§ 1. What Is Behaviour? § 2. Diverse Views as to Animal Behaviour. § 3. Activities of Unicellular Organisms. § 4. Special Case of Shell-building among Arenaceous Foraminifera. § 5. Reflex Actions. § 6. Tropisms. § 7. Non-intelligent Experimentation. § 8. Instinctive Behaviour. § 9. Theories of Instinct. §10. Evidence of Intelligent Behaviour. §11. Secondary Simplifications of Behaviour. § 12. Rational Conduct. § 13. General Impressions of Animal Behaviour.
IN our preceding studies we reached the conclusion that a matter-and-motion description of living creatures is far from being adequate, that it does not grip the kernel. While organisms are collocations of matter and energy, there has welled up within them a new aspect of reality which demands other than mechanical, chemical, and physical concepts. Wherein the newness precisely consists we have not discovered; but we recognise the living creature as a historic being which has enregistered the experiences and experiments of the past, which trades to good purpose with time, which has its conative bow often bent. In the hope of a further appreciation of the significance of life we turn now to a more systematic consideration of animal behaviour.
§ 1. What Is Behaviour?
The movements of the planets in their courses afford an object-lesson of orderliness on a grand scale, and yet the least in the realm of organisms is greater than theyin being an agent. The movements of a rolling stone are completely determined by its momentum and the circumstances, but there is a spice of unpredictability in the ways of most living creatures. The unexpected often happens. There are indeed uniformities of sequence in the reactions of organisms, otherwise no science of behaviour were possible, but there is an undeniable appearance of free agency. It is interesting to watch under the microscope the Brownian movement seen when minute granules of sepia or the like are jostled hither and thither, probably by the invisible ions which abut against them, but the scene changes in character when we put in a vigorous unicellular organism. It does not take charge like a gun torn from its attachment on board ship; it commands its course. Only in the realm of organisms is there true behaviourin which the creature is an agent and exhibits a correlated or concatenated series of acts, effective towards some definite result, favourable to the continuance and harmony of vital processes.
There are two master-activities in the animal organismfor the sake of which life is worth livingmovement and feeling, contractility and irritability, the functions in most cases of the muscular and nervous systems respectively. These master-activities are kept agoing, the relevant structures are kept in working order, by the other everyday functions of nutrition, circulation, respiration, excretion, and so on; never forgetting, in connection with Vertebrates at least, the fundamental trigger-pulling and regulative function of the organs of internal secretion. These everyday functions are the pre-conditions of behaviour; and growth and maturity may also condition behaviour. But behaviour itself is much more, it means that the organism is an agent and that it exhibits a correlated or concatenated series of actions.
§ 2. Diverse Views as to Animal Behaviour.
The difficulty of rightly interpreting the observed behaviour of animals is confessedly great. It is not easy for us to get mentally near them. In many cases the structure of their body in general and of their nervous system in particular is very different from ours; their sense-organs are often on another plan, and there are some whose functions we do not yet know; such words as a few of the higher animals have, we can only vaguely understand. How are we to get into mental contact with ants and bees? And even for more accessible animals, like horse and dog, there is a great difficulty involved in the simple fact that all our psychological terms are saturated with human meaning.
Some investigators have found a short and easy way out of difficulties by dogmatically declaring that there is no more mind among animals than there is among plants, and that the sensible course is to keep to physiological description. If that suffices for giving an account of the bryony climbing up the hedge with its exquisitely tactile and adaptively motile tendrils, will it not serve for the sea-urchin climbing up the rock, the squirrel climbing up the tree? This is the extreme of over-simplicity. It was indeed a wise saying of Spinoza:No one has yet learned from experience what the body regarded purely as a body is able to do in accordance with its own natural laws, or what it cannot do, but it seems to most naturalists to make the behaviour of higher animals magical if we do not credit them with an awareness and pre-awareness of meaning.
There are others who think that we get nearer the truth the more anthropomorphic we are, who believe that the behaviour of all animals shows evidence of mind. That is to say, the description that we give of an animal's behaviour, or of critical corners in it at least, is bound to be inadequate unless we use psychical terms. This is the other extreme. It was expressed by Hume when he said: No truth appears to be more evident than that beasts are endowed with thought and reason as well as men. That may be tenable generosity for horse and dog, but it cannot hold good for starfish and earthworm.
How are we to avoid the stern over-parsimony of Descartes on the one hand, and the delicious over-generosity of Montaigne on the other? We must not give a false simplicity to the facts by reducing the animal to the level of an automatic machine, but we must not read the man into the beast without critical hesitation. The hive-bees that make the honeycomb so symmetrically are not automatic machines, but neither are they little geometricians.
To keep to the via media of good sense must always be difficult, for the assumption of mind in an animal or of a psychological aspect in the behaviour of an animal cannot be demonstrated. There is no litmus paper for mentality. Every statement, says Bethe, that another being possesses psychic qualities is a conclusion from analogy, not a certainty; it is a matter of faith. Our assumption of mind in our fellow-men rests on the same sort of basis (though with inter-subjective corroborations); it is a necessary hypothesis and one that works. Is not a similar hypothesis indispensable in regard to animals if we are to understand them and make the most of them? But animal behaviour has such a long gamut that each case must be judged on its own merits. We ask in each case whether we can make sense of what we see without assuming mental factors, whether we can adequately describe what we see without using psychological terms. We inquire into the creature's power of profiting by experience, as to its pursuance of trial-and-error methods. The growth of experimental psychology has furnished many a welcome check to interpretations, showing some to be too simple and others to be too generous. Some weight will also be attached to the degree in which the nervous system of the animal in question resembles ours, or that of some types previously judged to be capable of, let us say, making an inference.
The sound practical rule is to try to re-describe the observed behaviour in as simple terms as possible without leaving out any essential feature. As Prof. Lloyd Morgan has put it, In no case may 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 (1894, p. 53). The simplest description is not necessarily the true one, as we know in human affairs; but the scientific method is to hold to it till facts force us to give it up.
§ 3. Activities of Unicellular Organisms.
Many unicellular organisms show restless movements which may be ranked at the foot of the inclined plane of behaviour. As long as certain combustible materials within the organism hold out, and as long as certain external stimuli continue to act, the creatures continue moving; and we see them responding to various calls which summon them now to one side and now to the other. This is like the unworked ore of behaviour, and not much beyond the level of everyday internal activities. We may speak of it as general organismal activity among unicellulars.
But even among the simplest creatures we notice three facts:(1) The movements of one animalcule are often quite different from those of another, even in the same medium: so soon is the note of individuality struck. (2) The restless roving is not at random, and when food is scarce it is intensified, passing insensibly into hunting. (3) Many of the simplest animals exhibit quite definite reactions to stimuli. They respond by particular movements to changes in temperature, in illumination, and in the chemical composition of the medium. As there is no nervous system, but simply a specific inborn protoplasmic organisation, we may use the phrase unicellular organic reaction for what is in a far-off way comparable to the reflex action of a higher animal.
The adherents of the mechanistic school point out that the Amba is a reservoir of energy which is tapped by various stimulisuch as the freshness of the water, the chemical composition, the temperature, and the illumination. Its locomotion represents the natural outflow of the stored energy, and the direction of its journeying depends on the continually varying stimuli with which it meets. But when we really study the Amba, as Professor Jennings (1906) has done, the possibility of any simple description of its doings disappears. Its general condition has much to do with its reactions; the direction of movement is not wholly determined by the position of the stimulus or the part of the body on which it acts; the moving Amba shows in its transient differentiation a trafficking with time; what it has done is an important factor in determining what it will do; the types of reaction are not stereotyped; it is not possible to predict the movements from a knowledge of the direct results of the external influence. So the life of the Amba is not such a simple affair as some would make out. We require a kind of description different from that which suffices for the potassium pill rushing about on the surface of the water.
The Amba encounters a hurtful stimulating influence affecting part of the cell; it withdraws the stimulated part, and that is related to the localisation of the influence. But it proceeds to send forth a finger-like process of its living matter in a new direction, and the issue of this is determined by internal conditions. If the new direction of movement leads to further stimulation, a new trial is made. Such trials are repeated till either there is no further stimulation, or if it is not possible to escape completely, until the stimulation falls on the posterior end, and the animal is retracted directly from the source of stimulation (Jennings, 1906, p. 22). The importance of this is great. A direction is taken because it relieves the Amba from hurtful stimulation. There is, Jennings says, selection from among the conditions produced by varied movements. Thus the behaviour of Amba is directly adaptive; it tends to preserve the life of the animal and to aid it in carrying on its normal activities (p. 23).
The writer is thoroughly convinced, after long study of the behaviour of this organism, that if Amba were a large animal, so as to come within the everyday experience of human beings, its behaviour would at once call forth the attribution to it of states of pleasure and pain, of hunger, desire, and the like, on precisely the same basis as we attribute these things to the dog. This natural recognition is exactly what Münsterberg (Grundzüge der Psychologie, Bd. I., 1900) has emphasised as the test of a subject. In conducting objective investigations we train ourselves to suppress this impression, but thorough investigation tends to restore it stronger than at first (p. 336).
Miss Washburn (1909, p. 41) has inquired into the nature of the Amba's mind, if haply it has one. There cannot be more than three or four qualitatively different elements in its experience; there is no evidence of memory images; it probably has not more than flashes of consciousness. The problem of the Amba's mind can wait, but it seems to us clear that when we allow as much as possible to physical properties (such as we see in the improvement of a violin in the hands of a master), to the response of chemical bodies to certain stimuli and no others (such as we see in photography), to the purely physiological registration of experience (such as we know in the improvement of well-exercised muscles), there is a new aspect of reality appearing in the behaviour of even an Amba.
The slipper-animalcule, Paramecium, abundant in water with decaying marsh plants in it, is a minute, cigar-shaped, ciliated Infusorian, just visible to the naked eye as an elongated whitish particle. Its rudimentary but very effective behaviour has been much studied, especially by Prof. H. S. Jennings. One of the commonest episodes is that in its swimming the Paramecium meets with something injurious in the water, and exhibits what is called the avoiding reaction. It reverses the action of its cilia and swims away from the stimulus; at a certain distance it moves so as to swing its anterior end in a circle, testing the water in different directions; when the sample from a certain direction no longer contains the obnoxious influence, the Paramecium goes ahead again in that direction, and may have a free course till the next stimulus is experienced. When the original stimulus is due to some mechanical obstacle, Paramecium can get no hint from testing the water; it tries going ahead in various directions, till it finds one in which there is no further obstacle to progress. In this direction it continues. Through systematically testing the surroundings, by swinging the anterior end in a circle, and through performing the entire reaction repeatedly, the infusorian is bound in time to find any existing egress from the difficulties even though it be but a narrow and tortuous passageway (Jennings, 1906, p. 49).
The behaviour of Paramecium is very instructive in its combination of effectiveness and simplicity. It drives itself forward in a narrow spiral, revolving on its long axis and swerving a little towards the aboral sidesuch is its action system; and most of its behaviour consists of slight variations on this simple tune. It constantly feels its way about, trying in a systematic way all sorts of conditions, and retiring from those that are harmful. Its behaviour is in principle much like that of a blind and deaf person, or one that feels his way about in the dark. It is a continual process of proving all things and holding to that which is good (Jennings, 1906, p. 106).
The behaviour and reactions of Paramecium consist on the whole in performing movements which subject the organism to varied conditions (using this word in the widest sense), with rejection of certain of these conditions, and retention of others. It may be characterized briefly as a selection from among the varied conditions brought about by varied movements (Jennings, 1906, p. 108). On the whole the animalcule rejects or avoids what is injurious and accepts or seeks what is beneficial, just as higher animals and men do. The behaviour is adaptive and purposive. Indeed Jennings goes the length of saying: In no other group of organisms does the method of trial and error so completely dominate behaviour, perhaps, as in the Infusoria.
A fact of great interest has been established by Professor Jennings, that the behaviour of unicellular organisms is modifiable by experience. He has experimented, for instance, with a trumpet-shaped ciliated Infusorian called Stentor which abounds in marshy pools, attaching itself by the narrow end to a water-weed, and surrounding the lower half of its body with a mucus-like sheath, the so-called tube. A cloud of carmine particles is introduced into the water-currents passing to the ciliated mouth of the Stentor. It bends to the aboral side, twisting on its stalk two or three times as it bends, and thus often avoids the cloud of particles. That is answer one. But if the particles continue to come, the ciliary movement is suddenly reversed and the water is driven away from the mouth. This may be repeated two or three times, and is answer two. If the Stentor does not get rid of the obnoxious stimulation in either of these two ways, it contracts into its tube and suspends activity, this being answer three. After half a minute or so it re-expands, and if the carmine particles still reach it, it contracts again. It will do this many times, and after each contraction it stays a little longer in its tube than it did before. Finally, if no improvement in circumstances rewards its trials, it breaks its attachment and swims forwards or backwards away from its tube. And this is answer four. The stimulus and other external conditions remaining the same, the organism responds by a series of reactions becoming of more and more pronounced character, until by one of them it rids itself of the stimulation (p. 176). The same individual does not always behave in the same way under the same external conditions, but the behaviour depends upon the physiological condition of the animal. The reaction to any given stimulus is modified by the past experience of the animal, and the modifications are regulatory, not haphazard, in character. The phenomena are thus similar to those shown in the learning of higher organisms, save that the modifications depend upon less complex relations and last a shorter time (p. 179). Our view of living creatures must make room for the new fact that behaviour reaches this level among the unicellulars.
Among unicellulars, then, we see the beginning of exploring and testing, the beginning of actual hunting, and this is on the main line of advance. We see, also, the enregistration of particular reactions to stimuli, an organisation of behaviour that is economical and in its rapidity of response often life-saving. But we see, also, a selection of reactions, a trying of one after the other till haply one meets the needs of the case, and this trial-and-error method is likewise on the main line of advance.
§ 4. Special Case of Shell-building among Arenaceous Foraminifera.
In the shell-building of some of the so-called arenaceous Foraminifera Mr. E. A. Heron-Allen and Mr. A. Earland have described (1915) what looks like constructive skill in the use of materials. As every one knows, many unicellular animals secrete shells of exquisite beauty, the organic crystallisation of which is as much of an unsolved problem as the adaptive internal architecture of bones, but in the case of the arenaceous Foraminifera the building materials are found ready-made in the environment and are utilised very effectively to form a casing. The points of special interest are two,(1) that a particular kind of material, such as sponge-spicules, is selected from the surrounding débris, from amid a multitude of apparent alternatives, and (2) that it is utilised in a fashion which is interpretable as peculiarly adaptive. The first point may be illustrated by the case of Technitella thompsoni, which covers itself with minute perforated Echinoderm platelets; the second point by the case of Marsipella spiralis, which arranges its encasement of sponge-spicules in a spiral, doubtless of considerable architectural value.
§ 5. Reflex Actions.
Among simple multicellular animals we find, as among the unicellulars, abundant illustrations of exploring, testing, and hunting. Perhaps we may recognise more staying power, persistence, and momentum, advantages naturally accruing from the acquisition of a body.
But the establishment of a nervous system opened the way to the organisation of reflex actions, which are the outcome of hereditarily prearranged linkages of nerve-cells and muscle-cells. These play an important rôle in behaviour. The sea-anemone's tentacles close upon their victim; the nestling's mouth opens at the touch of the food in its mother's beak; the earthworm withdraws into its burrow when it feels the tremor of a thrush's footstep; we cough in spite of ourselves when the crumb of bread is going the wrong way, and so on. These reflex actions are uniform reactions to a particular kind of external or internal stimulus; they are exhibited by all animals of the same kind in approximately the same way, though some individuals are quicker than others; they are independent of individual experience and do not require control on the part of the central nervous system; they depend on inborn structural linkages of particular sensory and particular motor nerve-units or neurons. In typical and simple cases, a reflex action involves (1) the receptor of a stimulusthe sensory or perceptory nerve-cell from which impulses pass in to the central nervous system; (2) a motor nerve-cell which connects the central nervous system with a muscle or a gland; and (3) between these two a communicating, or internuncial, or associative nerve-cell connecting them within the nervous system. The receptor neurone has its cell-body outside of the nerve-centre; the motor neurone, with its cell-body within the nerve-centre, sends a nerve-fibre to some peripheral effector organ; the associative neurone connects the two others. Thus is formed a reflex arc, the functional unit of the nervous system. In most cases the arrangements are more complex and several reflex arcs become interlinked. But the point is that reflex actions do not require individual correlation; that is pre-established. Yet it is important not to think of reflexes too simply.
Combination of Reflexes in Unified Behaviour. The perfection of reflexes is well illustrated in the behaviour of a sea-urchin, which has no nerve-ganglia. Its test is covered with mobile spines and snapping blades (pedicellariæ) which react in definite ways to definite stimuli and have an astonishing independence. For a single spine or pedicellaria on an isolated fragment of shell reacts very much as usual. In the uninjured creature the spines and other structures are all connected by a nervous network on the surface of the shell, and they act harmoniously, working into one another's hands, securing effective defence and locomotion. According to von Uexküll the sea-urchin is a republic of reflexes. The separate reflex arcs are so constituted and so put together that the simultaneous but independent course of the reflexes in response to an outer stimulus produces a definite general action, just as in animals in which a common centre produces the action. As von Uexküll puts it, when a dog runs, the animal moves its legs; when a sea-urchin crawls, the legs (spines) move the animal.
The astounding fact is the unification of behaviour that may occur in such a republic of reflexes. When a sea-urchin is placed upside down, a continuation of all the usual reactions would cause it to move on in an inverted position. But as von Uexküll has shown, the unusual physiological state induces a thoroughgoing change in the behaviour of the spines, they depart altogether from routine, and work adaptively to the needs of the organism as a whole, the animal being turned right again. This warns us not to think of reflexes woodenly; and if we need another warning we may get it in the extraordinarily subtle reflex by which a flatfish adjusts its coloration to that of the immediate environment of shingle.
Reflex Responses are Affected by the Physiological Condition of the Organism. In stinging animals, such as sea-anemones and medusæ, there are numerous reflex actions of an adaptive kind, concerned, for instance, with feeding. But it has been shown by Professor Jennings and others that the reaction is not simply a question of predetermined structure and an external stimulus. The answer depends on the relation of external conditions to internal processes. We cannot predict how an animal will react to a given condition unless we know the state of its internal physiological processes (Jennings, 1906, p. 231). Thus, to take a simple case, a sea-anemone cheated several times with false food ceases to exhibit the normal reflex. Many a sea-anemone, e.g., the large Stoichactis helianthus, will remove food from the oral disc if it is not hungry. A specimen of Metridium or Aiptasia will refuse to take bits of filter paper, though it will still take meat. After it has thus refused paper, two or three pieces of meat are given in succession, and taken readily. Now the bit of paper is placed again on the disc, and it too is swallowed. Clearly, the uninterrupted taking of a number of pieces of meat changes the physiological condition in some way, preparing the animal for the taking of any object with which it comes in contact. One cannot fail to note the parallelism with what occurs in higher animals under similar conditions (Jennings, 1906, p. 226). We see, then, that in relatively simple creatures, such as sea-anemones and starfishes, which have no nerve-ganglia, past stimuli and past reactions are important factors in determining present behaviour. Thus an elongated sea-anemone, Aiptasia annulata, which lives in crevices beneath and between stones, will bend into a new position if it is touched too often, and if it be molested still further will release its foothold and move to a new region (p. 206). In its natural surroundings it often has to cramp its body into quaint zigzag shapes, and a point of some interest is that this may become habitual and may persist for some time after the creature is removed to an unimpeded habitat. This illustrates what is meant by registration.
In his valuable study of the behaviour of the earthworm, Prof. H. S. Jennings shows that the response to a stimulus depends on external factors (such as the intensity and localisation of the stimulus), and on internal factors (such as the state of the animal at the time, its tendency to move in a certain way, e.g., head foremost, and the direction in which it was crawling at the time). But what is particularly interesting is the definite evidence that the behaviour at a given time depends on past experienceson former external conditions and on former actions.
Succession of Tentative Reflexes. Because of the abundance of reflexes in the simpler animals the impression has gained ground that behaviour in these lower reaches of life is very stereotyped. But this impression requires critical consideration. When a particular reflex action solves a particular problem at a stroke, there is no more to be done. But the problem is often more difficult, and what the creature does is to exhibit varied movements and to select certain resulting conditions. Even in such predominantly reflex creatures as sea-urchins, the tube-feet, spines, and pedicellariæ may in difficult situations continue in varied tentative movements, as if trying all expedients, and long after the original stimulation has ceased. The righting reaction of an inverted starfish is singularly varied and flexible. Professor Preyer repeatedly slipped a short india-rubber tube over one of the arms of a brittle star, and observed five different ways in which it was removed, including, it must be confessed, as one desperate method, the surrender of the arm itself. As the observer remarked, If one method does not help, another is used.
Professor Preyer's experiments on pegging down starfishes (of course without injuring them) revealed extraordinary flexibility of behaviour and also a shortening of the time required for escape. The number of useless movements, superfluous twistings, feelings about, and forward and backward motions, becomes less the oftener the individual has been placed in such a situation. If this is true (Prof. Jennings notes), we have in so low an animal as the starfish regulation through the selection of conditions produced by varied movements passing into a more directly regulatory action; in other words, what is commonly called in higher animals intelligence (Jennings, 1906, p. 241). In many cases we have to do with chain reflexes, one phase leading on to another, but the fact is that in most cases the succeeding phase is not invariably and irrevocably called up by the preceding one (Jennings, p. 251), the present action depending upon the entire physiological state of the organism, which, again, is determined by various factors.
One of the clearest results of the modern study of the behaviour of the lower animals is that the kind of action depends greatly on the physiological state as a whole, which, again, depends in part on history and experience. That experience can be enregistered in organisms with the simplest of nervous systems or with none is certain. Two individual Planarias (small ciliated fresh-water worms) often react in opposite ways to the same stimulus. What they do varies with their appetite, their freshness or fatigue, their recent stimulation and degree of excitement, and their history. After long study of Planaria, Professor Pearl concludes that it is almost an absolute necessity that a person should become familiar, or perhaps better, intimate, with an organism, so that he knows it in something the same way that he knows a person, before he can hope to get even an approximation of the truth regarding its behaviour (quoted by Jennings, 1906, p. 254).
If a reflex be an invariable reaction to a given stimulus, then there is much more than that in the behaviour of lower animals. For different answers to the same stimulus may be given by the same kind of creature or by the same creature at different times. The answer depends on the organismal condition as a whole. Moreover, the fact that stands out most clearly in the behaviour of the lower organisms is this:Each stimulus causes as a rule not merely a single definite action that may be called a reflex, but a series of trial movements, of the most diverse character, and including at times practically all the movements of which the animal is capable (Jennings, 1906, p. 280).
§ 6. Tropisms.
From chains of reflexes, suffused with awareness, it is not difficult to pass to the level of instinctive behaviour, but before we pass to that level we have to recognise the important rôle played by tropisms (see Loeb, 1918). Tropisms are obligatory or forced movements of the creature as a whole, which more or less automatically secure physiological equilibrium in relation to outside stimuli, such as light or heat, gravity or electricity, diffusing chemicals or water-currents. When a moth, constitutionally adapted to nocturnal activity, comes in its flight within the sphere of influence of a candle, and has one eye much more illumined than the other, owing to the direction in which it happens to be flying, more intense chemical processes are set up in the illumined eye. On that side there is therefore a relative increase in the mass of certain chemical products. But messages, impulses, stimulations, or waves of chemical reaction are always passing from the brain of the flying moth to the contracting muscles, and if the physiological symmetry of the brain has been disturbed by the unequal illumination of the eyes, the muscles on the more illumined side are thrown into a state of stronger tension or tonus, with the result that they will respond more forcibly to stimulation from the brain, and will therefore turn the head and body of the moth directly towards the candle near which it is flying. As soon as the plane of symmetry goes through the source of light, both eyes receive again equal illumination, the tension (or tonus) of symmetrical muscles becomes equal again, and the impulses for locomotion will now produce equal activity in the symmetrical muscles. As a consequence, the animal will move in a straight line to the source of light until some other asymmetrical disturbance once more changes the direction of motion (Loeb, 1918, p. 14). Such, in outline, is Prof. Jacques Loeb's ingenious and convincing theory of the tropism or forced movement which brings the moth into the candle.
Tropistic actions are obligatory in the sense that every creature of the same kind and in the same physiological state will in similar circumstances behave in the same way; there is no alternative. But it is a very notable fact, to be carefully thought over, that a tropism may be changed, reversed, or annulled by changes in the physiological condition of the body or by changes in the surrounding medium. The common Amphipod Crustacean Gammarus of fresh-water pools always moves away from lightthat is its tropism, but add the least trace of acid to the water, and it moves towards the lightas if a drop of philtre changed the creature's whole nature!
The tropistic movements often appear as if they had a very definite external aimsuch as the candlebut that is illusory. The orientation is physiologically coerced. There is no desire of the moth for the star. It should be noted that their general adaptiveness is not contradicted by cases like the moth flying into the candle, for organisms are not and could not be adapted to the altogether exceptional and unnatural. In some cases one tropism way thwart another, and it may be that a tropistic movement is sometimes interrupted by some strong internal stimulus such as a desire.
One of the criteria of organisms is the power of retention or registration which eventually finds remarkable expression in human memory, and the general view we wish to suggest, as a clue to the maze of animal behaviour, is that there has been at level after level a process of automatisation or organisation, which makes for economy of time and energy, and also, if it does not go too far, leaves the organism free for experiment and initiative. So in established reactions, in reflex actions, and in tropisms we see enregistrations which are, in a way, off the main line of advance.
Besides established reactions, reflex actions, and tropisms there are rhythmic activities adjusted to external periodicities, such as change of position in shore animals when the tide goes out or comes in. That these may be more than tropisms is shown by cases where the rhythm is so engrained in the creature's constitution that it persists for a time in periodic expression even when the external stimulus has ceased. The interesting green worms called Convolutas, well-known on some flat beaches, such as that of Roscoff, come up to the surface of the sand when the tide goes out, and retreat again when the tide comes in. Bohn has found that they will continue doing this for a couple of weeks in an aquarium away from the sea. Similarly, some hermit-crabs which make for the light at high tide and away from the light at low tide have been observed doing this for some time at the proper hours in a tideless aquarium.
It is a question for expert discussion and further experiment how far the conception of tropisms will carry us as an interpretation of the ways of the lower animals. Loeb believes it will cover most cases; Jennings thinks its scope is narrowly limited. Just as tropisms differ from ordinary reflexes in being usually adjustments of the animal as a whole, so behaviour differs from tropisms in being an effective concatenation or correlation of successive adjustments. In a tropism there is really but one adjustment, which is repeated over and over again. In behaviour there is trial after trial of different reactions, and a selection of the best available result.
§ 7. Non-intelligent Experimentation.
Preoccupation with reflexes and tropisms is apt to lead to an ignoring of the trial movements which are common among the lower animals. Unprejudiced observation of most Invertebrates will show that they perform many movements which have no fixed relation to sources of external stimuli, but which do serve to test the surroundings and thus to guide the animal (Jennings, p. 247). Prof. S. J. Holmes writes to the same effect and gives many illustrations:The lives of most insects, crustaceans, worms, and hosts of lower Invertebrate forms, including even the Protozoa, show an amount of busy exploration that in many cases far exceeds that made by any higher animal. Throughout the animal kingdom there is obedience to the Pauline injunction, Prove all things, hold fast that which is good (quoted by Jennings, p. 250).
Among simple multicellular animals there is, one must admit, not a little of that restless locomotion which we see in Infusorians and the like, which we have called general organismal activity. But at any moment this may give place to more definite behaviour. The creature commands its course and is neither blown hither and thither by every tropistic gust nor bound by reflex routine. It makes sensorimotor experiments which work towards an end, such as the systematic exploration of a corner in search of food. It shows control and selection. It may profit by experience, even though it has no brain.
The sea-anemone Antholoba reticulata, described by Bürger, usually lives on the back of a crab. If it be removed it fixes itself to the stony floor of the sea and spreads its tentacles, biding its time. After four or five days it frees itself and turns upside down. Now if the upturned base of the sea-anemone be touched by a crab's leg, it lays hold, folding itself about the limb. It now, in the course of several hours, climbs up the crab's leg to its back, where it establishes itself. The sea-anemone thus by its own activity attains the extraordinary situation where it is usually found. The whole train of action is like that shown in the complicated and adaptive instincts of higher animals (Jennings, p. 197).
As the type-case of what we propose to call simply organismal behaviour (or perhaps sensorimotor behaviour), we take the attack which the brainless, ganglionless starfish makes on the brainless, ganglionless sea-urchin (see Prouho, 1890). The starfish lays an arm upon the spinose surface of the sea-urchin and grips with its suctorial tube-feet. The sea-urchin responds by biting with its numerous snapping organs or pedicellariæ which close on the tube-feet. The starfish then draws away an arm, wrenching off the pedicellariæ. It repeats the process with the same or another arm until the sea-urchin is cleared of its weapons. The starfish then protrudes a portion of its highly elastic stomach over its victim, and the business is over. Now some of the items in the procedure are probably purely reflex, such as the attachment of the tube-feet, but the point is that the starfish exhibits a chain of actions, certainly not in the line of least resistance, which are mutually adjusted or correlated in such a way that they bring about an end. How the conative bow of the starfish was bent towards that end and kept towards that end, who shall tell us, but that we have here to do with behaviour seems undeniable. It appears to us to be an important fact that ganglionless animals show a trial-and-error method, a selection of the responses that put things right, and for a short time, at least, a profiting by experience. We cannot call this intelligent behaviour, but it is objectively the counterpart of intelligent behaviour.
This stage in the evolution of behaviour may be said to mark one of the great events in the history of life. As the organism became more differentiated it was open to a larger number of stimuli; as it gained a foothold in particular situations the door to choice was unlocked; as experience began to be garnered it became possible for an internal impulse to control the natural reaction to a stimulus. This was the dawn of freedom.
§ 8. Instinctive Behaviour.
When a spider makes a web or the bees a honeycomb, when a digger-wasp paralyses insects and stores them in its burrow as provender for its offspring, when a male stickleback builds a nest, when a young moorhen swims deftly the first time it touches the water, we have to deal with instinctive behaviour. It reaches its climax and its purest expression in Arthropods, such as ants, bees, and wasps; in birds and mammals it is more likely to occur in co-operation with intelligence.
There seems indeed to be a sharp contrast between what Sir Ray Lankester calls the big brain type, which reaches its finest development in birds and mammals, and the little brain type, the climax of which is in ants, bees, and wasps. The big brain type is relatively poor in ingrained capacities of instinctive behaviour but is eminently educable: the chick reared in an incubator in the laboratory does not recognise what water is, even when it is thirsty and standing in it; it does not know what its unseen mother's cluck means; and it will stuff its crop once or twice with worms of red worsted. But it learns to find its way about with prodigious rapidity. The little brain type is rich in ingrained capacities of instinctive behaviour, but is relatively non-educable. If a bell-jar be placed over the nest of a ground-wasp, from the door of which the inmates are wont to fly away, they are psychically unable to force a path out amid the herbage pressed down by the edge of the glass. Even when those outside force a way in, they cannot come out again, or give their fellows a hint how to escape.
It must not be supposed, however, that the little brain type is unable to profit by experience. We know in fact that they build up complex chains of associations. It is instructive to recall Professor Yung's experiments with hive-bees. Of 20 taken in a box into the country 6 kilometres from Geneva, 17 returned; of those 17 taken next day out on to the lake, none returned. This is to be contrasted with the successful return of the terns which were taken from Bird Key in the Tortugas to Cape Hatteras, 850 miles into seas never before visited; yet some returned in safety to their nests.
Those who incline to use in reference to ants and bees, crabs and spiders, the terms we need in describing our own activities, should remember the great differences in the plan of the nervous system in the respective ranks of Arthropods and Vertebrates. In the former there is much less centralisation; the cerebral ganglia are connected with a ventral chain of ganglia which are able to control many actions by themselves. We must remember that a wasp or a bee may go on feeding after its tail has been cut off, as Baron Munchausen's horse went on drinking after most of its body had been shot away. Even a decapitated insect can do a good deal, like St. Denis who walked round the town with his head in his hands. But whatever may have been the saint's reflections, we may be sure the insect has none.
Before going further let us take a thoroughly typical instance of instinctive behaviour, and there is no better than that of the Yucca Moth (Pronuba yuccasella), which has been often cited. When the large yellow bells of the Yucca open, each for a single night, the silvery moth, just emerged from her chrysalis, sets forth to visit them. From the anthers of one she collects pollen, which she kneads into a ball, and holds beneath her head. She flies to another flower, pierces the pistil with her ovipositor, lays her eggs among the ovules, and then places the fertilising pollen-pellet in the funnel-shaped opening of the stigma. Without the pollen thus brought by the moth the ovules would not develop. The larvæ of the moth eat a share of the developing ovules, but not more than about half are required. So that both plant and insect are served. In referring to this extraordinary case Prof. Lloyd Morgan writes: These marvellously adaptive instinctive activities of the Yucca moth are performed but once in her life, and that without instruction, with no opportunities of learning by imitation, and, apparently, without prevision of what will be the outcome of her behaviour; for she has no experience of the subsequent fate of the eggs she lays, and cannot be credited with any knowledge of the effect of the pollen upon the ovules. The activities also illustrate what is by no means infrequent in the more complex instincts, namely, the serial nature of the adaptation. There is a sequence of activities, and the whole sequence is adaptive in its nature.
What are the general characteristics of instinctive behaviour as exhibited by animals like ants, bees, and wasps, of the little brain type?
(1) Instinctive behaviour in its typical form is always specific or particulate. The garden-spider's web is not like the hedge-spider's web; the nest of one wild-bee is not like another's; each wasp has its own victims which it deals with in its own way; the female butterfly lays eggs on specific food-plants which are appreciated not by her but by the future caterpillars; and so on. Another aspect of the particulateness is a certain wooden lack of plasticity.
(2) The routine of instinctive behaviour has often a considerable degree of perfection the very first time, and while it may be improved by practice, it certainly does not require learning or experimenting. In other words, the instinctive behaviour depends upon a hereditary predisposition of the nervous system. Professor Driesch has defined instinctive behaviour as a complicated reaction that is perfect the very first time, but this inclines to be too hard and fast, for there is a certain amount of individual development in some instincts. None the less Paley expressed one of the characteristics of instinct when he spoke of it as a propensity prior to experience and independent of instruction. Instinctive behaviour just comes when the organism is exposed to the appropriate stimulation.
(3) The capacity for a particular piece of instinctive behaviour is shared with approximate equality by all like members of the species. All the female spiders of a given species make an equally fine web; all the males an equally inferior one. The capacity is often very markedly sex-linked, the one sex doing with perfect finish what the other does not do at all; thus the drones of the bee-hive take no part in comb-making. One must not, of course, suppose that instinctive capacities are not variable; the point is rather that instinctive equipment is much more uniform than intellectual endowment. It may be admitted that part of the individuality of intelligence is due to the fact that intelligence is as much made as born, which brings us back to the contrast that instinctive capacity is much more inborn than made.
(4) Instinctive behaviour is always adaptive to the normal conditions of the animal's life, though it may prove ineffective or misleading in face of peculiar exigencies. It has to do with particular events and circumstances, particular stimuli and configurations, which frequently recur, or, if not, are of vital moment (as in the escape from the imprisoning egg-shell); and a slight change in the conditions is likely to result in extraordinary nonplussing.
A study of these limitations tends to impress us with the difference between purely instinctive behaviour, and that experimental, inferential, or reflective kind of behaviour which we call intelligent. Let us illustrate.
The veteran French naturalist Fabre, who died in 1915 at the age of ninety-two, relates that he induced a long file of procession caterpillars to move round the circular parapet of a fountain, and by making the head of the leader touch the tail of the last member formed a living circle which continued for days circumambulating futilely. They knew nothing about anything. The grub of the mason-bee is hatched in a mortar-cradle with a lid through which it has to cut its way. This it does without difficulty. If the lid be artificially thickened by gluing on a piece of stout paper, this makes no difference to the success of the boring. But if a little empty paper box be placed over the lid the grub emerges into this, and having completed the boring part of its inborn routine cannot recommence it, and dies in its paper prison. Limitations of this sort are quite characteristic of purely instinctive behaviour and seem to remove it far from intelligence.
But the rigidity of instinctive routine must not be exaggerated. Professor and Mrs. Peckham have made a careful study of the instincts of wasps, both solitary and social. Several of the solitary forms go through the same general routine, but with interesting generic, specific, and even individual differences. When the femaleAmmophila, for instanceis ready to lay eggs, she makes a hole in the ground, closes it up, searches for some kind of prey (such as a caterpillar), stings it several times and pinches it, drags it to the nest, lays it down, opens the nest, drags in the paralysed victim, deposits an egg beside it, and then covers up the hole. On the whole it works like clockwork, but there may be variations and mistakes at every step! Moreover, in the wasp's routine there is probably help from intelligencein choosing a good site, in adapting the shape of burrow to the soil, in remembering the locality, in biting at the prey to suit the size of hole, and so on.
The general characteristics of instinctive behaviour have been admirably summed up by Prof. Lloyd Morgan.Instinctive behaviour is that which is, on its first occurrence, independent of prior experience; which tends to the well-being of the individual and the preservation of the race; which is similarly performed by all the members of the same more or less restricted group of animals and which may be subject to subsequent modification under the guidance of experience. Such behaviour is, I conceive, a more or less complex organic or biological response to a more or less complex group of stimuli of external and internal origin, and it is, as such, wholly dependent on how the organism, and especially the nervous system and brain centres, have been built through heredity under that mode of racial preparation which we call biological evolution (Instinct and Experience, 1912, p. 5).
It is confusing to use the term instinctive so loosely that it becomes almost equivalent to hereditary or inborn, as in phrases like instinctive pugnacity or instinctive gregariousness, for the usefulness of the term is in reference to specific behaviour. Yet it may be legitimate and useful to distinguish between general instinctive tendencies and specialised instinctive behaviour. A general instinctive tendency is the expression of an inborn impulsion which has not much particular content, such as is shown by mammals who are about to become mothers for the first time, or by an isolated hen-bird who fumbles at nest-making, or in the so-called sex-instinct. Thus we ourselves have many instinctive tendencies, but few instincts. These general instinctive tendencies are to be distinguished from fundamental appetites such as hunger, and also from general tropisms, illustrated, for instance, when young birds gather under a tea-cosy as under a motherwhere we have evidently to do with similar responses to similar stimuli.
§ 9. Theories of Instinct.
It is too soon to come to any hard-and-fast conclusion in regard to the nature of instinctive behaviour. We have not yet got the facts fully before us, and there is much need of more experimental study. It is almost certain that there are different grades of instinctive behaviour. There are three main theories at present in the field. (A) Some investigators rank instinctive behaviour as closely comparable to chains of reflex actions, and as due to non-cognitive hereditary predispositions to follow a certain routine when a number of stimuli present themselves. (B) Others regard instinctive behaviour as quite inseparable from intelligent behaviour. (C) According to a third view, instinct and intelligence are two radically different though often co-operative kinds of knowing, which have evolved along divergent lines.
(A) Some investigators rank instinctive behaviour as near compound reflex actions, as the outcome of non-cognitive hereditary impulsions or predispositions to enter upon a certain routine when a certain trigger is pulled, and to follow on in a perfectly definite manner, the result of one trigger-pulling leading to another trigger, and so on. This may be called the reflex theory of instinctive behaviour, and it is often held by mechanists in the strict sense. It may, however, be held by biologists who admit that vital processes cannot be adequately re-described in terms of chemistry and physics, who are, however, unwilling to admit in instinctive behaviour any reality beyond the physiological processes of the animal's nervous system. We shall call it, therefore, the reflex theory, rather than the mechanistic theory of instinctive behaviour.
Instinctive behaviour agrees with reflex acts in not requiring to be learned, in being dependent on hereditary nervous predispositions, and. in being exhibited approximately in the same way by all similar individuals of the species.
It differs from reflex acts in being the activity of the organism as a whole and in requiring (with few exceptions) an intact nervous system. It differs also in sometimes having some measure of plasticity or of variability, which is quite unknown in reflex actions. It differs also inasmuch as it does not always consist of acts soon over and done with and attaining a result useful in itself; it is often a unified many-linked concatenation of acts, working towards a distant result. In many of the chains of instinctive behaviour connected with parenthood, the end is very remote, sometimes never experienced; and making a dark burrow in a bank can hardly be its own reward. To describe instinctive behaviour as nothing more than, a series of intricately dovetailed reflex actions suggests a false simplicityslurring over the characteristic unification or concatenation. Considered physiologically, instinctive behaviour is based on neuro-muscular prearrangements, but to many naturalists it seems impossible to do descriptive justice to what takes place without supposing that the behaviour is suffused with awareness and sustained by endeavour.
According to Prof. W. McDougall, the higher or more complex instinctive activities are much more than compound reflexes. They are induced not by simple sense-impressions as reflexes are, but by complex groups of sense-stimuli, such as some scene. Thus insects visiting flowers show a total complex reaction to a total complex sense-impression. There is meaning or significance in it; and a sustaining conation or endeavour.
Prof. Lloyd Morgan holds an interesting view which seems more applicable to the big brain than to the little brain type. Instinctive behaviour he regards as physiologically akin to reflex action; it consists of concatenated reactions of the whole organism. The capacity for this in birds and mammals probably has its seat in parts of the brain below the cerebral cortex. But the lower centres stimulate the higher centres and intelligence qualifies the instinctive behaviour.
On Prof. Lloyd Morgan's view, intelligent guidance is the function of the cerebral cortex with its distinguishing property of consciousness; the co-ordination involved in instinctive behaviour, and in the distribution of physiological impulses to the viscera and vascular system is the primary function of the lower brain-centres; in instinctive behaviour as such, consciousness correlated with processes in the cerebral cortex is, so to speak, a mere spectator of organic and biological occurrences at present beyond its control; but, as spectator, it receives information of these occurrences through the nerve-channels of connection between the lower and the higher parts of the brain. Thus instinct and intelligence are different organs, but they co-operate, and as intelligence is kept more or less informed of the steps of instinctive behaviour it is sometimes on the spot to help the animal out if some critical situation arise which the routine-behaviour cannot meet.
Some who think that it is feasible to interpret instinctive behaviour biologically as a concatenation of reflexes are at the same time willing to admit that there may be a psychical accompaniment which does not rise to the cognitive level. Thus Minkiewicz, who has made very important experiments on animal behaviour, regards instinctive performance as implying a certain low form of unconscious, but none the less purposive psychical activity.
(B) Others regard instinctive behaviour as inseparable from intelligent behaviour. Thus Professor Stout regards instinctive behaviour as being biologically a concatenated series of reflex-like actions, dependent on hereditary neuro-muscular prearrangements. Subjectively, however, it shows conative impulse, unity and continuity of attention, perseverance with adaptive variation of behaviour corresponding to felt success or failure, and, in many cases, the evidence of having learned by experience. The congenital prearrangements of the neuro-muscular mechanism for special modes of behaviour do not of themselves suffice to explain the animal's conduct. Their biological utility depends from the outset on their operation being sustained, controlled, and guided by intelligent interest in the pursuit of ends. It seems to us that this view fits the blended instincts of birds much better than the pure instincts of bees.
According to Prof. C. S. Myers there is but one psychological functioninstinct-intelligence. In what is ordinarily called instinctive behaviour the innate mechanism is relatively fixed and given; in what is ordinarily called intelligent behaviour the mechanism is relatively plastic and acquired. But I maintain that such differences are only relative and that no mental state (or process) can be spoken of as solely instinctive or as solely intelligent.
(C) A third view, particularly associated with Professor Bergson, regards instinctive behaviour and intelligent behaviour as two quite different kinds of efficiency, implying different kinds of knowing.
If we define intelligent behaviour as that which involves objectively some trial-and-error experimenting and profiting thereby, and subjectively some perceptual inference, we may say that instinctive behaviour differs in being non-experimental (though it may improve as the result of experience) and non-inferential (though not necessarily destitute of awareness). It is the impression of many observers that instinctive behaviour differs from intelligent behaviour in the rigidity of the routine and in the absence of awareness of the end to be attained. In intelligent behaviour, as we know it in ourselves, there is an awareness of ends as ends, and there is a power of adapting old means to new ends. But it is only by an argument from analogy that we can speak about absence or presence of awareness, and even in intelligent behaviour the degree of awareness varies greatly in intensity.
According to Professor Bergson, instinct and intelligence differ in kind and have evolved on divergent paths. The ways of ants and bees cannot be described as intelligent. As Prof. H. Wildon Carr puts it, the fundamental difference lies in the mode of apprehension of reality, and the kind of knowledge that serves the activity of each. We can never know what this instinctive knowledge is. But we may approach it sympathetically in our power of intuitiona direct vision of reality that is not clothed, so to speak, with the categories of the understanding. This reality is the living activity itself apprehended as a real duration.
One of the fundamental sentences in L'Évolution Créatrice is this: The cardinal error which, from Aristotle onwards, has vitiated most of the philosophies of nature is to see in vegetative, instinctive, and rational life three successive degrees of the evolution of one and the same tendency, whereas they are three divergent directions of an activity that has split up as it evolved. The difference between them is not a difference of intensity, nor, more generally, of degree, but of kind. To this, M. Bergson has, indeed, immediately to add that intelligence and instinct are rarely to be caught pure, for instinct is often accompanied by gleams of intelligence (seen, for instance, when hive-bees nest in the open air), and there is no intelligence in which some traces of instinct are not to be discovered.
Intelligence uses unorganised instrumentstools; instinct uses inborn organised instruments. The innate knowledge in instinct is of things, of particular pieces of matter; the innate knowledge in intelligence is of relations, of forms. Instinct implies intimate and full awareness of a particular configuration of things; intelligence makes frames applicable to many things. If instinct has signs or words, they are adherent, invariably attached to a certain object or a certain operation. Intelligence has mobile signs, which can pass from things to ideas, and this language has been a great liberator. In short, instinct and intelligence are quite different expressions of life. As to the much-debated question whether instinct is conscious or not, Professor Bergson holds that there may be lively consciousness in some cases, and that it may be nullified in others. Consciousness is the light that plays around the zone of possible actions, in the interval between representation and action; it is associated with hesitation and choice. Therefore since there is much choice in intelligent behaviour and little in instinctive behaviour, the latter tends to be less conscious than the former.
The position that instinctive behaviour is on a different evolutionary tack from intelligent behaviour may be defended apart from Professor Bergson's particular view of the difference. When we observe a spider executing an extraordinarily complex and sharply punctuated series of movements which result in a web and doing this effectively the very first time, we seem to be in a world different from that of intelligence. And again when we observe insects continuing to go through a laborious routine which has lost all its point, and from bondage to which the least modicum of intelligence would deliver them, we seem to be in a world very different from that of intelligence.
While the frequent limitations of instinctive behaviour seem to us to point to a differentia between it and intelligent behaviour, we find further evidence in considering its achievements in preparing for the unforeseen and remotefor offspring which will never be seen, for the evasion of a winter which will never be experienced. There is an adjustment of means to ends which certainly does not rest on a basis of individual experience. It is possible to say that this organisation for the attainment of remote and unknown ends is the inherited result of an originally intelligent prevision, but there are great difficulties in face of this theory. There is certainly inherited organisation, but there is no evidence that the instinctive behaviour ever passed through an intelligent phase. In simple cases, we can imagine a sort of intelligent argument from analogy: thus the woodpecker-like bird, Colaptes mexicanus, feeds on insects while it can, but stores acorns against the day when no insects will be available. But no analogy can suggest making elaborate provision for offspring that are never seen.
If we rule out the theory that instinctive behaviour has no psychical side, for that is an outrageously false simplicity, we may say that there is a considerable amount of common ground between the various theories. There are plainly two aspects of instinctive behaviourobjective and subjective. There is the hereditary organisation of the nervous system which has been so prepared or evolved that the specific behaviour comes automatically when the organism is appropriately stimulated. But there is also the associated instinctive experience, some degrees of awareness of the situation, some memory of analogous past experiences, some more or less dim consciousness of an end. Along with this cognitive factor there is a conative one, a predetermined bending of the constitutional bow in a particular direction. And there may also be, in some cases, an evocation of associated emotions.
According to Professor McDougall, instinct is a functional unit which is transmitted as such from generation to generation, but it implies the existence in the creature's innate constitution of three thingsfirst, a specialised perceptual disposition; secondly, a specific conative tendency that is excited when this perceptual disposition is played upon by the appropriate sense-impression; and thirdly, some co-ordinated system of motor channels through which the conative tendency works towards its satisfaction.
Less technically we may say that there is (1) some degree of awareness of what is being done, (2) a feeling of activity and a bent bow, and (3) the constitutionally ingrained linkages which make a chain of reflex-like acts possible.
§ 10. Evidence of Intelligent Behaviour.
Especially among birds and mammals we find behaviour which cannot be adequately described without using psychological terms. It implies, objectively, some trial-and-error experiments and profiting thereby, some learning that is more than woodenly associative, something more than the dog's secretion of salivary juice when the dinner whistle is blown. We infer that it implies, subjectively, some perceptual inference, some working with ideas, some appreciation of the relations of things. It is reflective and experimental as contrasted with reflex and instinctive.
The Greek eagle lets the tortoise fall on the rocks so that it is broken, just as the rook does with the fresh-water mussel. The collie anticipates a possible straying of the flock and guards against its occurrence. Beavers cut a canal right through an island in a big rivera task not practically justified till it is completed. It is strictly impossible to prove that these animals really put two and two together as we do in perceptual inference, but no less generous interpretation seems adequate.
When Dr. G. T. Romanes's chimpanzee was asked for a number of straws up to five, it used to pick up the required number and present them with the ends exposed between finger and thumb. When it was right it got its reward. Sometimes, however, if asked for four straws, it would gather three to save time and double one of them so that four ends showed. When a reward was refused on such occasions, it would straighten out the doubled straw, pick up another one, and present the required number. In a case of this sort we are inclined to admit intelligence, for it was rather subtle and novel, and we know that the chimpanzee has a highly developed brain.
But pass to one of Miss Drzewina's experiments with hermit-crabs. She removed them from their borrowed shells and gave them similar shells which had been plastered up. The hermit-crabs spent a long time trying to get into these closed shells. Eventually, however, they gave it up as hopeless, as of course it was. When some shells of the same sort, but empty, were put into the aquarium the hermit-crabs would not look at them. The established association was too strong. Yet when some other shells of a different shape were introduced, the hermit-crabs tried them at once. The question is whether this also was intelligent behaviour, or whether it illustrated what we do not understand, a profiting by experience on a lower than an intellectual level, such as must form the basis of the very effective agency of the brainless, ganglionless starfish already referred to. And our inclination to be parsimonious in our interpretation is of course strengthened by the fact that the hermit-crab belongs to the little brain type of organisation on quite a different line of evolution from Vertebrates. Many spiders are readily deceived if a vibrating tuning fork is brought near their web. They rush out to deal with the situationresponding to the familiar tremor stimulus. They may be cheated over and over again. In one case, however, after a tantalising deception extending over fifteen days, the spider ceased to give any attention to the tuning fork. The question is whether we must in such a case postulate memory and perceptual inference, or whether some purely physiological interpretation is adequate. Thus the getting used to a stimulus may be in some cases due to fatigue, in the wide sense, including dulled sensation. Our inclination to a parsimonious interpretation in such a case as this is strengthened by the fact that even brainless and ganglionless animals illustrate a modification of activity by individual experience. Repeated stimulation alters the physiological condition of an animal so that it gives an intensified reaction to a moderate stimulus, as in the case of an earthworm that has been teased a little. Contrariwise, repeated stimulation that leads to nothing may result in the suppression of a reaction, as in sea-urchins that soon stop answering back to fruitless changes in light and shade, or in sea-anemones that cease to respond to the touch of false food. Even the carnivorous plant, Venus's Fly Trap, refuses to be duped many times in succession.
One of the marks of intelligence is profiting by experiencelearning. At a lower level there is temporary modification of behaviour, and this passes, insensibly we think, into lasting modification. A crab or a crayfish learns in a week or two to distinguish infallibly between the right way and the wrong way to food and freedom. How far down this capacity extends we do not know; perhaps it requires a nervous system of considerable complexity. If we obey the law of parsimony, we are led to the conclusion that the creature under sufficient stimulus of reward and of shortening a period of unpleasantness and unrest, forms a habit without knowing how, though probably with high-strung attention and delicate quivering sensitiveness, and precise registration of sequences of movements; and that after the trick has been learned it trusts itself, as a piano-player does who learns in quite a different way. Miss Washburn notes that an animal that has gone astray on the path will often find the way back to the starting-point, and from there traverse the whole road rapidly and unerringly, apparently in the same way that a piano-player who has a piece at his fingers ends, but has stumbled in a passage, can go through with entire success if he starts over again. As piano-players know, in such a case it is much better not to attend to stimuli at all, but to think of something else; the movements will take care of themselves better if consciousness intervenes as little as possible (1909, p. 231).
Many experiments have been made with rats, dogs, cats, chicks, and other creatures, which learn in the course of time to find their way out of labyrinths and puzzle-boxes. After some practice they are left in peace for a few days and then replaced in the previous situation. It is observed that they make fewer useless movements, that they sometimes make none. The question is whether ideas are at work, whether the creatures think. Have they remembered images of their successful movements, or do they obey the promptings of an organismal registration in which ideas have not been involved?
In her admirable book on The Animal Mind, Miss Washburn points out that images of a Hampton Court maze are difficult, and that the slow learning and the nature of the mistakes do not suggest working with ideas. Similarly, in regard to puzzle-boxes, she says that the slow learning, by gradual elimination of useless movements, suggests the absence of any guiding idea of the action, and Professor Thorn-dike, who initiated these experiments, corroborates this view by pointing to the entire lack of inferential or reflective imitation. That is to say, the successful behaviour of companions does not seem to be suggestive. But other observers, such as Professor Hobhouse, have come to the opposite conclusion, and in any case, as has been well said, We cannot conclude that an animal is incapable of ideas because it does not have them suggested to it under circumstances that would suggest them to our minds.
§ 11. Secondary Simplifications of Behaviour.
The difficulty of understanding animal behaviour is increased by the occurrence of secondary simplification. We are familiar with this in the individual habituation of exercises which originally required attentive selection and detailed control. What required conscious regulation from step to step becomes automatic, requiring very little attention, and the objective side of this is believed to be the establishment of nerve-paths of least resistance, of linkages such that one phase of the behaviour automatically evokes the next. One of the features is the dropping out of what is called implicit behaviour, a common name for the movements, too slight for detection, which seem at first to intervene, e.g., in type-writing and piano-playing, between the external stimulus and the overt reactions when more than a single reflex is involved.
There is not at present any convincing evidence that the direct results of habituation can be as such entailed on the offspring, and there are few available facts in support of the theory once widely held that instinctive predispositions to go through a certain routine are the hereditary results of the habituation of what was originally intelligent. That the instinctive capacities are inborn is certain, but it does not follow that they have been due to lapsed intelligence. At the same time, it is a fact of observation that the individual performance of a piece of instinctive routine may bring with it an increased perfection.
It is very interesting to find that particular reactions periodically repeated may take such a grip of the individual constitution that they are exhibited even in the absence of the liberating external stimulus. It is probable that this implies, in part at least, that long continuance of external periodicities has established internal rhythms in some important part of the metabolism of the individual creature.
In the simplest forms of behaviour, which imply little more than a co-ordination of a series of reflexes towards a desired result, there must also be organic registration. This is shown by the simple experiment of putting a starfish on its back, for it learns to right itself more and more quickly as time goes on. Although its effective behaviour is not instinctive, for it has to be learned, nor intelligent, since there are no nerve-ganglia, it improves with practice.
We shall return to the subject when we come to discuss the evolution of behaviour, but in the meantime, we notice the suggestion that it is, metaphorically speaking, part of the tactics of Animate Nature to economise mental activity for higher issues by a structural organisation or registration (badly called mechanisation) of capacities for effective agency. Thus instinctive capacity being a substitute for instruction may make emancipated experiment practicable, as birds well illustrate.
§ 12. Rational Conduct.
In the case of man, and probably in his case only, there is sometimes evidence of rational conduct as contrasted with intelligent behaviour. We cannot describe such conduct without using general terms; we know personally that, like original thinking, it involves experimenting with abstract ideas; it implies conceptual as distinguished from perceptual inference; it is controlled in reference to an ideal or purpose. We wonder whether even at this level there may not be a continuance of the organisation-process, for mathematicians of distinction and other original thinkers assure us of the reality of unconscious cerebration, and the absolute trustworthiness of the immediate ethical judgment of fine characters may be another illustration in a different field. There is an assured immediacy of reaction in many cases that makes the ordinary person marvel, and leaves the moral genius free to tackle more difficult problems.
§ 13. General Impressions of Animal Behaviour.
What we see is like a great staircase, exhibiting wonderful perfection at different levels, such as those of tropisms, instincts, and intelligence. We may distinguish a main line of experimenting, trial and error, and initiative from the side lines in which organisation or automatisation of behaviour predominates over immediately controlled direction. But while there is a hierarchy of activities, the diverse modes often overlap. Just as the surface relief of a countryside may show in one feature the outcrop of various strata of very different geological age, so in an animal's behaviour there is often a mingling of different kinds of activities unified in a way that baffles analysis. Instinct concatenates reflexes, and intelligence catches up tropisms. Instinctive capacities may form a basis for an advance in intelligence; and intelligently controlled behaviour may sink into habit. We have to distinguish in general (a) the ingrained or entailed hereditary capacities of responding effectively to certain stimuli, circumstances, and situations; (b) the individual tentatives, selections, adjustments, and learning which seem to many to imply some degree of awareness or pre-awareness and some conative element; and (c) the individual retention and registration of experience which facilitates for the individual the rapid repetition of effective reactions.
Our survey suggests some general impressions: