An Introduction to Social Psychology
Supplementary Chapter 1: Theories of Action
MY principal aim in writing this volume was to improve the psychological foundations of the social sciences by deepening our understanding of the principles of human conduct. In the three and a half years which have elapsed since the appearance of its first edition, I have discerned here and there in subsequent publications what seem to be traces of its influence. But none of the writers who have criticized or otherwise referred to the book seems to have noticed that it propounds a theory of action which is applicable to every form of animal and human effort, from the animalcule's pursuit of food or prey to the highest forms of moral volition. I therefore add this appendix to the present edition with a threefold purpose. First I desire to draw attention to this theory of action by throwing it into stronger relief ; secondly, I desire to present it in the form of a distinct challenge both to my colleagues the psychologists, and especially to writers on moral philosophy, to whose hands the positive theory of conduct has been too largely confided by the psychologists ; thirdly, I desire to help young students of psychology and ethics to understand the relation of the theory of action expounded in this book to other theories of action widely current at the present time. The execution of this threefold design involves a some-what technical and controversial discussion hardly suited
( 360) for the general reader; I have therefore preferred to present it in the form of an appendix, rather than to insert it in the body of the text.
I will first state dogmatically and explicitly the theory of action which is implied throughout this volume, and will then justify it by showing the inadequacy of the other theories of action that have been most widely accepted.
Human conduct, which in its various spheres is the topic with which all the social sciences are concerned, is a species of a wider genus, namely, behaviour. Conduct is the behaviour of self-conscious and rational beings; it is the highest type of behaviour; and, if we desire to understand conduct, we must first achieve some adequate conception of behaviour in general and must then discover in what ways conduct, the highest type, differs from all the lower types of behaviour.
We sometimes speak of the behaviour of inert or inorganic things, such things as tools, or weapons, or even the weather. But in such cases we usually recognize more or less clearly that we are using the word playfully—we playfully regard the object as alive —and the ground of our doing so is generally that. it seems to set itself in opposition to our will, and to strive to frustrate or hinder the accomplishment of our purpose. It is generally recognized that the word "behaviour" implies certain peculiarities which are only found in the movements of living things. These peculiarities are the marks of life; wherever we observe them, we confidently infer life. We form our notion of behaviour by the observation of the movements of living things ; and, in order to explicate this notion, we must discover by what marks behaviour is distinguished from all merely physical or mechanical movements. If
( 361) in imagination we construct a scale of types of behaviour ranging from the simplest to the most complex, we find that at all levels of complexity behaviour presents four peculiar marks.
1. The creature does not merely move in a certain direction, like an inert mass impelled by external force; its movements are quite incapable of being described in the language with which we describe mechanical movements ; we can only describe them by saying that the creature strives persistently towards an end. For its movements do not cease when it meets with obstacles, or when it is subjected to forces which tend to deflect it : such obstacles and such opposition rather provoke still more forcible striving, and this striving only terminates upon the attainment of its natural end; which end is generally some change in its relation to surrounding objects, a change that subserves the life of the individual creature or of its species.
2. The striving of the creature is not merely a persistent pushing in a given direction ; though the striving persists when obstacles are encountered, the kind and direction of movement are varied again and again so long as the obstacle is not overcome. Behaviour is a persistent trial or striving towards an end, with, if necessary, variation of the means employed for its attainment.
3. In behaviour the whole organism is involved. Every action that we recognize as an instance of behaviour is not merely a partial reaction, such as the re-flex movement of a limb, which seems to be of a mechanical or quasi-mechanical character ; rather, in every
( 362) case of behaviour, the energy of the whole organism seems to be concentrated upon the task of achieving the end: all its parts and organs are subordinated to and co-ordinated with the organs primarily involved in the activity.
4. The fourth mark of behaviour is equally characteristic and probably equally universal with the other three, though it is less easily observed; it is, namely, that, although, on the recurrence of a situation which has previously evoked behaviour, the creature may behave again in a very similar manner, yet the activity is not repeated in just the same fashion as on the previous occasion (as is the case with mechanical processes, except in so far as the machine has been in some degree worn out on the former occasion) ; there is as a rule some evidence of increased efficiency of action, of better adaptation of the means adopted to the end sought—the process of gaining the end is shortened, or in some other way exhibits increased efficiency in subserving the life of the individual or of the species.
When we survey the whole world of material things accessible to our perception, these seem, as a matter of immediate observation and apart from all theories of the relation of mind to matter, to fall into two great classes, namely, (1) a class consisting of those things whose changes seem to be purely physical happenings, explicable by mechanical principles; (2) a class of things whose changes exhibit the marks of behaviour and seem to be incapable of mechanical explanation, but rather to be always directed, however vaguely, to-wards an end—that is to say, are teleological or purposive ; and this class constitutes the realm of life.
The four peculiarities which, as we have seen, characterize behaviour are purely objective or outward marks
( 363) presented to the observation of the onlooker. But to say that behaviour is purposive is to imply that it has also an inner side or aspect which is analogous to, and of the same order as, our immediate experience of our own purposive activities. We are accustomed to accept as the type of purposive action our own most decidedly volitional efforts, in which we deliberately choose, and self-consciously strive, to bring about some state of affairs that we clearly foresee and desire. And it has been the practice of many writers, accepting such volitional effort as the type of purposive activity, to refuse to admit to the same category any actions that do not seem to be prompted and guided by clear fore-sight of the end desired and willed. When purposive activity is conceived in this very restricted way, and is set over against mechanical processes, as process of a radically different type, there remains the difficulty of assigning the place and affinities of the lower forms of behaviour.
One way of solving the difficulty thus created is that adopted by Descartes, namely, to assign all the lower forms of behaviour to the mechanical category. But this is profoundly unsatisfactory for two reasons: (i) As we have seen, behaviour everywhere presents the outward marks which are common to the lower forms of behaviour and to human conduct, and which set it so widely apart from mechanical processes; (2) this way of dealing with the difficulty creates a still greater difficulty, namely, it sets up an absolute breach between men and animals, ignoring all the unmistakable indications of community of nature and evolutional continuity between the higher and the lower forms of life.
The creation of this second difficulty has naturally resulted in the attempt to solve it by forcing the truly
( 364) purposive type of process into the mechanical category; that is, by regarding as wholly illusory the consciousness of striving towards an end which every man has when he acts with deliberate purpose ; by assuming that we are deceived when we believe ourselves to be real agents striving more or less effectively to determine the course of events and to shape them to our will and purpose. The demonstration that this view is untenable requires a very long and intricate argument, which can-not be presented here even in briefest outline. It must suffice to say that the acceptance of this view would be subversive of all moral philosophy, would deprive ethical principles and ethical discussion of all meaning and value; for if our consciousness of striving to achieve ends, to realize ideals, to live up to standards of conduct, if all this is illusory, then, to seek to determine what we ought to do and to be, or to set up standards or norms or ideals, is wholly futile; such endeavours can at best only serve to make us more acutely aware of our impotence in face of such ideals.
We can only avoid this difficulty and this impasse by recognizing that the commonly entertained notion of purposive activity is too narrow, and that it must be widened to include the lower forms of behaviour as well as the higher forms which constitute human conduct.
The only serious objection that can be raised to this widening of the notion of purposive activity is the contention that the word "purpose" essentially implies on the part of the agent consciousness of the goal that he seeks to attain, of the end he pursues; it may be said that
(365) it is only in so far as the agent may reasonably be regarded as clearly conscious of the goal he seeks that we can claim to understand in any sense or degree how the end determines the course of the activity, how, in short, the action is teleologically determined. And it may be said, with truth, that we are not warranted in believing that the lower animals are capable of conceiving, or of being in any way clearly conscious of, the ends of their actions; and therefore, it may be said, it is illegitimate to regard the lower forms of behaviour as purposive or to claim that our immediate experience of purposive activity in any way enables us to understand them.
This objection may be removed by the following considerations. Mental process seems to be always a process of striving or conation initiated and guided by a process or act of knowing, of apprehension; and this knowing or cognition is always a becoming aware of some-thing, or of some state of affairs, as given or present, together with an anticipation of some change. That is to say, mental life does not consist in a succession of different states of the subject, called states of consciousness or ideas or what not; but it consists always in an activity of a subject in respect of an object apprehended, an activity which constantly changes or modifies the relation between subject and object. Now this change which is to be effected, and which is the goal or end of action, is anticipated with very different degrees of clearness and adequacy at different levels of mental life. In many of our own voluntary actions the end is anticipated or fore-seen in the most general manner only ; to take a trival but instructive instance : you cough in order to clear your throat; or, experiencing a slight irritation in your throat, you put out your hand, take up a glass of water, and drink, in order to allay it. How very sketchy and ill-
(366) -defined may be your thought of the end of your action! And even in the execution of our most carefully thought-out, our most purposeful, actions, our anticipatory thought or representaton of the end to be achieved falls far short of its actual fulness of concrete detail. The anticipation of the end of action is, then, always more or less incomplete; its adequacy is a matter of degree. Therefore we ought not to assume that a clear and full anticipation or idea of the end is an essential condition of purposive action; and we have no warrant for setting up the instances in which anticipation is least incomplete as alone conforming to the purposive type, and for setting apart all instances in which anticipation is less full and definite as of a radically different nature.
It is important also to note that the representation or idea of the end is not truly the cause or determining condition of the purposive activity. The merely cognitive process of representing or conceiving the end or the course of action does not of itself suffice to evoke the action; we can imagine many possible actions or ends of actions, without carrying them out or feeling any inclination to pursue them; in fact it often happens that the more clearly we envisage the end and course of a possible action, the more strongly averse to it do we become. The truth is that the anticipatory representation of the end of action merely serves to guide the course of action in de-tail; the essential condition of action is that a conative tendency, a latent disposition to action, shall be evoked. Where the anticipatory representation of the end is vague and sketchy and general, there the action will be general, vague, imperfectly directed in detail ; where it is more detailed and full, there action is more specialized, more nicely adjusted to the achievement of its end.
From our own experience we are familiar with ac-
( 367) -tions in which anticipation of the end varies from that of the most clear and detailed nature through all degrees of incompleteness down to the most vague and shadowy, a mere anticipation of change of some undefined kind. We are therefore able to form some notion of the inner or subjective side of the action of animals, even of those lowest in the scale of organization. Putting aside a limited number of animal actions which owe their definiteness and precision to guidance at every point by new impressions falling from moment to moment upon the sense-organs (as in the most striking instances of purely instinctive action), we see that, as we go down the scale of life, actions become less precisely guided in detail, and present more and more the character of random or but vaguely directed efforts; in this corresponding to what we may legitimately suppose to be the increasing vagueness of the anticipatory representations by which they are guided. The theoretical lower limit of this series would be what has been well called (by Dr. Stout) anoetic sentience; a mere feeling or sentience involving no objective reference and giving rise only to movement or effort that is completely undirected. This lower limit is approached in our own experience when we stir uneasily or writhe or throw ourselves wildly about, under the stimulus of some vaguely localized internal pain. But we do not ourselves experience the limiting case, and it is questionable whether we can properly suppose it to be realized in the simplest instances of animal behaviour; it seems probable that the actions of even the lowliest animals imply a vague awareness of something, together with some vague forward reference, some vague anticipation of a change in this something.
Knowing, then, is always for the sake of action; the function of cognition is to initiate action and to guide it
( 368) in detail. But the activity implies the evoking, the coming into play, of a latent tendency to action, a conative disposition; every such tendency or conative disposition is either of a very general or of a more specialized or specific character ; and each such conative tendency, when awakened or brought into play, maintains itself until its proper or specific end is attained, and sustains also the course of bodily and mental activity required forthe attainment of that end. When, then, any creature strives towards an end or goal, it is because it possesses as an ultimate feature of its constitution what we can only call a disposition or latent tendency to strive to-wards that end, a conative disposition which is actualized or brought into operation by the perception '(or other mode of cognition) of some object. Each organism is endowed, according to its species, with a certain number and variety of such conative dispositions as a part of its hereditary equipment for the battle of life; and in the course of its life these may undergo certain modifications and differentiations.
To attempt to give any further account of the nature of these conative dispositions would be to enter upon a province of metaphysical speculation, and is a task not demanded of psychology. I will only say in this connection that we may perhaps describe all living things as expressions or embodiments of what we may vaguely name, with Schopenhauer, Will, or, with Bergson, the vital impulsion (l'élan vital), or, more simply, life; and each specifically directed conative tendency we may regard as a differentiation of this fundamental will-to-live, conditioned by a conative disposition. At the standpoint of empirical science, we must accept these conative dispositions as ultimate facts, not capable of being analyzed or of being explained by being shown to be instances of
(369) any wider, more fundamental notion. To adopt this view is to assert that the facts of behaviour, the empirical data of psychology, must be explained in terms of fundamental conceptions proper to it as an independent science. The physicist works, and explains his facts, in terms of the conception of mechanical process, not necessarily concerning himself with the metaphysical problem that underlies this conception; for example, he accepts as an ultimate fact the tendency of a moving mass to continue to move in a straight line without change of velocity. In a similar manner the psychologist may work, and explain his facts, in terms of the conception of purposive or appetitive process. The physicist studies mechanical processes of all kinds in order to arrive at the most general laws of mechanical process; and his explanation of any one fact of observation consists in exhibiting it as an instance of the operation of such general laws; that is, in showing that it conforms to the type, that it may be analytically regarded as a conjunction of simple mechanical processes obeying the most general laws of mechanism. Just in the same way the psychologist has to study appetitive processes of all kinds and of all degrees of complexity, in order to ascertain the most general laws of appetitive process. And his explanation of any process of the kind with which he is concerned must consist in exhibiting it as an instance of the operation of such general laws of appetition, in showing how it may be analytically regarded as a conjunction of appetitions according to the general laws of appetition that he has established. According to this view, then, the acts of human beings, all our volitions, our efforts, our resolutions, choices, and decisions, have to be explained in terms of the laws of appetition. When, and not until, we can exhibit any particular instance of conduct or of be-
( 370) -haviour as the expression of conative tendencies which are ultimate constituents of the organism, we can claim to have explained it.
Owing to the great development of physical science in modern times and to the immense success that has attended its attempts to explain physical facts in terms of the laws of mechanism, there obtains very widely at the present time the opinion that we understand mechanical process in some more intimate sense than we can understand appetitive process; and that, therefore, it is the business of all science to explain its facts in terms of the laws of mechanism, and that appetitive processes can only be rendered intelligible if they can be reduced to the mechanical type. But this is a delusion. Of the two types of process, we certainly understand the appetitive more intimately than the mechanical ; for we directly experience appetition, we have an inside acquaintance with it, as well as acquaintance of the purely external kind which is the only kind of acquaintance that we have with mechanical process. And when metaphysicians attempt to go behind the distinction of mechanical and appetitive processes (which for science is fundamental) and at-tempt to show that processes of the two types are really of like nature, the most plausible view seems to be that which regards mechanical process as reducible to the appetitive type or regards it as, perhaps, representing a de-gradation of process of the appetitive type. This, at least, is the view which has been and is maintained by some of the most distinguished metaphysicians and which seems to involve less serious difficulties than the acceptance of the converse view.
I have now stated explicitly the theory of action which is implied by the doctrines of instinct, of sentiment, and of volition, expounded in this volume; and it remains to justify it by showing the inadequacy of other theories of action.
The theory of action most widely accepted by psychologists at the present time is, perhaps, the theory which regards all organisms as merely machines and all behaviour as mechanically determined. I put this aside for the reasons already stated.
Of other theories, the one which has exercised the greatest influence in modern speculation is the theory of psychological hedonism; this is the theory of action which was unfortunately adopted by the founders of Utilitarianism as the psychological foundation of all their social and ethical doctrines. It asserts that the motive of all action is the desire to obtain increase of pleasure or diminution of pain. It claims to be an empirical induction from the undeniable fact that men do seek pleasure and do try to avoid pain. But its strange power to hold the allegiance of those who have once accepted it is to be explained by the fact that it seems to afford a rational explanation of all conduct, to show a sufficient cause for all action. Whenever an action can be regarded as an effort in pursuit of pleasure or in avoidance of pain, we seem to have an explanation which is ultimate and intelligible. We feel no need to inquire—Why should anyone prefer pleasure to pain, or seek to gain pleasure and to avoid pain? No other theory of the
( 372) ground of action seems at first sight so self-evident and satisfying.
It is, no doubt, possible to show the fallacious nature of the doctrine by careful examination of our own motives and unbiassed consideration of the conduct of other men. For such consideration shows that, when we de-sire any object or end, as, for example, food, what we normally desire is the object or end itself, not the pleasure that may attend the attainment of the end. But the complexity of the human mind is so great, its springs of action so obscure, that, in almost every instance of human behaviour, it is possible for the psychological hedonist to make out a plausible interpretation in terms of his theory. Two facts play into his hands: first, the fact that the attainment of any desired object or goal brings satisfaction or pleasure; for the desired object or goal can then be ambiguously described as a pleasure, and the agent can be said to have been mowed by desire for this pleasure: secondly, the fact that, even thou--': a man be really moved by the desire of pleasure, he may choose to sacrifice the pleasure of the immediate future (or even to suffer pain) in order to secure a greater pleasure at a later time. And the hedonist, when he can-not plausibly interpret an action—such as one involving the sacrifice of life in the cause of duty—in terms of his theory in any other way, can always assert that the agent was moved by his aversion to the pain of remorse which he foresees to be the consequence of neglect of duty.
For these reasons the easiest and surest refutation of the hedonist theory of action is provided by the consideration of animal behaviour. For we may observe numberless instances of action, of persistent striving towards ends, on the part of lowly animals which cannot
( 373) be credited with the power of anticipating or desiring the pleasure that may accrue from success.
A second theory of action, which claims to be of general validity, ascribes all conation, all mental and bodily striving, not to desire of future pleasure or aversion from future pain, but to the influence of present pleasure or pain; that is to say, feeling (in the sense of pleasure or pain) is regarded as an essential link between cognition and conation; it is maintained that cognition only moves us to action in so far as it evokes in us pleasurable or painful feeling. This may conveniently be designated the pleasure-pain theory of action. It is widely accepted at the present time ; it is more subtle and less easily refuted than the theory of psychological hedonism, which is no longer seriously to be reckoned with. The difficulty of refuting this doctrine arises from the fact that mental process has almost invariably some feeling-tone, is coloured, however faintly, with pleasure or with pain; so that it is possible to attribute with some plausibility almost every instance of activity to the feeling which accompanies and qualifies it. This theory rightly recognizes that what we normally desire and strive after is some object or end which is not pleasure itself, though its attainment may be accompanied by pleasurable feeling; that, for example, when we are hungry we normally desire food rather than the pleasure of eating. But it asserts that the moving power of the desire, that which prompts us to action, is the feeling, the pleasure or pain, which we experience at the moment of desire and of action; that, when we desire food, that which prompts us to strive after it is neither the pleasure which we anticipate from eating nor the pain which we anticipate from fasting, but the pleasure which arises
( 374) from the thought of eating or the pain which immediately qualifies the sensation of hunger.
The last sentence indicates the line of criticism by which this theory may be shown to be untenable. We must ask—Is the hungry man prompted to seek food by the pleasure of the thought of eating or by the pain of hunger? Some of the pleasure-pain theorists incline to the one view, some to the other, and some  boldly solve the difficulty by accepting both, asserting that desire always involves both pain and pleasure. These last assert, for example, that the desire of food is pleasant in so far as it is or involves the thought of eating, and that it is at the same time painful in so far as it is a state of unsatisfied appetite or craving. The assumption that consciousness may be at any one moment both pleasurably and painfully toned is one of very doubtful validity; but it is a further and perhaps more serious objection to this view, that the pleasure and the pain which are assumed to coexist should be assumed also to prompt to the same kind of action. And if the pleasure and the pain are assumed to alternate in consciousness, rather than to coexist, the same difficulty remains. As a matter of fact, every kind of desire or striving may be pleasurable or painful—pleasurable in so far as it progresses towards its goal, painful in so far as it is thwarted ; and yet the desire and the striving may persist while the feeling tone alternates from the extreme of pleasure to extreme of pain. Thus the desire of the lover persists, whether he be raised to the height of bliss by the expectation of success, or cast down to depths of torment by a rebuff.
If we consider the animals, we shall again be led to
( 375) the true view. It is now generally admitted that we can-not attribute to the lower animals "ideas," or any power of clearly representing, or thinking of, things not present to the senses; therefore we cannot attribute their actions to the pleasure of the idea of attaining the end pursued; yet such animals strive under the spur of hunger, as we say, and of other appetites. Therefore, in the lower realms of life all action must be attributed by the pleasure-pain Theory to present pain. But the pain of hunger seems to be in our own case the pain of unsatisfied craving; that is, the pain is conditioned by the craving, and presupposes it—if there were no craving, there would be no pain. But the craving is essentially a conation, a tendency to action, however vaguely directed. Hunger, then, is not a pain which excites to action; but it is fundamentally a tendency to action, which, when it cannot achieve its proper end, is painful; it is, in short, an appetition arising from a specific conative disposition. And it seems in the highest degree probable that this is true of the hunger of animals and of all the pains to which the pleasure-pain theory finds itself compelled to attribute their activities.
The assumption, necessarily made by the pleasure-pain theory, namely, that all the actions of animals (save possibly some of those of the highest animals) are prompted by pain, is, then, unsatisfactory, and seems to invert the true relation of feeling to conation. That human desires and actions are not exclusively or in any large measure due to present pain is obvious; the pleasure-pain theory, therefore, attributes them in the main to the pleasure which accompanies the thought of the desired end or goal. The necessity of assuming that the actions of animals and those of men are predominantly prompted by the opposite principles (pain and
( 376) pleasure respectively) should give pause to the pleasure-pain theory. But, if we waive this objection and inquire after the source or condition of the pleasure which is supposed to accompany the thought of the end of action and to prompt to action, we shall find that here too the theory inverts the true relation of feeling to conation. Desire, or the thought of the desired end, is pleasant in so far as an appetite or conation obtains some degree of ideal satisfaction through the belief in the possibility of presently achieving the act, or in so far as the activities prompted by the desire successfully achieve the steps which are the means to the end. Thus hunger, even acute hunger, is pleasant if we know that the bell will presently summon us to a well-spread table, or if we are in the act of obtaining the food we desire; yet, if the hungry man knows that it is impossible for him to obtain food, if, for example, he is a castaway in an empty boat, the thought of food is a torment to him, though he cannot cease to desire it, or prevent himself from dwelling upon the thought of it.
Both the pleasure and the pain of hunger seem, then, to be conditioned by the craving, the conative tendency, the specifically directed impulse or appetition. And this seems to be true not only of the desire for food, but of many other desires. When, for example, we desire the applause of our fellows, when we are consumed with what is called disinterested curiosity, when we desire to avenge ourselves or vent our wrath on one who has insulted us, when we desire to relieve distress, when we are impelled by sexual desire ; in all these cases the state of desiring is painful in so far as efforts are unavailing or attainment appears impossible, and pleasurable in so far as we are able to anticipate success or take effective steps towards the desired end. And in each case the
( 377) strength of desire, of the conative tendency, seems to be quite, or almost quite, independent of the quality and of the intensity of its hedonic tone ; while on the other hand the hedonic tone seems to be manifestly conditioned by the conative tendency, its quality by the success or failure of the striving, its intensity by the strength of the tendency. When, then, the pleasure-pain theorist tells us that feeling determines conation, we must ask what determines the feeling; and, if he replies that cognition of some object is the immediate condition of feeling, we point to these numerous instances in which the feeling-tone of the thought of the object varies from pleasure to pain, its quality and strength being obviously determined, not directly by cognition, but by the conation it evokes.
But if, for the purpose of the argument, we accept the thesis that the pleasure of the idea of the end, the pleasure that we experience in contemplating the end of action, is the spur that prompts and sustains action, and inquire why is the thought of the desired end pleas-ant, we find that two different answers are returned. Some of the pleasure-pain theorists tell us that the thought of the desired end or of the achievement of the end is pleasant because this end is in congruity with our nature. Now this can only mean that the end of action which on being contemplated appears pleasant is one to which we naturally tend, that is, is one towards which we feel impelled in virtue of a conative disposition directed to such an end. To give this answer is then implicitly to give up the pleasure-pain theory and to admit the truth of the view maintained in these pages.
The other answer to this question as to the source or ground of the pleasure we feel in contemplating the end of action, is to assert that all feelings are primarily the
( 378) pleasures and pains of sense, that certain sensations are intrinsically pleasant and others intrinsically unpleasant, and that all other pleasures and pains are derived from these by association. According to this doctrine, which has been most fully elaborated by G. H, Schneider, the sight of food is pleasant, because the pleasure of its taste has become associated with the visual impression according to the principle of contiguity; and the pleasure thus associated with the visual perception or representation of food is the condition of the desire for food, and prompts and sustains our efforts to obtain it. This answer may seem plausible when applied to explain de-sires whose satisfaction normally involves sense-pleasures ; though even in their case it is open to several very serious objections. First, the notion of the association of pleasure with ideas of objects according to the principle of contiguity is of very questionable validity. Secondly, the fact that the feeling-tone of desire for an object may vary, as we have seen, from the extreme of pain to the extreme of pleasure is irreconcilable with this view; for it shows that there is no fixed association of pleasure with the idea of the desired object, but that the feeling-tone of the thought of the object is a function of the way we think about it, being pleasant when we think of it as attainable, unpleasant when we think of it as unattainable. Further, this answer has no plausibility when applied to the many desires the satisfaction of which involves no sense-pleasure, such as the desires for applause, for revenge, for knowledge.
And we may attack the doctrine at the root by questioning its fundamental assumption, namely, that certain sensations are intrinsically pleasurable and others intrinsically painful. This assumption seems most plausi-
( 379) -ble in the case of what are called physical pains, but even in this connexion its validity may be seriously questioned. It may be maintained that what we call a painful sensation is essentially a sense-impression which evokes aversion, a conative tendency to escape or with-draw from the situation, a tendency which usually manifests itself clearly enough, as when the hand is snatched away from a hot surface or a pricking point; and that painful feeling only arises in so far as this conation fails to attain its end. It seems to be just for this reason that such sensations as toothache and other strong sensations from inflamed organs are so intensely painful. The various organs are endowed with their capacities for evoking these strong sensations, in order that they may be with-drawn from the influence of the excessive stimuli—the sensitivity of the teeth, for example, serves primarily to prevent our biting strongly on hard substances on which they might be broken. But when, as in toothache, tendencies which such strong sense-impressions excite fail to terminate the impression, and we vainly throw ourselves about, rock to and fro, or writhe in a thousand ways, the situation is intensely painful. Our power of voluntarily supporting sense-impressions that normally are painful points in the same direction. When for any reason we voluntarily submit to strong sense-impressions (as when we have a tooth filled by the dentist, making up our minds to submit to the necessary pain), we suppress by a strong effort of will partially or wholly the tendency to escape the strong sense-impression; and, in so far as we are successful in this, it loses its painful character. In this way also, I think, we must understand such extreme examples of fortitude as the calm behaviour of the Indian brave or the Christian martyr under torture; the training and beliefs of such persons render them ca-
( 380) -pable of voluntarily submitting to the torture and of sup-pressing by strong volition the tendency to struggle to escape evoked by the strong sense-impressions ; and, in so far as they succeed in this, the experience ceases to be painful—the stake and the rack are robbed of their terrors. For the same reason hunger, voluntarily submitted to (as when we fast for the sake of our health) is but a matter of small discomfort, though we are told that it is very painful when suffered involuntarily.
It may be maintained with equal plausibility that the pleasures of sense also are conditioned by conation. If we consider the case of the pleasures of the palate, we see that the pleasant tastes are those which stimulate us to maintain the processes of mastication and deglutition. According to the pleasure-pain theory, these activities are induced and maintained by the pleasure which the taste excites. But how can this view be maintained in face of the fact that the same taste qualities cease to be pleasing so soon as they cease to evoke these activities? Thus, one who likes sweet things finds the taste of sugar pleasant so long as it subserves its normal function of exciting the processes of ingestion; but as soon as repletion ensues, the tendency to mastication and deglutition can no longer be excited by the sweet taste (for this requires the co-operation of certain visceral conditions which are abolished by repletion), and the mastication of the sugar then ceases to be pleasant, and may even become decidedly unpleasant, if for any reason we persist in it.
It appears, then, that even in those instances most favourable to the pleasure-pain theory, the facts are difficult to reconcile with it, and are more consistently in harmony with the opposite view, namely, that pleasure and pain are always conditioned by the success and fail-
( 381) -ure of conation, respectively. And the superiority of the latter view will be established if we can point to in-stances in which activity is unmistakably independent of pleasure and pain; for by such instances, the pleasure-pain theorist would be compelled to admit that his theory of action holds good of some activities only, and that others require a different theory for their explanation, namely, the theory which makes feeling dependent on conation and which seems quite adequate to the explanation of the types of activity most favourable to the pleasure-pain theory. Such instances we may find at the two extremes of human behaviour; namely, in the actions implying the highest moral effort and in merely habitual actions. Whether or no we accept as true the story of the voluntary return of Regulus to captivity and death, we all recognize that it represents a possible type of conduct. Now while psychological hedonism has to explain such conduct by supposing that Regulus was more averse to the pains of remorse than to those of bodily torture and death, the pleasure-pain theory is driven to suppose that the contemplation of the heroic line of action yielded Regulus a high degree of pleasure, and that this pleasure impelled him to pursue this line of action even though he anticipated from it a painful death; or the alternative explanation might be suggested, that he found his absence from Carthage so painful that he was impelled by this pain to return thither. Surely, whether from the ethical or the psychological standpoint, this form of the hedonic theory when applied to such in-stances of hard choice, appears hardly less fantastic than psychological hedonism ! Surely it is obvious that men do often carry through a line of action which is to them painful in every phase, in the contemplation of it, :n deciding upon it, and in its execution and achievement !
( 382) Consider the more familiar instance of the father who feels himself impelled to inflict severe punishment upon a beloved child, such as the withholding from it the enjoyment of something to which they had both looked for-ward, hoping to enjoy it together. At every stage the father hates the necessity laid upon him, and knows that lie himself is sacrificing a keen pleasure and undertaking a painful task. It would be absurd to say that the father's conduct is sustained by the pleasure of the thought of the improved conduct or character of his son which the punishment may bring about. Even if at times he may find consolation in this thought, it can be but momentarily ; and such pleasure will be in the main wholly submerged and neutralized by his sympathetic pain and by the violence he does to the immediate promptings of parental love.
Instances of purely habitual and quasi-mechanical actions are not less decisive. We sometimes find ourselves performing some trivial familiar action, without having intended or resolved to do it, but merely because we hap-pen to be in a situation in which this action is habitually performed; as when one winds up one's watch on changing one's waistcoat. Such "absent-minded" actions involve a minimum of attention, but are nevertheless conation; they are the expressions of habits, and seem to be independent of pleasure and pain, whether anticipated or experienced at the moment. Such an action is immediately induced by the sense-impressions of the moment ; they bring into play the specialized conative disposition which is the habit. Such actions, better perhaps than any others, enable us to understand in some degree the way in which many of the actions of the animals are per-formed.
We may pass on to consider other theories of action;
( 383) and we may notice first the only remaining theory which makes any claim to be applicable to human behaviour of all types and levels. This is the intellectualist theory of action which attributes action immediately to "ideas," ignoring the obvious fact that the development and organization of character, or of the conative side of the mind, is largely distinct from and independent of the development and organization of knowledge, the cognitive side of the mind. Prominent among older exponents of this theory was Herbart, and, among contemporaries, Professor Bosanquet and (if I have not wholly failed to understand his writings) Mr. F. H. Bradley.
According to this theory, the mind consists of a more or-less highly organized system of ideas; and every idea is both an intellectual entity and a tendency to action. The type of all the higher forms of action is the so-called ideo-motor action, the action which is supposed to result directly from the presence in consciousness of the idea of that action. Volition is merely a somewhat complicated instance of such ideo-motor action.
Now, it may be seriously questioned whether any action really conforms to the alleged ideo-motor type. Actions proceeding from so-called fixed ideas have usually been regarded as examples par excellence of ideo-motor action. But the modern developments of psycho-pathology are making it clear that in all such cases the fixed idea is fixed, and is capable of determining action, just because it is functionally associated with some strong conative tendency. But, putting aside this objection and accepting for the purpose of the discussion the notion of ideo-motor action, I urge that it would be manifestly absurd to say that action which is carried out with painful effort against inner and outer difficulties of all sorts, is simple ideo-motor action. We have to ask—What
( 384) gives the one idea of action the power to prevail over other ideas of action equally vividly conceived? Bradley's answer to this question is that the self identifies it-self with the end the idea of which prevails. Bosanquet answers that it is attention to the one idea. Both answers are true if the "self" and "attention"are understood in the true sense; that is, if the self is understood as the vast organization of conative dispositions which is the character, and if attention is understood as conation revealing itself in cognition. But for Bosanquet attention is merely apperception in the Herbartian sense, the fusion of an idea with a mass of congruous ideas; and since conation is not recognized, the congruity implied is logical congruity. Whatever idea of action, then, is congruous with other ideas of action is apperceived or attended to, and therefore predominates over other ideas; and this is volition. Bosanquet adds that "in cases of deliberative action at a high level of consciousness, the self or personality participates, i.e., one of the ideas which are striving for predominance reinforces itself by the whole mass of our positive personality." But he explains that the whole self or personality is merely a mass of ideas with their accompaniments of feeling, "a fabric of ideas accompanied with their affections of pleasure and pain, and having a tendency to assert them-selves in so far as they become partly discrepant from reality." And in Bradley's view also the self seems to be merely a "fabric of ideas." In this intellectualist theory of action, then, conation, or will, which, as has been maintained throughout this volume, is the very foundation of all life and mind, is simply ignored; and my
(385) criticism of it must consist in pointing to all that has been said of instinct, sentiment, and volition in this book. Unless all this is the purely fanciful construction of a. diseased brain, this intellectualist doctrine is radically false. I will only point out in addition that, when we turn to the lower forms of life, the impotence of this theory is at once clear ; for, since at that level we cannot postulate "ideas," all action has to be interpreted as purely mechanical reflex action; and we are then faced with the problem of evolving intellect and will from unconscious mechanism, a task to which, as is generally recognized, the ingenuity of Herbert Spencer himself proved inadequate.
All other theories of human conduct may be classed together in virtue of the fact that they place moral conduct in a separate category, apart from all other forms of behaviour, and attribute it to some special faculty peculiar to human beings, which they call "conscience," or "the moral sense," or "reason," or the "rational will," or "the sense of duty"; a faculty which seems to be conceived as having been implanted in the human mind by a special act of the Creator, rather than as being the product of the slow processes of evolution. Most of those who attribute moral conduct to any such special faculty recognize that human nature comprises also certain lower principles of action, which they call animal propensities, instincts, or passions; and these are regarded as regret-table survivals of our animal ancestry, unworthy of the attention of a moral philosopher.
All these doctrines are open to two very serious objections: (1) that they are incompatible with the principle of the continuity of evolution; (2) that they are forms of the "faculty doctrine" whose fallacies have so often been exposed. But a few words must be said
(386) about the more important of them. When authors tell us that "reason" is the principle of moral action, it is necessary to point out that the function of reason is merely to deduce new propositions from propositions al-ready accepted. Suppose a hungry man to be in the presence of a substance which he does not recognize as food; by the aid of reason he may discover that it is edible and nutritious, and he will then eat it or desire to eat it; but, if he is not hungry, reason will not create the desire or impel him to eat. And in the moral sphere the function of reason is the same. Reason aids us in determining what is good, and in deducing from our knowledge of the good conclusions as to what actions are right. But, unless a man already hungers for righteousness, already desires to do whatever is right, to be whatever is virtuous, unless, that is, he possesses the moral sentiments and moral character, reason cannot impel him to do right or to desire it. To create desire is a task beyond its competence ; it can only direct pre-existing tendencies towards their appropriate objects. It is therefore a grave error on the part of some authors to say that reason may create a desire for a moral quality; or to say (as Sidgwick said) that in rational beings as such the cognition or judgment that this is right or ought to be done "gives an impulse or motive to action." For this is not true of rational beings as such—in Satan, we may suppose, no such impulse would be awakened by this is-sue of the reasoning process. It is true only of moral or moralized beings as such, beings who already desire to be
(387)virtuous and to do the right. It is only by arbitrarily and implicitly defining the "rational being" as one who desires to do right, that the doctrine is made to seem plausible. Nor is this doctrine, that moral conduct proceeds from - the reason, appreciably improved when "the rational will" is put in the place of "reason." This may seem to avoid the intellectualist fallacy of assigning intellectual processes as the springs of action. But, unless some further account of the will is given, this doctrine is in no way superior to the doctrine of "conscience ;" for the "rational will" remains a mere word, by which we denote the fact that we do make deliberate moral choices and decisions, and that such choice is not merely the issue of a brute conflict of opposed desires.
Though the intuitionist doctrines which attribute moral judgment, moral choice and effort, to a special faculty. have been variously stated, and though the supposed faculty has received a variety of names, they are essentially similar and need not be separately considered. We may consider that form which derives from Kant and attributes our moral judgments and conduct to "the sense of duty." It is no longer seriously contended that all the actions of any moral being spring from the "moral faculty." It is admitted that upon most of the ordinary occasions of life our actions spring from other principles or sources. But it is maintained that, in deliberation which issues in moral decision, this issue is determined by the co-operation of "the sense of duty." The "sense of duty" is in fact the last refuge of intuitionism, of those moralists who insist upon making of man's moral nature a mystery, separate from the larger mystery of mind, and implying laws of an order radically different from those which govern behaviour in general. Canon Rashdall writes : "In claiming for the idea of duty not
( 388) merely existence but authority, we have implied that the recognition that something is our duty supplies us with what we recognize upon reflection as a sufficient motive for doing it. . . . The recognition of the thing as right is capable of producing an impulse to the doing of it." And he speaks of the "sense of duty" as being "the one all-sufficient motive present to the consciousness" at moments of moral crisis.
This doctrine, if true, obviates the need for all psycho-logical investigation or reflection on the part of the moral philosopher; except in so far as he desires to expose the errors of his predecessors, by showing how they proceed from a false and unnecessarily complicated psychology, such as that of Kant or that of the founders of Utilitarianism. For the whole of the positive psychology required by him is contained in a nutshell, in the sentence : "Reason proclaims my duty, and my sense of duty impels me to do it." But some of the modern exponents of intuitionism, unfortunately for the consistency of their doctrine, are not content to leave their "sense of duty" an utterly mysterious faculty of which nothing more can be said. Sidgwick asserted that the notion of "ought" or duty is too elementary to admit of formal definition ; and in the same spirit Dr. Rashdall tells us that the idea that something ought to be done "is an unanalysable idea which is involved in all ethical judgments." But he ventures further and tells us that "Duty means precisely devotion to the various kinds of good in proportion to their relative value and importance" ; and again : "At bottom the sense of duty is the due appreciation of the proportionate objective value of ends."
(389) From this it appears that, by the admission of a prominent exponent of the intuitionist doctrine, "the sense of duty" is not an ultimate element of the moral consciousness, is not an unanalysable idea and at the same time an impulse to action ; rather it appears as the highly abstract name for all that immensely complex part of the mental organization which is the moral character, and which comprises the system of the moral sentiments and the developed self-regarding sentiment. For it is the possession of developed moral character, and this alone, that enables us to judge rightly of the relative values of moral goods and impels us to pursue the best ; and, as I have tried to show in this book, and as indeed is now generally admitted, this complex organization which is moral character is only acquired by any individual by a slow process of growth continued through many years under the constant pressure of the social environment and of the moral tradition. Our "sense of duty" is, in short, at the lower moral level our sense of what is demanded of us by our fellows ; and, at the higher moral level, it is our sense of what we demand of ourselves in virtue of the ideal of character that we have formed. How and why we respond to these demands made upon us by our fellows and by ourselves, and how we come to make these demands, I have tried to show by means of a general theory of action, a theory of the moral sentiments and a theory of volition.
Before dismissing the theory of "a moral faculty," I must add that in one respect the intuitionist doctrine is true ; namely, it is true that when we have acquired moral sentiments we do frequently both pass moral judgments and make moral efforts without any weighing of the con-sequences of action. But to admit or to establish this is neither to justify the doctrine of "a moral faculty," nor
( 390) to deny that our moral judgments frequently need correction by reference to the consequences of action upon human welfare, the only true and ultimate criterion of moral value.
We may admit also the possibility that, though the moral sentiments are in the main built up anew in each individual in the way roughly sketched in the pages of this volume, some predisposition to their formation may be inherited, and that, in so far as this is the case, the capacity of moral judgment, which is rooted in them, may be said to be innate and, in that sense, a priori.
It only remains to show that the theory of action here set forth is implied in the doctrines of some eminent philosophers (although it has not been explicitly stated by them), and most clearly perhaps by T. H. Green and Prof. Stout. These authors recognize the actions of animals as true conations or expressions of will, in the wider sense of the word "will." They recognize that human nature is capable of, or liable to, similar modes of primitive conation ; and that desire is a comparatively complex mode of conation of which, perhaps, in the proper sense men only are capable. But they do not claim that volition or moral conduct is nothing more than the issue of a conflict of desires. They rightly tell us that these simpler modes of conation, blind impulses, cravings, and desires, are something that each man experiences as, in a sense, forces acting upon him, impelling him towards this or that line of action ; and that he knows that his true self can either oppose such tendencies, or can accept them ; and that only when the self thus intervenes to accept or resist desire or impulse do we perform a volitional act. And by the self they do not mean an abstract entity of which no account can be given. Green tells us
(391) that by the true self he means the character of the man ; he uses also the term "conscience" to convey the same notion ; and by conscience he means something which has a history in the life of the individual, something that is slowly built up in the course of moral training and under the influence of the social environment; conscience or moral character is, in short, in Green's view an organized system of habits of will.
Stout also tells us that volition is distinguished from mere conflict of desires by the decisive intervention of self-consciousness; and that this self, which in moral conflict self-consciously throws itself upon the side of one desire and against others, is a unified system of interests. Now an interest is, for Stout, a conative tendency with the accompanying potentialities of feeling; and the self, therefore, is a unified system of conative tendencies.
These authors, then, have put forward in very general terms the theory of action which I am defending. They recognize will as a fundamental faculty co-ordinate with cognition; they recognize that in all organisms (animals and men alike), this faculty of striving is directed either vaguely or with more or less of precision towards certain kinds of action which tend to secure specific ends; that when these conative tendencies are brought into play in relative isolation, sporadic impulse, desire, or action is the result ; and they recognize that moral volition and moral conduct depend upon the systematic organization of such tendencies ; that in short, moral volition ex-presses character or is character in action. Their doctrines, then, imply the thesis here maintained ; namely, that in order to explain or understand any action we have to exhibit it as the expression of some single conative disposition, or of a conflict of, or of some conjunction of,
( 392) such tendencies, according to the plan of organization of the character ; and that, when we thus show it to be an instance of conation or appetition conforming to the most general laws of appetition, we do all that as men of science we can be called upon to do.