53/ ^7/ iarntll IRaimsitg Jitatg 4.,..//3>-5'fi nj.H.lmJ' THE NATURE OF EMOTION A THESIS Accepted by the Faculty of Cornell University I FOR THE Degree of Doctor of Philosophy DAVID IRONS, M.A. BOSTON QINN & COMPA-NY 1897 Cornell University Library The original of tiiis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924029198541 THE NATURE OF EMOTION A THESIS Accepted by the Faculty of Cornell University FOR THE Degree of Doctor of Philosophy DAVID IRONS, M.A. BOSTON GiNN & com;pa.ny 1897 A US '^70 j'%-4>-^:Chs- fj CONTENTS. CHAPTER I. Need of an inquiry into the nature of emotion — Explanation of tlie present diversity of opinion on this subject — Introspective examination of concrete cases — Emotion contrasted with cognition — Emotion distinguished from pleasure- pain (i) as reaction, (2) as dependent on special conditions, (3) as involving qualitative distinctions, (4) as factor in conduct — The distinction between emo- tion and conation — Summary of results. CHAPTER II. The identification of emotion with pleasure-pain — Emotion as impulse — The analysis of emotion into pleasure-pain and conation — The James-Lange theory — Professor Dewey's position — The views of Professors, Baldwin, Ribot, and Kiilpe — Mr. Marshall's theory — Emotion as ' excitement ' — Emotion as explicable in terms of pleasure-pain and cognition — Professor Wundt's theory ^^ Conclusion. CHAPTER III. Method to be followed in determining the primary emotions — Objections considered — The tentative nature of the inquiry — Satisfaction and dis.satisf ac- tion as the emotional elements in joy and grief — Discussion of anger — Fear in its various forms — Definition of hope — Ill-feeling in all its forms — Kindly feeling (i) as race characteristic, (2) as exemplified in gratitude, pity, affection, etc. — Emotions of repugnance — Admiration and scorn — The self-feelings — The so-called 'intellectual emotions' — The aesthetic, moral, and religious feel- ings — Conclusion. Note. — The references in the headlines to volume and number apply to the Philosophical Review, from which this thesis has been reprinted. The paging has been made consecutive throughout. THE NATURE OF EMOTION. THE object of this article is to set forth, as fully as space permits, the arguments in favor of a position which was mentioned incidentally in the January number of Mind, 1894. An inquiry into the nature of emotion cannot be deemed altogether superfluous at the present time. While most psy- chologists agree in maintaining that the phenomenon is expli- cable in terms of more primary aspects of mind, they appear to be wholly at variance in regard to the elements from which the supposed product is derived. Emotion has been described as ' a coalition of sensational effects with one another and with ideas ' (Bain) ; organic sensation (Lange) ; pleasure-pain, pure and simple (Horwicz); a fusion of feeling and organic sensation (Kiilpe); pleasure-pain in combination with cognition (Lehmann, Hoffding) ; a mass of sensuous and representative material with a predominant affective tone (Sully) ; tendency to activity (God- fernaux) ; a strong primary feeling of pleasure or pain, accom- panied by a change in the course of ideas, and strengthened by the feelings associated with the ideas called up (Wundt); rank feeling of excitement (James ^); a complete psychosis involving cognition, pleasure-pain, and conation (Ward). Few writers, moreover, keep consistently to one point of view. Bain begins with a combination of various sensations and ideas, though the process of ' transformation ' is admitted to be some- what obscure. It appears, however, that the sensational stimu- lus is not always necessary, and that love, anger, and possibly fear, are "original fountains of sentiment or feeling."'^ Then emotion is classed, along with Intellect and Action, as one of the chief divisions of the mental powers.* Finally, we find that particular emotions are defined in terms of pleasure-pain or impulse.* Wundt, after expounding the theory already 1 Psychological Review, i, 5, p. 525. • Ibid., p. 95. 2 Emotions and Will (3d ed.), p. 73. • Ibid., pp. 142, 173, 177. THE NATURE OF EMOTION. 3 referred to, tells us that "the universal animal impulses are the earliest forms of emotion." 1 Hoffding and others show the same tendency to depart from their original position, and to identify emotion more or less closely with impulse.^ The root of the trouble seems to be that emotion is explained away before a serious effort is made to ascertain with accuracy its real nature. In attempting to rectify this omission, we must begin with an examination of the fact as it appears in consciousness. It is evident that direct observation alone affords absolutely reliable information with regard to the quali- tative distinctions of psychical states. We can, of course, sup- plement the results obtained by the use of this method. If, for example, the phenomenon under investigation not only seems fundamentally distinct from other aspects of mind, but is found in addition to have special conditidhs and effects, the testimony of introspection gains in objective validity. The application of the direct method to emotion is not so hopeless an undertaking as one is accustomed to imagine. All emotions do not involve commotion, and the less violent states can be accurately observed. The simplest and most natural way of opening the inquiry will be to take concrete instances. When hate, for example, is aroused, into what constituents can the total consciousness of the moment be resolved } Under certain circumstances we can distinguish at once the cognition of the object, the pain of the injury or series of injuries, the awareness of certain organic disturbances, and an impulse towards a definite course of action. But the statement that we hate a person is not equivalent to the assertion that we are pained, know the cause, have various organic sensations, and feel impelled to act in a particular manner. Nor can hate be identified with any one of these elements or any combination of them. It implies that, in con- sequence of his actions with reference to us, we feel disposed towards some one in a certain way, and that, as a result of this feeling-attitude, actions of a special sort and these alone appeal 1 Menschen- und Thierseele (1892), pp. 421-2; Creighton and Titchener's trans., pp. 386-7. ^ Psychology, English trans., p-,235. 4 THE PHILOSOPHICAL REVIEW. [Vol. VI. to us. The ' plain man,' in describing his state of mind, will say that he has bad feelings towards the individual, ' feels as if he could ' inflict on him all manner of harm. Similarly, liking or affection for a person is not the idea of the object, the pleasure received, the expectation of further benefits, the aware- ness of various physical processes, or any combination of these. Here again we find a disposition towards the object which cannot be expressed in terms of mere intellectual or practical attitude. It is a feeling in reference to the object which is dependent on the intellectual attitude, and conditions the practical. The observation of every emotional state will yield the same result, and we seem to have here a mental fact fundamentally distinct from the ultimate aspects of mind commonly recognized. ^ The existence and nature of this distinction will become more evident if we imagine an individual who is composed psychically of cognition, pleasure-pain, and conation. Such a being, when injured, might judge it advisable to retaliate, on the ground that if he does not he may be attacked again. Or, on general principles of abstract justice, he might deem it right to act towards the offender as the latter has acted towards him. But he would have no intense ill-feeling towards the aggressor, prompting him, apart from all deliberation and calculation of consequences, to adopt a particular line of conduct. His atti- tude would be similar to that of the ideal judge in the act of sentencing a criminal according to the general principles of law. We might assume that he had inherited an instinct of retalia- tion, or that he was so constituted as to take pleasure in giving pairi in return for injury. ^ But in neither case would his action 1 I am aware that few psychologists identify emotion with any one of the men- tal aspects mentioned above. This circumstance, indeed, justifies the present investigation. Taking my stand on the difference which is almost universally admitted, I seek to determine what it is before proceeding to estimate its significance. 2 As a matter of fact, this pleasure can only be explained by reference to emo- tion. Retaliation is pleasant when hate is in the ascendant; otherwise there is no pleasure in the mere act of inflicting pain. The delight in cruelty as such is but an apparent exception to the rule. Habitual indulgence in an emotion will create a special tendency to action. If this propensity be firmly established, the No. 3-] THE NATURE OF EMOTION. 5 be accompanied by any malevolent feeling towards the object. Or, again, he might be moved to activity by the pain involved, but in that case he would seek immediate relief.^ It cannot be asserted that he would invariably attain this end by inflict- ing suffering on the author of the pain ; but if he resorted to retaliation with this purpose in view, he would proceed with the same lack of ill-will which the hunter displays when he kills an animal to relieve the pangs of hunger. As this imaginary individual would be destitute of all malev- olent feeling, he would likewise be incapable of affection or gratitude. If he were agreeably affected by one of his fellows, he might cognize the pleasure-giving agent as a worthy kind of being, and return the favor from a sense of justice, or with a view to future benefits. His action, however, would not be prompted by any kindly disposition towards the individual con- cerned. He would regard the latter in a purely intellectual manner as a series of sensations involving, according to all probability, a personality like himself — a personality which has been the means of causing him enjoyment. On the other hand, he would turn on himself the same cold, cognitive gaze. His actions would give him pleasure if they harmonized with his interests or ideals, but there would be no responsive pride or self-complacency. It is evident that a being of this sort would lack one of the springs of action which plays so large a part in our life, both need of satisfying it will of itself be a sufficient stimulus to activity. Purely malevolent action, therefore, though it may seem to be independent of malevo- lent emotion, is yet derived from it. It can be shown also to bear traces of its origin, for (i) ill-feeling tends to appear during the course of the action, and (2) the individual who derives pleasure from cruelty is delighted if he can find or make some pretext which will rouse ill-feeling against his victim. The habit of malevolent action is strengthened also by such incidental accompaniments as the pleasures of power and contrast. If the position here advocated is not adopted, we must' simply assume, as ultimate and inexplicable facts, pleasure in paining others and pleasure in giving pleasure. This in itself is rather a lame conclusion, and it becomes still more unsatisfactory when we note that it renders insoluble the problem of explaining why one of these principles comes into play at one time and its opposite at another. 1 In regard to this statement with reference to pain-prompted action, see Ward, Encyclopaedia Britannica, xx, p. 7 1 ; Spencer, Principles of Psychology, i, p. 280, etc. 6 THE PHILOSOPHICAL REVIEW. [Vol. VI. directly and indirectly. And though he might perform the actions to which emotions prompt, he would never be driven along, regardless of everything, simply because one course alone seemed to satisfy his feeling at the moment. Further, there would always be this fundamental difference between him and the actual human being, that his attitude to other persons and things could only be intellectual and practical.^ Since pleasure-pain ends in the self, and has no outward reference,^ when he faced a person or event he could only cog- nize the fact in a certain way, and act accordingly. The nature of the cognitive attitude would undoubtedly be influenced by the pleasure-pain received, but it would still remain purely cognitive. When we thus eliminate everything but cognition, pleasure- pain, and conation, the distinctive character of' emotion comes to light more prominently. I have used the term ' feeling- attitude ' to indicate, not to define, this apparently unique aspect of mind. The word 'feeling' expresses subjectivity and diffu- sedness. Emotion is subjective in much the same sense as pleasure-pain. It is a centrally initiated reaction, however, while the latter i^ pure receptivity. Briefly, the one is subjectivity as reaction ; the other is subjectivity as receptivity. The word ' attitude ' is employed to mark this distinction, and to empha- size the fact that emotion, in virtue of its character as reaction, has what might be termed ' objective reference.' We may say, then, that emotion is the subjective response which appears when we react in view of a situation instead of being passively affected by it. This response is not a mere impulse or tend- ency to act ; it is a mood or state of feeling in regard to the object, on account of which special modes of conduct appeal to us with a force they do not possess on other occasions. Im- pulses and volitions arise in consequence, for particular motor- ideas thus become dominant. These ideas gain this ascendency, 1 This would be more generally acknowledged if psychologists would substitute their various formulas for the ordinary terms used in reference to the emotions. The fact is surreptitiously introduced under cover of the word, and statements thus appear plausible that would otherwise seem rather strained. 2 Bradley, Mind, i888, p. 3 ; Hoffding, Outlines of Psychology, p. 234. No. 3.] THE NA TURE OF EMOTION. 7 however, simply because they are in harmony with that phase of subjectivity which is prominent at the time.^ In order to determine the nature of emotion more accurately we shall contrast it with the other aspects of mind in turn. In this connection cognition can be treated briefly. A cogni- tive state is not merely a subjective event with a place in the series of psychical occurrences. It reports the existence and nature of something, and thus has a reference beyond itself. This reference is its most prominent feature, and its factual nature as a psychical event falls into the background. Emo- tion, it is evident, has not the objective reference which per- tains to cognition. It has the outward direction which is characteristic of all reaction. Moreover, it is not so much an isolated activity as the actual being of the individual in a cer- tain phase or aspect. For this reason its factual character as a psychical event is obtruded on consciousness. Naturally this awareness of subjectivity is not so strong as in the case of pleasure-pain, for there is an outward direction and a fixing of the attention on some object. This consciousness is always present, however, except in a few exceptional cases, and even then the obliteration is but momentary. This leads naturally to the discussion of the contrast between emotion and pleasure-pain, for the latter is also subjective in a special sense. It is pure receptivity, however, and thus lacks that outward direction which emotion as reaction possesses. It is mere result, an effect coming from the object and ending in the self. Emotion, on the contrary, is an attitude we adopt. We are pained or pleased by something, and have emotion towards something. The distinction is embodied in the struc- ture of language. That emotion has an outward reference has frequently been admitted,^ but the significance of the fact has not been fully ^ What the agent cannot do in 'cold blood' seems to him the only thing to do when his passions are aroused. 2 Cf. Hume, Treatise (Green & Grose ed.), ii, pp. 78, 121 ; Hbffding, Psychology, p. 234; James, Principles of Psychology, ii, p. 313; Lehmann, Hauptgesetze des Gefiihlslebens, pp. 19, 11 6-7. 8 THE PHILOSOPHICAL REVIEW. [Vol. VI. recognized. It is often supposed, for instance, that a combina- tion of cognition and pleasure-pain explains everything. If simple coexistence be all that is meant, it would be easy to show, on the evidence of introspection, that emotion is not merely the simultaneous presence of the elements in question. On this view, moreover, it is difficult to explain how pleasure- pain and cognition can be present together while emotion is absent, for this is the state of affairs when we are ' hurt,' i.e., pained but not angry, at the conduct of a friend. If, on the other hand, some process is implied, it would surely be better to say that when the factors mentioned are present they con- dition the appearance of something else. If we treat pleasure- pain and cognition, not as antecedents or conditions, but as constituents which enter into some quasi-chemical combination, emotion would have a double claim to be regarded as a unique element of consciousness. As product it would have a charac- ter all its own, and the process would be miraculous. That emotion is a central reaction, and in this respect dif- ferent from pleasure-pain, is another position which has been admitted ^ without a full appreciation of its consequences. As it is of vital importance, it will be well to indicate the argu- ments in its favor. In the first place, it is obvious that emotion is subject to that law of activity which we call the principle of habit. The more frequently an emotion is indulged in, the smaller is the stimulus necessary to call it forth. When we say a person is timorous, irascible, or affecticfnate, we mean that he has a certain emotional habit, in virtue of which fear, anger, or affection is readily aroused. Such habits constitute the temper- ament or disposition of the individual, and may be acquired or in- herited. As temperament is simply tendency to reaction, it is easy to see how it can be inherited. Now pleasure-pain does not become habitual, in the sense that it becomes by repetition more easily excited. On the contrary, it comes under the influ- ence of the principle of habituation (which is a very different thing), and thus tends to become weaker when repeated. Further, while the circumstances are in themselves emotion- ' Cf. Sully, Human Mind, ii, pp. 91, 93. No. 3-] THE NATURE OF EMOTION. 9 ally exciting, emotion may be inhibited, or at all events appre- ciably affected, by a consideration of the propriety of indulging it. Anger may be checked or totally inhibited by a perception of the consequences it might entail, and any passion tends to subside if the agent feels that he is making himself ridiculous. On the other hand, no abstract reflections on the uselessness of the thing will get rid of the pain of toothache, if the nervous agitation continues. No general consideration of the evil results of pain will affect the disagreeable effect produced by an ill-proportioned building so long as the ugly object remains within the field of vision. We can influence pleasure-pain only by dealing with its conditions, whereas we can restrain emotion when its normal conditions are present. And if certain points of view have become habitual, certain emotions will be habitu- ally suppressed on all ordinary occasions. It is scarcely neces- sary to point out that a merely passive effect cannot be con- trolled in this way, and that a reaction alone can be affected by a perception of the consequences it involves. In view of these fact-s, it is significant to observe that we feel more responsible for emotion than for pleasure-pain. We may be ashamed of being angry, but we are never really ashamed of being pained. This is inexplicable save on the assumption that emotion is reac- tion, for we can identify ourselves only with our own activity, not with an effect imposed on us from without. It is worthy of note in this connection that individuals differ much more in emotion than in hedonic sensibility. Finally, attention may be drawn to certain phrases and modes of speech. We talk of ' instinctive ' ^ fear or dislike, ' involuntary ' or ' grudging ' admir- ation, 'unmotived' hate, 'unbridled' passion, 'outburst' of emotion. This language would be absurd if emotion were not a reaction, and it is absurd if applied to pleasure-pain^ We cannot speak of an ' unmotived ' pain, or an ' outburst ' of pleasure. We are justified in concluding, therefore, that pleasure-pain is simply the way in which things affect us, while 1 It is the feeling which is Instinctive, not any action. The feeling may arise instinctively in circumstances which render a new line of action necessary. In such cases, of course, there can be no instinctive action, and in other cases all Siction appropriate to the emotion may be inhibited. lO THE PHILOSOPHICAL REVIEW. [Vol. VI. emotion is the manner in which we react. The latter is an element of our own character, while the former is mainly an expression of the psychical nature which we share with others. If we turn now to the conditions of the two phenomena we shall find that they are entirely distinct, as might indeed be inferred from the results already attained. Pleasure and pain depend upon harmony i and discord respectively. Lehmann maintains that anything causes pain which is in conflict with the conditions of mental or physical existence, and that the opposite holds in the case of pleasure.^ This is true, but it does not go far enough. Many things affect us agreeably or the reverse, merely in virtue of their relation to the aspect of our nature which is for the time being in evidence. If we are gloomy, gayety displeases us ; and the same action will cause pleasure or pain according as hate or kindly feeling is dominant. As Shaftesbury puts it, "the man in anger has a different happiness from the man in love." ^ While mere discord and harmony are thus the conditions of pleasure-pain, they do not determine emotion. Whatever is at variance with the individual's interests or wishes causes pain, but anger does not appear until he feels that he is ' injured.' Any bodily disorder is painful, but the sense of danger must be present before fear is aroused. Admiration and contempt imply the recognition of worth and its opposite respectively. Hate presupposes that the object of the feeling is cognized as a hostile personality. A situation becomes emotionally effec- tive, therefore, only if it is viewed under some general aspect. The detail as such, while possessing hedonic potentialities, is irrelevant. That emotion depends upon the point of view adopted, is obscured by the fact that it may become habitual. It may thus take the form of a reflex response to presentation, and, in extreme cases, may burst forth without any external provocation whatever. 1 ' Harmony ' is taken in its widest sense, to include all that is not discordant. Cf. Bradley, Mind, 1888, p. 7. 2 Die Hauptgesetze des Gefukhlebens, pp. 1 50—1 . » Characteristics (5th ed.), i, p. 296. No. 3-] THE NATURE OF EMOTION. II A variety of emotions is possible, therefore, in a situation that remains objectively the same. The sight of suffering causes pain, but, while the disagreeable effect remains, anger, pity, or contempt may arise according to the way in which the situation is regarded. Similarly while defeat is, and remains, unpleasant, we may admire an opponent if we attend to the skill displayed, or dislike him if he appears to us merely as the cause of a disagreeable result. Of course these different points of view bring new elements of discord and harmony into promi- nence, and thus, in an incidental and secondary way, condition pleasure-pain as well. On the other hand, they influence emo- tion directly. In the one case, they change the object which stands in relation to us, and thus indirectly affect its hedonic power ; in the other, they supply a reason and justification for a particular reaction. We have seen that an emotion may be inhibited, by a vivid perception of the effects it might produce, even when its normal conditions are in operation. This possibility of direct control does not exist in the case of pleasure-pain ; and it thus illustrates in a striking way the difference in^he conditions of the two phenomena. In certain cases, also, an emotion can be accounted for only on the ground that a need was felt for reacting in that way. As Bain remarks, "the irascible temper in a state of surcharge does not need an actual offender" ;i " the temperament overflowing with tender emotion finds many things to love." 2 It is important also to note that the same consideration may affect emotion and pleasure-pain in different ways. That one has been injured by a friend, is a reflection that may check anger or ill-feeling, while adding to the pain. Finally, as we have already pointed out, emotion itself may be in one sense a condition of pleasure-pain. It is at times the self in relation to which things are harmonious or discordant. Since hedonic effect is determined by harmony and discord, and the latter can differ only in intensity, it is natural to suppose that pleasure and pain have only quantitative differences. This is the position which is most generally accepted by psychologists 1 Emotions and Will, p. 189. ^ Ihid., p. 281. 12 THE PHILOSOPHICAL REVIEW. [Vol. VI. at the present day .^ On the other hand, evidence can be adduced to prove that emotion exhibits qualitative distinctions. First there is the testimony of introspection, and in a case like this, as Kulpe maintains,^ it is to introspection that the final appeal must be made. So far as direct observation is concerned, it may confidently be asserted that fear, hate, contempt, are as distinct from one another as blue, red, green. In both cases the assertion of qualitative distinction rests on the same basis, and has the same justification. Further, while every pain may be set down to discord and every pleasure to harmony, the pri- mary emotions must be referred to conditions that are in each case specifically distinct. The recognition of worth, for instance, is entirely different from the sense of injury or the conscious- ness of danger. What holds for the conditions holds for the effects. While pleasures or pains always prompt to activity of the same sort,^ each primary emotion gives rise to a special mode of behavior. There is nothing in common between the actions due to anger, fear, contempt, admiration, hate, and affection. It is also noteworthy that the different emotions are marked off by special names, while the various pleasures and pains are not thus distinguished from one another. Ordinary language is not an infallible guide in matters of detail, but when it presents us with a broad general fact of this sort its testimony cannot be entirely ignored. It is scarcely necessary to add that emotions cannot be classed simply as pleasurable or painful. No emotion has the same hedonic character throughout. A certain degree of fear is stimulating, and therefore pleasurable ; and an outburst of rage may be distinctly agreeable.* Affection may be painful in the sense that it may be accompanied by greater pain than pleasure.^ While hate in its initial stages involves a large element of pain, it may in the end become exclusively pleasurable. ' The thought of revenge is sweet,' especially when ^ Cf. Ward, Encyclopaedia Britannica, xx, p. 71 ; Bradley, Mind, 1888, p. 2; Sully, Human Mind, ii, p. 7 ; Lehmann, Hauptgesetze des Gefilhlslebens, p. 124; Kiilpe, Grundriss der Psychologic, p. 246. ^ Grundriss, p. 333. ' Cf. Ward, Encyclopaedia Britannica, xx, p. 71. ♦ Sully, Human Mind, li, pp. 94, 95. ' Cf. Horwicz, Psychologische Analysen, ii, pp. 448, 460, No. 3-] THE NATURE OF EMOTION. 13 backed by the consciousness of power, and the emotion heightens general activity and is thus in itself a source of pleasure. We may now consider the influence which pleasure-pain and emotion exert on action. If a state of consciousness is pleas- ant, we seek to retain it ; if painful, we seek to get rid of it.^ When we act from pleasure or pain, we act in reference to them. An effect on our own subjective state is invariably the end in view. Can we apply this in the case of emotion > Is vengeance desired because it puts an end to hate.' Can a parent be said to make sacrifices for his children's benefit in order to maintain his affection for them ? Is it true that a patriot lays down his life for the purpose of preserving his love of country .'' This seems somewhat absurd, and the reason is not hard to find. We have already pointed out that an emo- tion has no constant hedonic character. It may be pleasurable at one time and painful at another. Yet the action which it prompts is always the same. Whether hate be painful or the reverse, it always impels the agent to actions of hostility. This proves conclusively enough that emotion is a principle of action entirely different from pleasure-pain in its mode of operation. The distinction is specially prominent when the two principles come into direct conflict. Pity is thus at variance with the pain which is always associated with it. The former directs atten- tion to the disagreeable object, while the latter moves us in the opposite direction, and impels us to banish the unpleasing spectacle from consciousness. Moreover, the intensity and extent of emotion-prompted action are often out of all propor- tion to the pleasure-pain involved. The actions of hate and affection alike usually bear no quantitative relation to the amount of pleasure or pain received or expected.^ It is not difficult to ascertain the exact nature of the distinc- tion which thus obtrudes itself. Pleasure-pain is purely sub- jective, concentrates attention on itself, and naturally enough arouses activity dealing with the hedonic condition of the agent. 1 Cf. Ward, Encyclopaedia Britannica, xx, p. 7 1 ; Spencer, Principles ^f Psy- chology, i, p. 280; Hoffding, Psychology, p. 274 ; Sidgwick, Methods of Ethics (4th ed.), pp. 44, 45. ^ Sidgwick, Methods of Ethics, p. 50. 14 THE PHILOSOPHICAL REVIEW. [Vol. VI. Emotion, on the contrary, is an attitude or disposition towards something, directs attention outward, and therefore impels to action that has reference to an object. When the agent is entirely under the influence of emotion, he acts as he feels dis- posed towards the object, just because he is so disposed, and for no other reason. He is concerned with the object alone. The subject is in the background, the object all in all. The malevolent passions are as disinterested as the others. We ' lose ourselves ' in hate as in love. When hate is dominant the conduct of the individual is unselfish in precisely the same sense as it is unselfish when he is impelled by the opposite emotion. He may know that he is acting contrary to his own interests, but that does not affect him so long as hate retains its power.i Several writers have emphasized this characteristic of emo- tional action. Shaftesbury contends that interest does not govern the world, and that it is hard to believe that nothing is done in pure good-nature or kindness.^ Hutcheson maintains that there is something in our nature which determines us to actions " without any conception of them as good or as the means of preventing evil." ^ He analyzes anger, and finds in it, besides the desire of obtaining reparation of the wrong and security for the future, " a propensity to occasion misery to the offender, a determination to violence, even when there is no intention of any good