeed when he tried making a series of separate volitions. Thus for the number 5, he would "will" each separate push of the button on the part of the dog. Even so, however, he never got beyond 9, for then the dog would become impatient and would ring the bell continuously. Anything that diverted the dog's attention, such as noises, etc., also entailed failure. In these tests master and dog had faced each other, each gazing steadfastly at the other. Mr. Kretchmer was convinced, however, that the dog was not guided by any sort of sign, but rather by suggestion. He based his belief on the following two observations. After some practice, he says, the tests were also successful when he did not look at the dog, but stood back to back with it, or when he screened himself from the dog's view by stepping to one side behind a curtain. The tests were unsuccessful, on the other hand, whenever he was mentally fatigued or had taken some alcoholic drink. The arguments do not appear to me to be adequate. If he turned his back upon the dog and no other observer was present, he had no means of knowing whether the dog did not, after all, peer around to get a peep at him. If others who knew the desired number, were present, the dog might have gotten his cues from them. And there may be some doubt whether the curtain adequately served the purpose for which it was intended. At any rate, it was added that all attempts to influence the dog from an adjoining room--which would thus exclude effectively all visual signs--were utter failures. I am also strengthened rather than weakened in my belief, by the second argument which Mr. Kretschmer makes, viz.: that mental fatigue or the use of alcohol on the part of the questioner tends to make the result unsatisfactory. We noted a similar effect in the case of the horse (page 150), where a disturbance of the "rapport" between the questioner and the horse was invoked by some by way of explanation. The facts were explained by us much more simply. We attributed the result to the close correlation between the type of mental concentration and the nature of the expressive movements--a correlation which we have shown experimentally to exist. I cannot, therefore, subscribe to the view that this dog did not require either visual or other sensory signs. The tests which were made for the purpose of strengthening that view, are on a par, I believe, with those mentioned on page 45. And since auditory, olfactory, and other stimuli, though not impossible, still are improbable, I believe that our Hans, Huggins's dog, and the one belonging to Mr. Kretschmer, differ from one another only in this, that the first taps, the second barks, and the third presses a bell-button.

And finally I have access to a letter from the Rhine Province in which there is a brief account of a dog that would promptly obey any command that was given without a sound and supposedly without the accompaniment of the slightest kind of gesture. It is specially mentioned that the animal steadily watched its master during these tests. The perception of the slightest involuntary expressive movements is in all probability the secret in this case also. Here, too, suggestion has been invoked by way of explanation, but there was not the slightest attempt made to find for it a more specific foundation, and we cannot suppress an objection based on the matter of principle. It is incumbent upon anyone who uses a term so ambiguous, to define what content he desires to have put into it. If he does not do this, he is giving us, instead of a concept, a bare word, instead of bread, a stone.

While we must reject the explanation based on suggestion,[AC] we believe, on the other hand, that we have here again, evidence of the presence of visual signs, given unwittingly and involuntarily, just as I am sure that they were involved in the two preceding cases, and similarly in the case of the Huggins dog. Since the effective signs were discoverable in none of these canine predecessors of Hans, an investigation would be desirable, based upon the insight gained as a result of these experiments upon Mr. von Osten's horse. Unfortunately this is impossible, since the dogs in question are dead. But others like them undoubtedly exist in many places. We might mention that when Hans first came under the limelight of public attention, there was also frequent reference to the Huggins dog, but he soon dropped out of the discussion again.[63] And this for two reasons. The dog never took his gaze from his master and appeared to be entirely dependent upon him in his reactions. Hans, on the other hand, seemed to give evidence of a high degree of independence and never appeared to look at the questioner. But we know now that, though he was never dependent upon the will of his master, he, too, abjectly hung upon the man's involuntary movements and never for a moment lost him from view. But since the horse is able to observe with one eye alone, and needed to direct only it and not the entire head toward the questioner, in order to focus comfortably, one could not conclude as to his line of vision from the direction of the head. Since, furthermore, in the horse the pupil is hardly distinguishable from the darkly pigmented iris and since the white sclerotic is hidden by the eyelids, except when the eye is turned very much, it is difficult to determine what direction the eye is taking. I once purposely stepped backward to the horse's flank, so that he had to turn his eye far back and thus the outer border of the iris and the white sclerotic coat became visible and all doubt concerning the line of vision was removed. This doubt could never arise in the case of the dog, the median plane of whose head is always directed toward the object fixated, and Zborzill is justified in saying, as he does, in his discussion of training of the kind mentioned on page 177, "But any careful observer can immediately guess the manner in which such a dog has been trained."[64] If Hans had chanced to possess so-called "glass-eyes"--in which the dark pigment is wholly or partly lacking, so that the black pupil is clearly defined against the lighter background,--then no doubt could ever have arisen concerning the direction of the eye, and Hans never would have come to be regarded as the "clever" Hans.

[Footnote AC: I can find examples of supposed suggestion in the case
of animals given only by Rouhet.[61] He says that by means of
suggestion he taught a half-year old half-blooded mare-colt which he
had raised himself, to fetch and carry, and this in a very short
time. In order to indicate to the colt what was wanted, Rouhet would
concentrate with his whole mind upon the object intended (a watch),
and at the same time he would bend forward slightly. In the third
test, that is at the end of fifteen minutes, he had accomplished his
purpose, and in the tenth lesson, no more mistakes occurred. The
colt would fail to respond, however, as soon as he refrained from
making any gestures, or was in a laissez faire frame of mind, or
when he thought of other things. He therefore believes that there
must have been some kind of immediate, though inexplicable,
connection between the brain of the trainer and that of the horse. I
think the explanation is evident: the connection was not as he
thought, an immediate one, but arising through the mediation of the
man's attitude ("attitude un peu baissee"), and of his movements
("gestes"), both resulting from his intense concentration ("tension
de la pensee").

In general we may say that, no matter what content we may wish to
put into the term "suggestion," not a single fact has since come to
light which would justify, and much less demand, the application of
the term to lower forms, unless we would expand the definition of
the term to the extent of comprising every kind of command, every
arousal of ideas, whatsoever. But it would then be nothing but a new
name for old knowledge[62] and would lose all explanatory value.
(Hypnotism, so-called, in the case of horses, I shall discuss
elsewhere in another connection.)]

After the publication of the December report, Hans acquired a reputation for excellence in thought-reading and thus the discussion of thought-reading among animals in general became once more the order of the day. That is to say that many of our domestic animals are--like the human mind-reader (a la Cumberland),--supposed to have the ability to infer the thoughts of their masters from slight, involuntary movements. They are thus aware when the feeding hour approaches, when they may go out in the open, etc. They also appear to be aware that their welfare lies in our hands, and therefore would seem to have a vital interest in divining our intentions and our wishes. Not only our spoken words, but also numberless movements--usually without our knowing it and often contrary to our desire--speak a clear language. As is well said by the American neuropathologist, Beard,[65] (who first explained the phenomenon of thought-reading, on the basis of the perception of very minute muscular jerks, and therefore called it "muscle-reading" or "body-reading"): "Every horse that is good for anything is a muscle-reader; he reads the mind of his driver through the pressure on the bit,--though not a word of command is uttered." We know that in the case of perfectly trained horses the rider's mere thought of the movement which he expects the horse to make, is seemingly sufficient to cause the animal to execute it.[AD] Such cases are of course very much like that of our Hans, excepting that instead of visual signs they involve aids of a mechanical nature, which, however, does not alter the general principle, since both of them are of the nature of sensory stimulation. But we must not overlook the essential difference between this so-called thought-reading on the part of animals and that which is done by man. The human thought-reader can interpret movements, for he is familiar with the ideas which are their source. Thus when at the second tap, I notice a very slight jerk of the subject's head, and a stronger one at the fifth tap, I infer that he thought of the problem 2+3=5. While the experimenter thus cannot be said to read thoughts, he still infers them. The animal, on the other hand, we may be reasonably sure, draws no such inferences. In its conscious life it remains ever on the sensory level. If we could ask Hans about it, he would probably answer: "As soon as my master stoops forward, I begin to tap; as soon as he moves, I stop. The thing which induces me to act thus is the carrot which is given me; what it is that induces my master to make his movements, I do not know."--It is therefore erroneous to believe that animals require the power of abstract thinking in order to utilize the signs which are consciously or unconsciously given them, as is argued by Goldbeck[68] when he says with reference to the training for visual signs, which we have already mentioned before: "There the dog has consciously interpreted the visual impression in terms of the conclusion that he is expected to bring forth the leaf indicated." Nor was there any justification for the critic who thought he could put the essence of the report of December, given in Supplement IV, into the following words: "He (Hans) showed that he has the power of attention, can draw logical conclusions, and can communicate the result of his thinking,--and all this independently." Yet none of this had been asserted. The whole thing may be explained satisfactorily by means of a process of simple association established between the signs observed in the master and certain reactions on the part of the horse. The fact that the movements made were so exquisitely minute does not change the matter in the least. Such signs call for a high degree of sensory keenness and great concentration of attention, but by no means an "extremely high intelligence."

[Footnote AD: An illustration is given by Babinet[66] concerning the
horse of an English lord. Mr. Burkhardt-Foottit, also, that
excellent trainer, who has been master for more than forty of the
most highly-trained horses, tells us that while sitting on a
well-managed horse it sometimes happened that he had merely thought
of making a certain turn, when the horse immediately executed it,
before he, the rider, had to his knowledge given any sign or aid. An
observation belonging under this head is also made in Tolstoi's
"Anna Karenina"[67], this perfect mine of acute psychological
observation. In the famous description of the race we are told
concerning Count Wronskij riding his Frou-Frou just behind Machotin
mounted upon Gladiator, who was leading the race: "At the very
moment when Wronskij thought that it was time to overtake Machotin,
Frou-Frou, divining her master's thought, increased her pace
considerably and this without any incitement on his part. She began
to come nearer to Gladiator from the more favorable, the near side.
But Machotin would not give it up. Wronskij was just considering
that he might get past by making the larger circuit on the off-side,
when Frou-Frou was already changing direction and began to pass
Gladiator on that side." Similar experiences might be gathered
elsewhere. Not infrequently the reflection of the rider that his
horse had not for a long time indulged in some trick peculiar to
him, will immediately call it forth; or doubts on the part of the
rider concerning the possibility of crossing some barrier, are often
the cause of the horse's fall or of his refusal to leap and of his
running away.]

Let us turn now from the consideration of visual perception to that of auditory perception in the horse. We saw that the fact that Hans was able to respond to commands which were only inwardly enunciated, that is, commands which were merely thought of but not spoken, was not proof of great acuity of hearing, but rather that hearing was not at all involved. If Hans had been deaf he would, none the less, have promptly obeyed the commands. Blind and near-sighted horses try to overcome their deficiency by means of the sense of hearing, and hence show a pronounced play of ears. In the case of the Osten horse, however, attention has been diverted from auditory stimuli in the process of habituation to visual signs, and as a result ear-movements are almost completely wanting. One is not of course permitted to deny _a priori_ that perhaps some associations might have been formed between objects and the vocal signs belonging to them, e. g., between the colored cloths and the names of the colors if both had been presented together oftener than was the case.

But there is a dearth of reliable observation as to how far auditory associations of this sort may be established in horses. Usually the following is cited. Horses learn to start off, to stop, and to turn about in response to calls. They are able to distinguish properly between the expressions "right" and "left", or equivalent terms. Upon command they will start to walk, to trot or to run. And they also know the name by which they are usually called. All authors agree that cavalry horses understand the common military commands; one writer even avers that they excel the recruits in this respect.[69] Some believe that in riding schools the horses pay closer heed to the calls of the riding-master than to the control of unpractised riders, even when the two are at variance with one another.[70] My experience with the Osten horse and a number of other pertinent observations aroused in me the suspicion that much that is called or spoken in the process of managing a horse may possibly be just so much labor lost. In consequence I made a series of relevant experiments. I have thus far tested twenty-five horses of different kinds, from the imported Arabian and English full-blood, down to the heavy draft-horse. The experiments were made partly in the courtyard of military barracks, partly in the circus, and partly in a riding-school or in private stalls. I am specially indebted for kind assistance to Messrs. von Lucanus, Busch, and to H. H. Burkhardt-Foottit and E. Schumann, the two excellent trainers connected with the Busch Circus. During these tests, the horses were always amid circumstances familiar to them, whether free or bridled, under a rider or hitched to a wagon. All aids or signals, except the calls, were eliminated in so far as it was possible.

The results of those tests were in substance as follows: Many horses react to a smack of the lips by a rather fast trot. Many stop on the cry "Hola" or "Brr". This last was nicely illustrated in the case of two carriage horses supplied with large blinders and held with a loose rein, and hitched to a landau. One of them regularly stopped when the "brr" was given by the driver, whereas the other, which had not been habituated to this signal, kept serenely on the trot, so that the vehicle regularly veered off the track--a sure sign that no unintentional aid was being given by means of the reins. Other horses, again, were accustomed to halt in response to a long-drawn-out "hola", but it was the cadence of melody rather than the word that was effective, since any other word, or even a series of inarticulate sounds, would produce the same result, provided they were given with the proper inflection. When this was changed, then the response would fail.

The result was not so apparent when it came to controlling the kinds of gait. One riding-school horse, when lunged and in a gallop, could be induced by a friendly call--the word again was a matter of inconsequence--to slacken his pace into a trot and from a trot into a walk. But this reaction was by no means very precise. Another, a full-blood, contrary to the trainer's expectation and to his great astonishment, failed to respond to any kind of spoken command as soon as the one who carried the reins refrained from making any movements which might indicate what was wanted. (To refrain from all expressive movements of this kind is by no means an easy matter). The slightest move, apart from any help by means of the reins or the whip-handle, was sufficient to evoke a response. The results in the case of the military horses, differed in many particulars. Thanks to the courtesy of Captain von Lucanus I had the opportunity of testing three cavalry horses, two geldings and one mare, aged nine, thirteen, and nineteen years respectively, and all of them in the regiment ever since their fourth year. They had been selected as the "most intelligent" in the squadron, and we were assured that they would obey punctiliously all the usual commands. They were ranged behind one another, with the customary distance of two horses' lengths between, and were ridden each by his accustomed rider. Both starting and stopping upon command were tested. The horses were held by the reins, but the riders were cautioned to refrain from giving any aid that might cause the horse to start when starting was to be tested, or that might restrain him when stopping in response to the spoken command was to be tested. If a suspicion arose--a thing which happened only twice, however--that a rider had actively aided in his horse's reaction, then an officer would mount into the saddle. If it appeared that one of the horses was simply imitating the others, then the others were purposely restrained by their respective riders. The commands were given by the corporal who usually had charge of the horses. In a few cases the sergeant of the squadron gave the commands, but this made no difference in the success of the experiment. Now as to the results. Whenever the horses were trotting or walking, all commands, without exception, were in vain. They effected neither an increase nor a decrease in the pace. A result was obtained only when the horses were standing when the test began; and this result was simple enough,--upon certain calls the animals would respond by beginning to walk. This was the only reaction that was obtained. The most effective of the commands appeared to be "Squadron,--march!" But the command "Squadron!" or "March!" alone, were quite as effective; yet none of these commands was obeyed without exception. Reactions were occasionally obtained in response to "trot!", "gallop!" "retreat!", (the usual introductory "squadron" was purposely omitted here, because it alone sufficed to start the horses). But the reactions were always the same, viz., to start on a walk. Another series of commands (such as those which are addressed to the rider alone, e. g., "Lances down!") had no effect whatever; a certain amount of selection therefore did seem to take place. In all these tests the order of the horses with reference to each other's position was repeatedly changed. One of the horses, the youngest, and reputed to be the most "intelligent", (he was as a matter of fact the most spirited), gave evidence of a gregarious instinct, intensified by habit, which, if it had been overlooked, might have become a source of serious error. Not being accustomed to go at the head, when so placed it started properly in only 18% of all such cases. When, however, (other conditions remaining the same,) he was put in second or third place, he started properly in 67% of the tests, and if we take into account only those cases in which the three most effective commands were used ("Squadron!", "March!", and "Squadron--march!") he reacted correctly in 91% of the cases. (The number of tests was 17, 36 and 22 respectively for the three groups mentioned.) The horse, therefore, almost always began to step properly when he stood behind one of his companions, but seldom when he stood at the head. And when he stood at the head and began to walk at the proper moment, it was plain that it was a case of imitation and not initiative, for the horse was still able to see the others, owing to the extent of his field of vision backward, and he was always the last to move, whereas otherwise he was always the first to move, and always difficult to restrain. So when the horses to the rear were restrained or when the intervening distance of two horses' lengths was lessened, so that this gelding could not see the one in the rear, he failed completely to respond. Accordingly these three horses did little to justify the faith which their squadron had placed in them.

Now a few words on the manner in which horses react upon the call of their names. We are not concerned with those that are seldom or never called by name (such as those in the cavalry). I have not discovered one horse that constantly and unequivocally reacted upon the mention of its name (though I would not assert that there are none that would do so.) I was nearly always able to convince the owners or grooms, who at first had maintained a contrary opinion, that any inarticulate sound was capable of producing the same effect as the calling of the name. What the significance of inflection may be, I am not at all certain. When a certain one of a number of horses standing in the same stable was called, all of them responded by pricking their ears, raising their heads, or else turning about. For this reason the reaction of the horse specifically called lost all significance. Likewise the call which is ordinarily used in lunging when the man in the center of the circle wishes the horse to change its gait, or to advance toward him, also proved ineffectual as soon as the man inhibited every sort of movement. A slight nod, on the other hand, was always effective. Several times I have tried to call horses to me, when they were free and running about in the arena, but was unsuccessful. After I had given them some sugar, however, they would always come to me--whether I had called or not--and would then refuse to leave my side. But this is a matter of common observation.

I would, however, regard all of these tests as merely provisional. In spite of the greatest effort, it was not always possible to control all the conditions of the experiment, and furthermore, the number of tests would have to be materially increased in order to yield an appreciation of the difference due to race, age, and the individual variation and training of horses. But we may, even now, be sure of one thing. Over against the certainty with which horses react to visual stimuli (in the form of movements perceived), it does not appear that the formation of auditory associations is greatly favored by nature in these animals,--indeed, auditory associations are far less common than is generally supposed.[AE] Horses compare very unfavorably with dogs in this respect. The latter easily learn to react with a high degree of precision to auditory signs,--as I learned from a series of experiments which I was enabled to perform. The Osten horse, therefore, does not stand alone among his kind in his inferior auditory equipment, as one might be tempted to believe at first blush.

[Footnote AE: All the authors who have given practical suggestions
for the training of horses, whether free or with lunging reins, have
great faith in the efficacy of calls, but usually recommend a
mingling of calls and movements in the way of signs, (thus
Loiset,[71] Baucher,[72] von Arnim[73]). It therefore cannot be
stated just in how far the calls really effect anything. In other
cases I am inclined to doubt outright the influence which is
ascribed to the auditory signs. Meehan[74] gives an account of a
horse that was exhibited in London in the early 90's of the last
century. Pawing with his hoof, this horse apparently was able to
count and answer questions in arithmetic, and among other
accomplishments he was supposed also to be able to understand
something of language. In reality, however, he merely responded to
cues which were disclosed to the reporter by the trainer. In pawing,
the horse was guided by movements of the trainer, and in nodding or
shaking the head he reputedly got his cue from the inflections of
the man's voice. Is it not probable that in this latter case it was
the movements which accompanied speech that were alone effective in
inducing the nod or the shake of the head, so that the exhibiter was
deceiving not merely the public, but also himself? Perhaps we may
also doubt the exposition made by the well-known hippologist,
Colonel Spohr.[75] He tells us that it is easy to train horses to
raise the left foot or the right foot in response to the commands
"Left--foot!" or "Right--foot!" and that it will be the fore foot
when one is standing in front of the horse, and the hind foot if one
stands near the rear. It cannot be so very difficult, he thinks,
even to get the horse to understand the commands "Left (or
right)--fore foot!" and "Left (or right)--hind foot!"--and all
without any other aids but the spoken words. Should this really be
possible without even the slightest kind of designating
movement?----The following case, again, I believe is undoubtedly
based upon a misinterpretation. Redding[76] relates concerning his
nineteen-year old horse that he himself had owned for thirteen
years, and had always kept in single harness,----that this horse not
only understood the meaning of a long list of words, such as:
bureau, post-office, school, churchyard, apple, grass, etc., but he
also knew a number of persons by name, as well as their places of
residence. If he were told in advance to halt at a certain
residence, he would do it without any further aid from the driver.
For this reason the happy owner felt certain that the animal
possessed a high order of intelligence and "that this horse does
reason." What sources of error were here operative, whether signs
were given by means of reins, or head or arm movements, could be
determined only by a careful examination of the case.

And finally we would exercise some reserve in entertaining the
suggestions for the acoustic education of horses which have come
from various sources. Colonel Spohr[77] whom we have just been
mentioning, thinks that it would not be a difficult matter to get a
horse to respond with a walk to one smack of the lips, with a trot
to two smacks, and with a galop to three, and then he could be made
to slacken his pace once more into a trot in response to one
long-drawn "Pst!" and to stop in response to two. Others have gone
even further. Decroix,[78] at one time leader in veterinary affairs
in France, conceived the idea of working out a universal language as
regards the commands that are given to horses, in the humane purpose
of sparing them the whip. He called it "Volapuek hippique." For the
commands "go," "right," "left," and "halt," he suggests these: "Hi!"
"Ha!" "He!" and "Ho!" respectively. From these it was possible to
make eight combinations, such as "Hi! Hi!" for "Trot!" "He! He!" for
"Left about" (while the single "He" was to mean "Forward, to the
left!") "Ho! Ho!" for "Back!" etc. Decroix thought that the whole
system could be inculcated in a very few lessons. He even had a
medal struck which was to be awarded to the driver or rider who
should first exhibit a horse, thus instructed, to the Societe
Nationale d'Acclimatation de France (of which Decroix was
president). Eight years have elapsed since then, but we have heard
of no one who has earned the medal mentioned. In the future greater
care will probably be exercised in the putting forth of such
suggestions, and two sources of error may be guarded against, viz.:
involuntary movements on the part of the rider or driver, and
imitation of the horses amongst themselves. (One horse, guarded by
an experienced rider, may serve as copy for ten others with
inexperienced men in the saddle.)]