The Animal Mind: A Textbook of Comparative Psychology

As one might expect, among the earliest experiments upon animals were those made by physiologists with a view to determining the functions of sense organs. The experimental movement in psychology was slow in extending itself into the field of the animal mind. Romanes, whose adherence to the anecdotal method we have noted, made in 1881, rather as a physiologist than as a psychologist, a number of exact and highly valued experiments on ccelenterates and echinoderms, which were summarized in his book entitled “‘Jelly-fish, Star-fish, and Seaurchins,” published in 1885. He has also recorded some rather informal experiments on the keenness of smell in dogs. Sir John Lubbock, in 1883, reported the results of some experiments on the color sense of the small crustacean Daphnia, and his book on ‘‘Ants, Bees, and Wasps,” containing an account of experimental tests of the senses and “intelligence” of these insects, appeared in the same year. A German entomologist, Vitus Graber, experimented very extensively at about this period on the senses of sight and smell in many animals. Preyer, the authority on child

psychology, published in 1886 an experimental study of the behavior of the starfish. Loeb’s work on the reactions of animals to stimulation began to appear in 1888. Max Verworn, the physiologist, published in 1889 an exhaustive experimental study of the behavior of single-celled animals. With the exception of Preyer and Romanes, all these men had but a secondary interest in comparative psychology : Bethe, indeed, as we shall see, wholly rejects it. Lloyd Morgan, who has written instructively on comparative psychology, makes but a limited use of the experimental method. Wesley Mills, professor of physiology in McGill University, has studied very carefully the mental development of young animals such as cats and dogs, but is inclined to criticise the use of experiment in observing animals. The work of E. L. Thorndike, whose ‘‘ Animal Intelligence” appeared in 18098, represents, perhaps, the first definite effect of the modern experimental movement in psychology upon the study of the animal mind. Thorndike’s aim in this research was to place his animals (chicks, cats, and dogs) under the most rigidly controlled experimental conditions. The cats and dogs, reduced by fasting to a state of “utter hunger,”’ were placed in boxes, with food outside, and the process whereby they learned to work the various mechanisms which let them out was carefully observed. Since the appearance of Thorndike’s work the performance of experiments upon animals has played much part in the work of psychological laboratories, particularly those of Harvard, Clark, and Chicago universities. The biologists and physiologists have continued their researches by this method, so that a very large amount of experimental work is now being done in comparative psychology. Despite the obvious advantages of experiment as a method for the study of animal behavior, it is not without

its dangers. These were clearly stated by Wesley Mills in a criticism of Thorndike’s ‘‘ Animal Intelligence” (492). They may be summed up by saying that there is a risk of placing the animal experimented upon under abnormal conditions in the attempt to make them definite and controllable.1 Did not, for example, the extreme hunger to which Thorndike’s cats and dogs were reduced, while it simplified the conditions in one sense by making the strength of the motive to escape as nearly as possible equal for all the animals, complicate matters in another sense by diminishing their capacity to learn? Were the animals perhaps frightened and distracted by the unusual character of their surroundings? Thorndike thinks not (707); but whether or no he succeeded in averting these dangers, it is clear that they are real. It is also obvious that they are the more threatening, the higher the animal with which one has to deal. Fright, bewilderment, loneliness, are conditions more apt to be met with among the higher vertebrates than lower down in the scale, and the utmost care should be taken to make sure that animals likely to be affected by them are thoroughly trained and at home in their surroundings before the experimenter records results.

The ideal method for the study of a higher animal involves patient observation upon a specimen known from birth, watched in its ordinary behavior and environment, and occasionally experimented upon with proper control of the conditions and without frightening it or otherwise rendering it abnormal. The observer should acquaint himself with the individual peculiarities of each animal studied, for there is no doubt that striking differences in mental capacity occur among the individuals of a single species. At the same time that he obtains the confidence of each individual animal, he should be able to hold in check the tendency to humanize it and to take a personal pleasure in its achievements if it be unusually endowed. This is, to say the least, not easy. Absolute indifference to the animals studied, if not so dangerous as doting affection, is yet to be avoided.

We may now turn from the problem of discovering the facts about animal behavior to the problem of interpreting them. If an animal behaves in a certain manner, what may we conclude the consciousness accompanying its behavior to be like? As we have seen, the interpretation is often confused with the observation, especially in the making of anecdotes; but theoretically the two problems are distinct. And at the outset of our discussion of the former, we are obliged to acknowledge that all psychic interpretation of animal behavior must be on the analogy of human experience. We do not know the meaning of such terms as perception, pleasure, fear, anger, visual sensation, etc., except as these processes form a part of the contents of our own minds. Whether we will or no, we must be anthropomorphic in the notions we form of what takes place in the mind of an animal. Accepting this fundamental proposition, the students of animals have yet differed widely in the conclusions they have drawn from it. Some have gone to the extreme of declaring that comparative psychology is therefore impossible. Others have joyfully hastened to make animals as human

as they could. Still others have occupied an intermediate position. Descartes and Montaigne are the two writers antedating the modern period who are most frequently quoted in this connection. The latter had evidently a natural sympathy with animals. In that most delightful twelfth chapter of the second book of Essays, “An Apology of Raymond Sebonde,”’ he gives free rein to the inclination to humanize them. I quote Florio’s translation: ‘“‘The Swallowes which at the approach of spring time we see to pry, to search and ferret all the corners of our houses; is it without judgment they seeke, or without discretion they chuse from out a thousand places, that which is fittest for them, to build their nests and lodging? . . . Would they (suppose you) first take water and then clay, unlesse they guessed that the hardnesse of the one is softned by the moistness of the other? ... Why doth the spider spin her artificiall web thicke in one place and thin in another? And now useth one, and then another knot, except she had an imaginary kind of deliberation, forethought, and conclusion?” To ascribe such behavior to the working of mere instinct, ‘‘with a kinde of unknowne, naturall and servile inclination,” is unreasonable. ‘‘The Fox, which the inhabitants of Thrace use” to test the ice on a river before crossing, which listens to the roaring of the water underneath and so judges whether the ice is safe or not; “might not we lawfully judge that the same discourse possesseth her head as in like case it would ours? And that it is a kinde of debating reason and consequence, drawne from natural sense? ‘Whatsoever maketh a noyse moveth, whatsoever moveth, is not frozen, whatsoever is not frozen, is liquid; whatsoever is liquid, yeelds under any weight ?’”

Descartes, on the other hand, writing some sixty years later, takes, as is well known, the opposite ground. He says in a letter to the Marquis of Newcastle, ‘‘As for the understanding or thought attributed by Montaigne and others to brutes, I cannot hold their opinion.” While animals surpass us in certain actions, it is, he holds, only in those ‘‘which are not directed by thought.... They act by force of nature and by springs, like a clock, which tells better what the hour is than our judgment can inform us. And doubtless when swallows come in the spring, they act in that like clocks. All that honey bees do is of the same nature” (183, pp. 281-283). The statement of Descartes, contained in the letter to Mersenne of July 30, 1640, that animals are automata, is often misunderstood. Descartes does not assert that animals are unconscious in the sense which that term would carry to-day, but only that they are without thought. Sensations, feelings, passions, he is willing to ascribe to them, in so far as these do not involve thought, “It must however be observed that I speak of thought, not of life, nor of sensation,” he says in the letter to Henry More, 1649; “I do not refuse to them feeling . . . in so far as it depends only on the bodily organs”’ (183, p. 287). In this he does not go so far as some modern writers, who decline to assert the presence of any psychic process in the lower forms of animal life.

Turning to recent times, we find arguments very like those of Montaigne used by the earlier evolutionary writers. Darwin, for instance, says in ‘The Descent of Man,” “As dogs, cats, horses, and probably all the higher animals, even birds, have vivid dreams, and this is shown by their movements and the sounds uttered, we must admit that they possess some power of imagination” (169, p. 74). “Even aw 2, Umatinet brute beasts,” says Montaigne, ‘“. . . are seen to be subject to the power of imagination; witnesse some Dogs. . . whom we ordinarily see to startle and barke in their sleep” (sor, Bk. I, ch. 20). “Only a few persons,” Darwin continues, ‘‘now dispute that animals possess some power of reasoning. Animals may constantly be seen to pause, deliberate, and resolve.” And he states that his object in the third chapter of the work quoted is ‘‘to show that there is no fundamental difference between man and the higher mammals in their mental faculties” (169, p. 66). Romanes is evidently guided by the same desire to humanize animals.

Now these writers were not led to take such an attitude merely out of general sympathy with the brute creation, like Montaigne; they had an ulterior motive; namely, to meet the objection raised in their time against the doctrine of evolution, based on the supposed fact of a great mental and moral gulf between man and the lower animals. They wished to show, as Darwin clearly states, that this gulf is not absolute but may conceivably have been bridged by intermediate stages of mental and moral development. While this argument against evolution was being pressed, the evolutionary writers were very unsafe guides in the field of animal psychology, for they distinctly “‘held a brief for animal intelligence,” to use Thorndike’s phrase. In more recent times interest in both the positive and the negative sides of the objection drawn from man’s superiority has died out, and such special

: Fee: has become unnecessary. On the other hand, the fact that the greater part of the y* experiments on animals were until the last ten or fifteen years performed by physiologists has given rise to an opposite tendency in interpreting the animal mind: the tendency to make purely biological concepts suffice as far as possible for the explanation of animal behavior and to assume the presence even of consciousness in animals only when it is absolutely necessary to do so. Loeb in 1890 suggested the theory which he has since elaborated, that the responses of animals to stimulation, instead of being signs of “sensation,” are in évery way analogous to the reactions of plants to such forces'as light and gravity ; hence unconscious “tropisms” (421). Bethe in 1898 attempted to explain all the complicated behavior of ants and bees, which the humanizing writers had compared with our own civilization, as a result of reflex responses, chiefly to chemical stimulation, unaccompanied by any consciousness whatever (51). This revival, in an altered form, of the Cartesian doctrine has met with energetic opposition, especially from writers having philosophical interests. When the first edition of the present work appeared, the parties in the controversy could be divided into three groups: those who believed that consciousness should be ascribed to all animals; those who believed that it should be ascribed only to those animals whose behavior presents certain peculiarities regarded as evidence of mind; and those who held that we have no trustworthy evidence of mind in any animal, and should therefore abandon comparative psychology and use only physiological terms. Of recent years, the tendency has been towards the survival only of the two extreme parties: it has been more and more recognized that there exists no evidence of mind which is not either equally bad or equally good in the case of all animals.

Among the authorities who would ascribe mind to all animals belong Claparéde of Geneva, the Swiss naturalist Forel, and the Jesuit Wasmann. They maintain this position from widely different philosophical points of view. The first-named is what is called a parallelist; that is, he believes that mental processes and bodily processes are not causally related, but form two parallel and non-interfering series of events. In the study of animals, both the physical and the psychical series should, he thinks, be investigated. Biology should use two parallel methods, the one ascending, attempting to explain animal behavior by physical and chemical laws; the other descending, giving an account of the mental processes of animals. Ultimately, it may be hoped, according'to Claparéde, that both methods will be applied throughout the whole range of animal life. At present the ascending method is most successful with the lowest forms, the descending method with the highest forms. We cannot afford to abandon the psychological study of animals, for our knowledge of the nervous processes underlying the higher mental activities is very slight; physiology here fails us, and psychology must be left in command of the field. The danger besetting the attempt at a purely physical explanation of animal behavior is that the facts shall be unduly simplified to fit the theory. Thus Bethe’s effort at explaining the way in which bees find their way back to the hive as a reflex response, or tropism, produced by ‘‘an unknown force,” is highly questionable; the facts - seem to point toward the exercise of some sort of memory by the bees. Itis always possible, further, that the tropism is accompanied by consciousness. A physiologist from Saturn might reduce all human activities to tropisms, says Claparéde in a striking passage. ‘‘The youth who feels himself drawn to medical studies, or he.

who is attracted to botany, can no more account for his profoundest aspirations than the beetle which runs to the odor of a dead animal or the butterfly invited by the flowers; and if the first shows a certain feeling corresponding to these secret states of the organism (a feeling of ‘predilection’ for such a career, etc.), how can we dare to deny to the second analogous states of consciousness?” (122). If it is argued that we have no direct, but only an inferential, knowledge of the processes in an animal’s mind, the argument is equally valid against human psychology, for the psychologist has only an inferential knowledge of his neighbor’s mind (124).

Wasmann defends the animal mind from a different position. For one thing, he believes that mental processes may act causally upon bodily states: He accepts, in other words, what is called interactionism, as opposed to parallelism. Further, although he strongly opposes the doctrine that the reactions of animals are unconscious tropisms and constantly emphasizes their variability and modifiability through experience, he nevertheless believes that a gulf separates the human from the animal mind. The term. ‘intelligence’? which most writers use to designate merely the power of learning by individual experience, Wasmann ~ would reserve for the power of deducing and understanding relations, and would assign only to human beings (761, 762). Although animals have their instincts modified by sense experience, man ‘‘stands through his reason and freedom immeasurably high above the irrational animal that follows, and must follow, its sensuous impulse without deliberation”’ (763).

Forel, in the third place, is what is called a monist in metaphysics. That is, he does not believe either that mind and body are parallel, or that they interact causally, but that they are two aspects of the same reality. ‘‘Every psychic phenomenon is the same real thing as the molecular or neurocymic activity of the brain-cortex coinciding with ae ON it” (233, p. 7). The psychic and the physical, on this theory, should be coextensive; not merely should consciousness in some form belong to all living things, but every atom of matter should have its psychic aspect. On such a basis, Forel takes highly optimistic views of the animal mind. In insects, of which he has made a special study, it is, he thinks, ‘‘possible to demonstrate the existence of memory, associations of sensory images, perceptions, attention, habits, simple powers of inference from analogy, the utilization of individual experience, and hence distinct, though feeble, plastic individual deliberations or adaptations” (233, p. 36).

A peculiar position on the problem of mind in animals is occupied by the “‘vitalists,” of whom Driesch (191) is the foremost representative. They regard the reactions of organisms as requiring the operation of psychic forces or “entelechies’”’; they hold that as physical phenomena such reactions cannot be explained save through the working of these psychic forces. A living being is forever distinguished from a lifeless creature by the presence of such entelechies. Thus the vitalist is an interactionist and a dualist: the worlds of the lifeless and the living are to him forever distinct.

The opposite camp is represented by Bethe, Beer, von Uexkiill, Loeb, and other physiologists, as well as by Watson. The eminent neurologist Bethe, in his study of the behavior of ants and bees, refuses to allow these animals any ‘psychic qualities” whatever, and suggests the term . “chemo-reception” instead of “‘smell,” to designate the influence which directs most of their reactions, — ‘‘smell”’ implying a psychic quality (51). In a footnote to a later article he says: ‘‘ Psychic qualities cannot be demonstrated.

the capacity of modifying our behavior [i.e. of learning], and every one knows from his own experience that psychic qualities play a part connected with this modifying process. Every statement that another being possesses psychic qualities is a conclusion from analogy, not a certainty; it is a matter of faith. If one wishes to draw this analogical - inference, it should be made where the capacity for modification can be shown. When this is lacking, there is not the slightest scientific justification for assuming psychic qualities. They may exist, but there is no probability of it, and hence science should deny them. Hence if one ventures to speak of a Psych@ in.animals at all, one should give the preference to those which can modify their behavior” (51). But that Bethe himself prefers not to make the venture is evident from statements in the text of the same article. The psychic or subjective, he says, is unknowable, and the only thing we may hope to know anything about is the chemical and physiological processes involved. ‘‘These chemo-physical processes and their consequences, that is, the objective aspect of psychic phenomena, and these alone, should be the object of scientific investigation” (51).

Together with Beer and von Uexkiill, Bethe shortly afterward published ‘Proposals for an Objectifying Nomenclature in the Physiology of the Nervous System.” The main purpose of this paper was to suggest that all terms having a psychological implication, such as sight, smell, sense-organ, memory, learning, and the like, be carefully excluded from discussions of animal reactions tq,stimula- .tion and animal behavior generally. In their stead the responses to stimulation where no nervous system exists, the term antitypes; for those involving a nervous system, antikineses; the latter are divided into reflexes, where the response is uniform, and antzklises, where the response is modifiable. A sense-organ | becomes a reception-organ, sensory nerves are receptory-nerves, and we have phonoreception, stibo-reception, photo-reception, instead of hearing, smell, and sight. The after-effect of a stimulus upon later ones is the resonance of the stimulus (39).

Loeb (434) agrees with Bethe that physico-chemical processes and not states of consciousness are the proper objects of investigation for the “psychologist.” These men evidently regard the universe as essentially uniform throughout — there exists for them no gulf between living and lifeless things; the behavior of living beings will be reduced to a series of chemical reactions as soon as science has progressed sufficiently far. They are “mechanists.” It is, however, perfectly possible to be a mechanist so far as the explanation of animal behavior is concerned, and still admit that animals have consciousness and that their behavior is accompanied by inner, mental states which it is the business of the psychologist to investigate. One does not have to be a vitalist to believe that animals have minds: one may hold that every action of an animal will gome day be explained as the result of physico-chemical processes, and yet maintain that the actions of animals are conscious. The consciousness would be an accompanjr. ment, an inner aspect, of the physico-chemical processes.

The views of Loeb and Bethe have gained much ground lately among certain American psychologists, notably Watson (771). The position of these ‘“behaviorists” seems not to have been fully thought out in its philosophical aspects, but is somewhat as follows. The difficulties of interpreting an animal’s mind from its behavior are so great that such inferences have no scientific value. We may therefore proceed as if animals had no minds; or rather, as if mind were a kind of behavior, observable by outside means. Since it is obvious that the difficulty of interpreting an animal’s mind from its behavior is only greater in degree than, not unliké in kind, the difficulty of interpreting other humah minds from behavior, human psychology also should confine itself to the observation merely of the actions of other persons, and permit no inferences as to the inner aspect of such actions. In fact, there is no inner aspect to such actions— thoughts and feelings, human as well as animal, are only behavior, and if we have at present no instruments for inspecting and measuring the movements which are thoughts and feelings, such instruments will in time be discovered.

In opposition to these views, we shall in this book maintain the following position. There exists an inner aspect to behavior, the realm of sensations, feelings, and thoughts, which is not itself identical with behavior or-with any form | of movement. Thoughts probably always have as their “accompaniment bodily movements, but the thought is not identical with the movement. If a physiologist perfected an instrument by which he could observe the nervous process in my cortex that occurs when I am conscious of the sensation red, he would see nothing red about it; if he could watch the bodily movements that result from this stimulation, say, for instance, the slight contraction of the articulatory muscles that occurs when I say “‘red” to myself, he would not see them as red. The red is in my consciousness, and no devices for observing and registering my movements will ever observe the red, though they may easily lead to the inference that it exists in my con-

sciousness. And precisely the same is true of all my sensations, thoughts, and feelings. Since an inner world of experience exists, we may legitimately try to investigate it. For this purpose we possess a method, which is called introspection. We can, that is, attentively and, if we have had practice, dispassionately and scientifically, observe what goes on in our own consciousness when we receive certain stimuli and make certain movements. Further, we can, by the use of the same kind of inference from one case to another similar case, upon which all scientific generalization is based, infer that

“when a being whose structure resembles ours receives the same stimulus that affects us and moves in the same way as a result, he has an inner experience which resembles our own. Finally, we may extend this inference to the lower animals, with proper safeguards, just as far as they present resemblances in structure and behavior to ourselves. Our object in this book will always be the interpretation of the inner aspect of the behavior of animals; we shall be interested in what animals do only as it throws light upon what they feel. To the true psychologist, no challenge is so enticing as that presented by the problem of how it feels to be another person or another animal; and although we must sometimes give up the problem in despair, yet we have also our successes. We have wonderfully advanced, within the last twenty-five years, in knowledge as to how the world looks from the point of view of our brother animals.

We may now note briefly some of the special precautions that must be observed in interpreting the conscious aspect of animal behavior], First, there is no doubt that great caution should be used in regarding the quality of a human conscious process as identical with the quality of the corresponding process in the animal mind. For example, we might say with a fair degree of assurance that an animal consciously discriminates between light and darkness; that is, receives conscious impressions of different quality from the two, yet the mental impression produced by white light upon the animal may be very different from the sensation of white as we know it, and the impression produced by the absence of light very different from our sensation of black. Black and white may, for all we know, depend for their quality upon some substance existing only in the human retina.

“\| Asecond precaution concerns the simplicity or complexity ‘of the interpretation put upon animal behavior. Lloyd Morgan, in his ‘‘Introduction to Comparative Psychology,” formulated a conservative principle of interpretation which has often been quoted as ‘‘Lloyd Morgan’s Canon.” The principle is as follows: “In no 0 Case may we we interpret an action as the outcome of the exercise of a higher psychical faculty, if it can be interpreted as the outcome of the exercise of one which stands lower in the psychological scale” (505, P- 53)»“In other words, when in doubt take the simpler interpretation. “For example, a dog detected in a theft cowers and whines. One possible mental accompaniment of this behavior is remorse; the dog is conscious that he has fallen below a moral standard, and grieved or offended his master. A second is the anticipation of punishment; the dog has a mental representation of the consequences of his action upon former occasions, and imagining himself likely to experience them anew, is terrified at the prospect. A third possibility is that the dog’s previous experience of punishment, instead of being revived in the form of definite images, makes itself effective merely in his feelings and behavior; he is uncomfortable and fright-

ened, he knows not definitely why. It is evident that these three possibilities represent three different grades of complexity of mental process, the first being by far the highest. Lloyd Morgan’s canon enjoins upon us in such a case to prefer the third alternative, provided that it will really account for the dog’s behavior. Now why should the simplest interpretation be preferred? We must not forget that the more complex ones remain in the field of possibility. Dogmatic assertions have no place in comparative psychology. We cannot say that the simplicity of an hypothesis is sufficient warrant of its truth, for nature does not always proceed by the paths which seem to us least complicated. The fact is that Lloyd Morgan’s principle serves to counterbalance our most important source of error in interpreting animal behavior. It is like tipping a boat in one direction to compensate for the fact that some one is pulling the opposite - gunwale. We must interpret the animal mind humanly if we are to interpret it at all. Yet we know that it differs from the human mind, and that the difference is partly a matter of complexity. Let us therefore take the least complex interpretation that the facts of animal behavior will admit, always remembering that we may be wrong in so doing, but resting assured that we are, upon the whole, on the safer side. The social consciousness of man is very strong,.and his tendency to think of other creatures, even of inanimate nature, as’sharing his own thoughts and feelings, has shown itself in his past to be almost irresistible. Lloyd Morgan’s canon offers the best safeguard against this natural inclination, short of abandoning all attempt to study the mental life of the lower animals.

In this chapter we shall try to show that there exists no evidence for denying mind to any animals, if we do not deny it to all; in other words, that there is no such thing as an objective proof of the presence of mind, whose absence may be-regarded as proof of the absence of mind. To begin with, can it be said that when an animal makes a movement in response to a certain stimulus, there is an accompanying consciousness of the stimulus, and that when

it fails to move, there is no consciousness? Is response to (i stimulation evidence of consciousness? In the case of man, we know that absence of visible response does not prove that the stimulus has not been sensed; while it is probable that some effect upon motor channels always occurs when consciousness accompanies stimulation, the effect may not be apparent to an outside observer. On the other hand, if movement in response to the impact of a physical force is evidence of consciousness, then the ball which falls under the influence of gravity and rebounds on striking the floor is conscious. Nor is the case improved if we point out that the movements which animals make in response to stimulation are not the equivalent in energy of the stimulus applied, but involve the setting free of energy stored in the animal aswell. True, when a microscopic animal meets an obstacle in its swimming, and darts backward, the movement is

not a mere rebound; it implies energy contributed by the animal’s own body. But just so an explosion of gunpowder is not the equivalent in energy. of the heat of the match, the stimulus. Similarly it is possible to think of the response made by animals to external stimuli as involving nothing more than certain physical and chemical processes identical with those existing in inanimate nature. result of external stimulation regularly involve withdrawal ‘ from certain stimuli and acceptance of others, it is natural

to use the term ‘‘choice”’ in describing such behavior. But if consciousness is supposed to accompany the exercise of choice in this sense, then consciousness must be assumed to accompany the behavior of atoms in chemical combinations. When hydrochloric acid is added to a solution of silver nitrate, the atoms of chlorine and those of silver find each other by an unerring “‘instinct’”’ and combine into the white precipitate of silver chloride, while the hydrogen and the nitric acid similarly ‘‘choose” each other. Nor can the fact that behavior in animals is adapted to an end be used as evidence of mind; for ‘‘purposive”’ reactions, whfch contribute to the welfare of an organism, are themselves selective. The search for food, the care for the young, and the complex activities which further welfare, are made up of reactions involving “choice” between stimuli; and if the simple “choice” reaction is on a par with the behavior of chemical atoms, so far as proof of consciousness goes, then adaptation to an end, apparent purposiveness, is in a similar position.