The Future of the Brain

Scientific Research & Outreach into Politics, Education & Society

Susan Greenfield

As Professor of Synaptic Pharmacology at Oxford University, Susan Greenfield leads a multi-disciplinary team investigating the physical basis of the mind and its implications for our understanding of human behaviour, work and society. Through her reputation as a leading scientist, author and thought leader in the dynamics and evolution of the human brain, Susan Greenfield has also extended her reach to influence and inform other aspects of our society.

Susan Greenfield: http://www.susangreenfield.com/

Being Properly Affected: Virtues and Feelings in Aristotle’s Ethics

By: L. A. Kosman

I

A moral virtue, or as we might say, a good state of character, is for Aristotle an established disposition for free and deliberate conduct of the right sort, a hexis prohairetikē, as he puts it. (1) In providing an account of the moral life in which the concept of a virtue or state of character is central, Aristotle reveals what is clear throughout the Ethics: that he, like Plato, thinks of the question of moral philosophy as not simply how I am to conduct myself in my life, but how I am to become the kind of person readily disposed so to conduct myself, the kind of person for whom proper conduct emanates characteristically from a fixed disposition. Of course the good life is a life of activity – a life, that is, in which such dispositions are realized and not simply possessed by persons of worth. Otherwise a perfectly desirable life might be spent asleep. But the good life is one whose activities are not simply in accord with the virtues but are the appropriate realizations of these virtues, and the virtues themselves are ready dispositions toward these activities. The good person is not simply one who behaves in a certain way, but one who behaves that way out of a certain character.

II

When we recognize the Aristotelian virtues to be dispositions toward deliberate and proper human conduct, it becomes tempting to take Aristotle to be thinking in terms of human action, of the virtues as fixed tendencies toward modes of praxis. But is that right? It is clear that no thoughtful reading of the Ethics can help but note the centrality of praxis in Aristotle’s moral theory and its important connections to the other key notions in that theory – for example, responsibility, choice, and practical reasoning. But virtues are not in Aristotle’s view dispositions solely toward modes of acting. Throughout his discussion of the moral virtues, and practically in his earliest account of them in Book 2, Aristotle makes clear that the activities for which virtues are dispositions are of two sorts, actions and feelings, praxeis kai pathē. He rarely mentions virtue in Book 2 with respect to action alone, but rather in terms of this dual phrase. The first account of the notion of disposition with which moral virtue will be identified, indeed, is exclusively in terms of feelings: kath’ has pros ta pathē echomen eu ē kakōs. (2) So the virtues are dispositions toward feeling as well as acting.

Acting and feeling are not simply two modes of human conduct among others. This can be seen if we use the older and more etymologically parallel English renderings of Aristotle’s praxis and pathos and describe the moral virtues as dispositions toward action and passion – toward characteristic modes of conduct, in other words, in which the virtuous person acts and is acted upon – is the moral subject, as it were, of active and passive verbs.

A view of feelings and emotions as passions or affections – that is, as instances of a subject being acted upon, of an agent, so to speak, as patient – is embedded in the very word which we translate by “feeling” or by “emotion”, the Greek word pathos. This word, like the verb paschein, of which it is a derivative form, has an earliest sense of what is experienced or undergone by way of misfortune or harm – what is, as we say, suffered – but comes subsequently, as with “suffer”, to have a general sense simply of what is experienced, a mode of a subject’s being acted upon. (3)

Insofar as Aristotle sees fear, anger, desire, pleasure, and pain as pathē, as passion, he views what we would call feelings or emotions as modes of a subject acted upon. This fact is revealed in the list Aristotle offers us of emotions with which moral virtues are concerned and in which there can be excess, deficiency, and right measures. The majority of items on this list are described by passive verbs; in thinking of fear, anger, pleasure, or pain, Aristotle is thinking of being frightened, being angered, being pleased, being pained. When I am afraid, something is frightening me; when I am angry, something is angering me. When in general I am experiencing an emotion or feeling of the sort which Aristotle would call a pathos, something is affecting me; I am being acted upon in some way, where the concept of being acted upon is reciprocal to that of my acting in some way. (4) Actions and feelings are thus for Aristotle modes of human being – action and passion – seen in terms of reciprocal concepts basic to our understanding of entities in general, the concepts of acting and being acted upon.

The opposition between acting and being acted upon, that is, the praxis/pathos opposition, is, to be sure, peculiar to human activity, because praxis is according to Aristotle peculiar to human activity. But it is only a special instance of a more general structural duality, that of poiein and paschein, doing and being done. (5) This dualism is encountered early in and throughout Aristotle’s writings, from its special appearance as an apparently reciprocal pair of relatively minor modes of being in the Categories to the important cosmological role recognized in the treatise On Generation and Corruption. (6)

This more general opposition of poiein/paschein, insofar as it governs and structures the most basic forms of being, understood as activity, is thus a central and fundamental structural principle of human activity. “Activity” here must take on a sense broad enough to include both acting and being acted upon – must, that is, include modes of active human being in which the human individual is both subject and object of the action, both agent and patient.

The notion of including as an aspect of human activity what an individual suffers, what is done to an individual in addition to what the individual does, may go against our intuitions. It is only an individual’s actions, we might feel, which truly belong to the being of a human individual in its most basic sense. This intuition is something to which we shall have to return. But it should not blind us to the fact that the affections and passive sufferings of entities seem to be among the authentic characteristics which such entities have, and this is true for both human and nonhuman beings.

If the kinds of human activity now understood in this broader sense include not only what are, more strictly speaking, instances of human action but also instances of human passion in the sense of being acted upon, and these latter are understood as paradigmatic instances of feelings or emotions, then Aristotle’s moral theory must be seen as a theory not only of how to act well but also of how to feel well; for the moral virtues are states of character that enable a person to exhibit the right kinds of emotions as well as the right kinds of action. The art of proper living, we should say, includes the art of feeling well as the correlative discipline to the art of acting well.

III

Moral philosophy, then, should count as an important question that of sentimental education and should recognize the proper cultivation of our feelings as within the domain of our moral concerns. But how could this be? It appears to be a distinction between our actions and our passions that actions are within our control, whereas passions are not; we are the initiating principle of what we do, but not of what is done to us. Aristotle seems to make exactly this point in claiming that we are angered or frightened, for example, not by choice – aprohairetōs, as he puts it. (7)

But if we give allegiance to the meta-ethical dictum that “ought” implies “can”, then it seems problematic how a moral philosophy could concern itself with questions of our feelings in contrast with question of how we ought to act in response to or in light of those feelings. What is, as Aristotle would say, eph’hēmin – within our power – is surely what we do, but not what is done to us.

One way in which this objection might be met is by pointing out what I have emphasized earlier: the important role of the concept of virtue. Since we are primarily called upon by a moral theory not to act and be acted upon in certain ways but rather to be kinds of persons disposed so to act and be acted upon, the question of what feelings are or are not immediately within our powers may be moot.

But could a moral theory be concerned with sentimental education even in this sense, namely, with the cultivation of proper sentimental dispositions? For in what sense might such dispositions themselves be said to be within the domain of choice or prohairesis? Why, in other words, wouldn’t the question be pushed one step further back? And how then could we be said in the first place to have disposition to be acted upon rather than to act in certain ways? Let me first say something preliminary about this latter question.

Aristotle’s claim that virtues are dispositions toward feeling as well as action can be seen, in light of our recognition that feelings are passions, to rest upon a theory of potentiality which recognizes the existence of passive as well as active powers. For since a disposition is a power – that is, a potentiality that renders an entity capable of a mode of actual being – and since feelings are passions, the virtues in questions must be passive powers – the potentialities, that is, for being actively affected in a certain way.

That Aristotle has a general theory of passive powers is evident. The discussion of potentiality in Book 9 of the Metaphysics makes this clear (in the context of a discussion that once again reveals the reciprocal relation and basic dialectical unity of poiein and paschein). Passive potentialities, we there learn, are elements within an entity’s nature, not simply external to it, and are connected to other more obvious integral aspects of the entity’s being. To say that oil is burnable is to ascribe to the oil a power that belongs to it and is a potentiality for being affected in a certain way, for having something done to it, and is at the same time to link this capacity to certain positive states and characteristics of the oil. (8)

More obvious forms of potentialities with respect to being affected are those by virtue of which an entity is capable of withstanding or not suffering a certain affection. Aristotle often refers to such a type of power both in the discussion in the Categories and in that in the Metaphysics. (9) In both cases, however, we are considering powers exhibited by an entity for discrimination among affections, for being affected in this way and not that.

Another class of such powers, and one that for Aristotle is paradigmatic and important, includes the faculties of the soul. The perceptual capacities, and the faculties of reason and thought as well, are potentialities of the sensitive and intelligent subject to be affected in certain ways, to be acted upon by the sensible and intelligible forms of objects in the world. (10) When we think of them in this way, there is nothing particularly mysterious about these powers: they are simply the abilities to be open to certain affections and closed to certain others – the reciprocal capacities, we might say, of being discriminatingly receptive and resistant.

The doctrine of passive potentiality enables Aristotle to envision a state of character by virtue of which an individual has the power to be affected in certain ways, the capacity to undergo certain passions and avoid others. A moral virtue with respect to feelings or emotions is just such a capacity; it is the power to have and to avoid certain emotions, the ability to discriminate in what one feels.

IV

But this suggests only how there might be said to be powers to be acted upon in certain ways; what concerned us was whether such powers might be said to be dispositions involving choice. We still therefore need to be concerned with the question of choice, and that question may now be put this way: how is it possible for Aristotle to see a virtue as a disposition toward a feeling or emotion, given the following facts about choice? In his first characterization of virtue Aristotle denies that virtues might be feelings themselves by pointing out that a feeling is not the sort of thing that is chosen; our anger and fear are, as he might have said, “aprohairetic”, whereas the virtues are kinds of choice or at least not aprohairetic, not devoid of choice. But then we should ask: how could choice be involved in a fixed tendency toward that which does not involve choice, indeed for which we are in no wise praised or blamed? Won’t there have to be a radical asymmetry between the relation among dispositions, actualizations, and choice in the case of feelings and that relation in the case of actions? For in the latter case a virtue appears to be a disposition toward prohairetic acts; the actions themselves are instances of choice, and it seems to be only by virtue of this fact that the hexis or disposition which is a virtue with respect to these acts is said to be prohairetic. But in the case of feelings, virtues would, on the view we are considering, themselves be chosen but at the same time be dispositions toward actualizations that are not chosen.

One solution to this problem which I hinted at before would be to abandon the notion that virtues are dispositions with respects to feelings in the sense of being fixed tendencies to feel in a certain way. It is not that courage as a virtue with respect to fear disposes us to feeling fear in certain ways or in certain circumstances or to a certain degree but rather that it disposes us to certain actions with respects to and in light of our fear. Some support may appear to be given to this reading by Aristotle’s description of a disposition as a state by virtue of which we are ill or well disposed with respect to the emotions: kath’ has pros ta pathē echomen eu ē kakōs. (11)

This is the view of Joachim in his commentary on the Ethics; he writes:

In the development of the orectic soul there is hexix when the soul echei pōs (viz. eu ē kakōs) pros ta pathē: i.e. when a permanent attitude towards his emotions (towards any possible disturbances of his orectic self) has been reached – an attitude which expresses itself in actions which are either the right or the wrong responses to such disturbances. (12)

But this reading is surely wrong: Aristotle, in the passage in question, immediately goes on to explain:

if for instance, with respect to being angered we do so excessively or insufficiently [sphodrōs or aneimenōs: if we’re too violent or too easygoing] we’re badly disposed, but if moderately [mesōs] then we’re well disposed. (13)

Aristotle’s more detailed discussions of the virtues make clear that it is with respect to how one feels and not simply how one acts in light of one’s feelings that one is said to be virtuous. The courageous person is one who is frightened by the right things, in the right way, in the right circumstances, and so on, and who is not frightened when it is appropriate not to be. The temperate person is one who is pleased by the right things, in the right ways, and so on. The so-called “gentle” person (whose virtue Aristotle finds to have no proper name) is angered by the right things, in the right way, and so on. (14) There is here no indication that these moral virtues defined in terms of feelings are dispositions toward some range of actions appropriate in light of these feelings, and, on the contrary, every indication that they are dispositions toward appropriate feelings themselves.

What emerges in addition, however, is the recognition that these feelings are accompanied by concomitant actions. I have talked so far as though feelings could be understood as particular affects of an individual independent of anything else that might be true of the individual. In a sense this is exactly Aristotle’s theory and is why he says that pathē are those things with regard to which we are said to be moved. (15) But considered more broadly, there is no way to identify a feeling or emotion without taking into account (1) what we might call the cognitive element in emotions and (2) actions on the part of the agent which are characteristically and naturally associated with such feelings. (16) Fearing is related to fleeing, desiring to reaching for, anger to striking out at, in no accidental way. In each of these cases, a certain action or range of actions is connected to a pathos in some important logical sense. That connection is defeasible, but it is not a merely accidental connection.

These considerations suggest that there may be two elements to the actuality corresponding to any given virtue. A virtue is a complex disposition in the sense that its actualization is complex, and specifically in that its actualization consists of a characteristic set of feelings and a correspondent characteristic set of actions.

But if this is true, our discussion of virtues has been seriously misleading. The fact that virtues are dispositions with respect both to actions and to feelings ought not to suggest to us that there are a number of virtues that are dispositions with respect to actions and a number of virtues that are dispositions with respect to feelings, any given virtue being a disposition with respect to one or the other, but rather that a given virtue is a disposition with respect to a characteristic set of actions and feelings. These feelings are not, as in the view we have just been considering, merely the occasion for actions that are the proper realizations of the virtue; they are part of the concept of that virtue considered as a disposition. But neither are they the sole realizations of the virtue; they are part of the set of corresponding actions and feelings for which the virtue is a disposition.

V

With this understanding in mind, let us, as Aristotle would say, make a fresh start. Recognizing that virtues are dispositions toward feeling as well as toward action, we found ourselves perplexed by Aristotle’s apparent claim that virtues involve choice, while feelings do not. It appeared to us that there was an unexplained asymmetry in Aristotle’s understanding, such that virtues might involve choice without their appropriate feeling-realizations involving choice, even though virtues seemed to involve choice in the first place because they are dispositions toward actualizations that are deliberate and chosen. Is it possible to make sense of the notion that a virtue involves choice even though the feelings that are its realization do not? Could a person be said to be courageous in a deliberate and chosen manner without it being the case that the feelings which such a person characteristically exhibits by virtue of being courageous are themselves chosen?

Consider a parallel situation in the case of actions. A person may exhibit a vice such that the actions performed as a result of that vice are involuntary considered in themselves, and we may still want to say that the vice itself is voluntary. Thus someone, as Aristotle notes, may in one sense not be responsible for individual actions performed while he way intoxicated or ignorant of what he was doing or in general acting in accordance with some trait of character that made it impossible or extremely difficult for him to do the right thing, and yet in another sense clearly be responsible, at least for being in the first place unaware or drunk or in general of such and such a vicious character. (17)

These examples provide only a partial parallel to our case. In the first place, they concern the somewhat weaker notion of voluntariness, not that of choice. In the second place, they concern only instances of vice and the concomitant breakdown of moral action. Nowhere do we find an instance of Aristotle’s characterizing as voluntary a virtue whose resultant actions could be said to be involuntary. Finally, the examples that come most readily to mind – case of drunkenness, ignorance, or negligence – concern what seem to be in a sense “meta-vices”: states that condition our general capacity for virtuous conduct. As a consequence the relationship between the action and the vice that occasions it is not a straightforward relationship of disposition to realization. Only under a partial description could an act of negligence be said to be the actualization of the vice of negligence in the same sense in which an act of courage is the realization of an agent’s courageous disposition.

But we could extend the basic point to cases of specific vices. It seems a part of Aristotle’s views on moral responsibility that vicious actions are in some important sense not chosen, and indeed under certain descriptions not voluntary, though the vices that occasions them are, and are therefore the responsibility, like their actions, of the moral agent.

This is an enormously complex and difficult topic in Aristotle’s thought, and an important one for Aristotle and for moral philosophy in general; the respect in which vicious acts are not willed but nevertheless our responsibility may be central to an understanding of the moral life (compare Kant). I introduce the question here only because it suggests that certain predicates which we might think of as features of actions and only derivatively of characteristics or dispositions relative to those actions may be applied independently to one or the other in a variety of ways. Most importantly we see here a sense in which predicates appropriate to dispositions (in this case the predicate of being voluntary) may apply to these dispositions not only with reference to their actualizations but also with reference to the mode of their acquisition.

It is a part of Aristotle’s moral theory which I have alluded to but not stressed that virtues are acquired. They are, as he says, not phusei, but neither are they para phusin: neither natural nor contrary to our nature. (18) This is what makes them dispositions – hexeis – rather than simply potentialities – dunameis. They must therefore be acquired, and their acquisition cannot be effected simply by an act of choice; we do not decide to be virtuous and straightaway become so. Virtues are cultivated and not chosen in any simple sense, for it is not as a direct result of calculation, deliberation, resolution, or any other relatively simple mode of human activity that we become courageous, temperate, or wise. We become these through a process of ethismos, or habituation, through the habitual acting out and embodying of those actualizations which the dispositions are dispositions toward. (19) This mode of acquisition is part of the logic of a virtue, because it is part of the logic of a hexix. It is therefore part of the logic of a moral theory which, as I have suggested, places the notion of virtue at its center.

Perhaps this sense in which we would want to say that a virtue is not chosen – think how easy the moral life would then be! – may have been what led us to say that a virtue’s involving prohairesis must depend on the actualizations of that virtue being prohairetic. But there now emerges an alternative way of thinking about the isse: instead of supposing that the quality of being prohairetic flows backward, as it were, from actualization to disposition, suppose we say that a virtue is prohairetic which is acquired through acts that are themselves prohairetic. What happens when we try this model?

Note first that there is a sense in which a virtue might be said to be chosen, and on a specific occasion. A person might decide on such an occasion to act virtuously and see that act as the first in a series designed to effect a transformation in her life to being virtuous. So she chooses on this occasion to be virtuous and so acts, acting virtuously now this time, now another time, now another time, until fixed disposition of virtue becomes a hexis that characterizes her moral life. Why shouldn’t we say in such a case that virtue has been chosen, from the beginning, and precisely because the acts that fix that virtue are chosen?

This notion accords well with the view of Aristotle’s we have just been looking at. On this view one becomes virtuous by impersonating a virtuous person, and in that impersonation, through the process of habituation, becomes the virtuous person whom one impersonates. The direction of self-constitution is seen as leading from actions to states of character as well as the other way around; indeed, we should perhaps say that the other way around is subsequent, is only a logical feature of the relationship between our character and the individual moments of our conduct once they have been established.

But how does this help us with the question of virtues and feelings? For our initial problem with feelings was that they are said by Aristotle not to be chosen; an account according to which virtues are prohairetic because the acts that fix them are prohairectic seems therefore to be of little help in understanding the relation among choice, virtues, and feelings. This question supposes that virtues whose actualizations are feelings may be acquired only through the direct and deliberate choice of feelings. But given our recognition that the actualization of a virtue is a complex and related set of actions and feelings, a different view is now open to us. A person may act in certain ways that are characteristically and naturally associated with a certain range of feelings, and through those actions acquire the virtue that is the disposition for having the feelings directly. Acts are chosen, virtues and feelings follow in their wake, though in logically different ways.

In this view the structure of becoming virtuous with respect to feelings reveals itself to be of the following sort: one recognizes through moral education what would constitute appropriate and correct ways to feel in certain circumstances. Once acts in ways that are naturally associated with and will “bring about” those very feelings, and eventually the feelings become, as Aristotle might have said, second nature; that is, one develops states of character that dispose one to have the right feelings at the right time. One does not have direct control over one’s feelings, and in this sense the feelings are not chosen; but one does have control over the actions that establish the disposition, the virtues, which are the source of our feeling in appropriate ways at appropriate times and in appropriate circumstances. Although we may in some narrow sense not be responsible for our feelings, we are responsible for our character as the dispositional source of those feelings.

This picture has much to recommend it. In the first place, it seems correct. My anger or jealousy may not be an emotion which I choose, and yet it may be true that I have become a person disposed to such anger or jealousy by a series of actions that would make it perfectly reasonable to describe my character as something I have chosen. In the second place, it points to what seem important differences between feelings and actions which become clear only when we talk about the modes of acquisition of virtues. Much of the picture I have sketched depends upon the fact that one can simulate an action but can only pretend to have feeling. Thus the mode of inculcation with regard to feelings must be by some other method than that of habituation. There is, in other words, no way to come to feel a certain way by practicing feeling that way, and this is precisely because in some sense our feelings are not in our control. But it is nonetheless possible to engage in a certain range of conduct deliberately designed to make one the kind of person who will characteristically feel in appropriate ways, at appropriate times, and so on. And in this sense, feelings are deliberate and chosen, since the hexis from which these feelings emanate are deliberate and chosen, since (in turn) the actions that lead to these hexis are deliberate and chosen, and deliberately chosen to make one the kind of person who characteristically will have the appropriate feelings.

VI

This picture seems to me on the whole correct and, broadly speaking, faithful to Aristotle’s version of the moral life, but it fails in important respects as a solution to the problems that initially perplexed us. Note in the first place how far we have come from our initial characterization of virtues as a disposition for deliberate and chosen conduct. That characterization interpreted Aristotle’s description of a virtue as a hexis prohairetikē to mean that the virtue was the agent’s ready capacity to act (in the broadest sense) prohairetically, not that the virtue had been acquired through some modes of prohairetic action. In the second place, the rather elaborate account into which we have been led seems curiously otiose; once we recognize that actions and feelings are linked together as the realizations of a virtue, why not simply say that a virtue is prohairetic because it is a disposition for those elements of its realizations which are actions? For there is no special problem, apart from the general problem of giving an account of what prohairesis is, about actions being prohairetic. Finally, the most important flaw in the account I have suggested is the following. The account depends upon actions that I lead to the establishment of a virtue but are not the expressions of an already established virtue being prohairetic. But this is not a view which Aristotle seems to hold. Such actions are voluntary, but it is by no means clear that they are for him in the fullest sense prohairetic. Just as acts are in a sense virtuous when they are acts of the kind which a virtuous person (one who performs such acts in the way that a virtuous person would perform them – that is, out of the fixed character that is virtue), (20) just so an act may be like prohairetic act, but not be one because it is not properly embedded in the larger context of the character and disposition of moral agent – because it is not a realization of that agent’s virtue. There is, Aristotle says, “no prohairesis without intelligence and thought, nor without moral character.” (21) Prohairesis involves, as we might say, not simply deciding, but willing, where the notion of will is sufficiently rich to demand reference to the larger context of central and properly integrated goals and habits of an agent’s moral life.

The recognition that choice for Aristotle is a concept governing not individual actions in a life divorced from this framework of goals and habits, but rather actions only as moments within a larger context of the character and intentions of the moral subject, reveals our picture to be inadequate as a solution to the problems I have raised. But it should, I think, contribute to the dissolution of these problems. For it should suggest the necessity of rethinking our initial denial that feelings may be chosen. That denial arose, I think, form our attending to a description by Aristotle of passions without reference to the context of virtue and character in which they occur. But why should we not be prepared to say that a person of steadfast and cultivated virtue who exhibits appropriate feelings in circumstances which he understands correctly and in which those are precisely the feelings which he would want to exhibit, taking into account the entire fabric of his goals, plans, and hopes for himself – why should we not be prepared to say that such a person has chosen those feelings?

Nowhere, I believe, does Aristotle say this. What we would like, but do not find, is an extension of the theory of deliberation and practical reasoning to account for the ways in which virtuous persons might be said to have the proper feelings which they have by prohairesis. Such an account would need to provide a sense in which we might be freed to feel what would be appropriate to feel by something like deliberation and choice, by some mode of coming to understand properly the circumstances in which our feelings arise, the place of these feelings and circumstances in our experience, and the ways in which we hold these circumstances and feelings in the larger contexts of our lives. In a sense, the theories behind certain religious traditions, psychoanalysis, and disciplines that promise self-transformation and self-mastery might be thought to represent attempts at such an account.

But any such account, I think, would have to recognize the primacy of praxis in the shaping and execution of our moral lives. It is this primacy that finally dominates the concerns of the Ethics. To say then that there is a strict parallelism for Aristotle between actions and passions would not be quite correct. On the one hand, he is led by the paschein/prattein structure and by the understanding of emotion as pathos to view virtue as a disposition equally for action and feeling, and as a consequence to recognize correctly the important place which our feelings occupy in the structure of our moral conduct. But on the other hand, the distinction between passion as something that happens to one and action as something that one does, coupled with the recognition that ethics must be concerned with prohairesis, leads him to turn his attention from feelings in the latter parts of the Ethics. It is not that he simply leaves the question of feelings out, but that their importance fades in the context of a particular theory of deliberation and choice and their place in moral conduct.

Nor is the case, however, as I have argued, that feelings are for Aristotle simply not chosen. The reason for this, I have suggested, is that choice for him is not a concept having to do with individual moments in an agent’s life, nor with individual single actions, but with the practices of that life within the larger context of the character and intentions of a moral subject, ultimately within the context of what it has become fashionable to call one’s life plan.

The question of moral choice in the deepest sense finally concerns questions of creating the conditions in which our actions and our feelings may be as we would wish them. These conditions include our states of character – the virtues – which we acquire through the complex practices of our moral life. So long as we consider moral questions in terms of individual moments in the agent’s life, we will not be able to understand this fact. This is why it should be clear that questions of virtue and feelings as moral categories are importantly connected.

This essay was published in Essays on Aristotle's Ethics, edited by Am

-->ēlie Oksenberg Rorty, University of California Press, Berkeley Los Angeles London

NOTES

1. Nichomachean Ethics 2.6. 1106b36; Eudemian Ethics 1.10.1227b8

2. Nichomachean Ethics 2.5. 1105b26; 2.2. 1104b14; 2.6. 1106b17, 1106b25, 1107a9.

3. See the article on paschō in G. Kittel, Theological Dictionary of the New Testament (Grand Rapids, Mich., 1964-1974), 5:904 ff.

4. Nichomachean Ethics 2.6. 1106b17 ff. Compare English afraid as originally the passive participate affrayed.

5. All this is on the assumption that prattein is a special case of poiein and not, as Aristotle comes to understand it, vice versa. Compare his uses of kinesis and energeia for a similar reversal of the common understanding.

6. Categories 9. 11b1 ff.; On Generation and Corruption 1.7. 323b1 ff.

7. Nichomachean Ethics 2.5. 1106a2.

8. Metaphysics 9.1. 1046a11 ff., 1046a13.

9. Categories 8. 9a23; Metaphysics 9.1. 1046a13.

10. On the soul 2 and 3. passim.

11. Nichomachean Ethics 2.5. 1105b26

12. H. H. Joachim, Aristotle, The Nichomachean Ethics (Oxford, 1951), p. 85.

13. Nichomachean Ethics 2.5. 1105b26 ff.

14. See the table of the virtues in Nichomachean Ethics 2.7. 1107b1 ff., and the specific discussions of virtues in Books 3 and 4.

15. Nichomachean Ethics 2.5. 1106a4.

16. Thus On the Soul 1.1. 403a25, and the discussion of the emotions in the Rhetoric. See W. W. Fortenbaugh, “Aristotle: Emotion and Moral Virtue,” Arethusa 2 (1969), 163-185.

17. Nichomachean Ethics 3.5. 1113b30 ff.

18. Ibid., 2.1. 1103a14 ff.

19. Ibid., 1103a28 ff.

20. Ibid., 2.4. 1105a26 ff.

21. Ibid., 6.2. 1139a33. See G. E. M. Anscombe, “Thought ans Action in Aristotle,” in New Essays on Plato and Aristotle, ed. Renford Bambrough (London, 1965).

Morality and emotions

By Martha Nussbaum

Emotions such as anger, fear, grief, envy, compassion, love and jealousy have a close connection to morality. Philosophers have generally agreed that they can pose problems for morality in a variety of ways: by impeding judgment, by making attention uneven and partial, by making the person unstable and excessively needy, by suggesting immoral projects and goals. The place of emotions in moral theories depends on whether they are conceived of merely as impulses without thought or intentional content, or as having some sort of cognitive content. Plato argued that emotions form a part of the soul separate from thought and evaluation, and moved, in the course of his writings, from a sceptical view of their contribution to morality to a more positive appraisal. Aristotle connected emotions closely with judgment and belief, and held that they can be cultivated through moral education to be important components of a virtuous character. The Stoics identified emotions with judgments ascribing a very high value to uncontrolled external things and persons, arguing that all such judgments are false and should be removed. Their cognitive analysis of emotion stands independent of this radical normative thesis, and has been adopted by many philosophers who do not accept it.
Modern theories of emotion can be seen as a series of responses and counter-responses to the Stoic challenge. Descartes, Spinoza, Kant and Nietzsche all accepted many of the Stoics’ normative arguments in favour of diminishing the role played by emotions in morality; they differed, however, in the accounts of emotion they proposed. Focusing on compassion or sympathy, Hutcheson, Hume, Rousseau, Adam Smith and Schopenhauer all defended the role of some emotions in morality, returning to a normative position closer to Aristotle’s (though not always with a similarly cognitive analysis).
Contemporary views of emotion have been preoccupied with the criticism of reductive accounts that derive from behaviourist psychology. By now, it is once again generally acknowledged that emotions are intelligent parts of the personality that can inform and illuminate as well as motivate. Philosophers’ views have been enriched by advances in cognitive psychology, psychoanalysis and anthropology. Feminist accounts of emotion differ sharply, some insisting that we should validate emotions as important parts of moral character, others that emotions shaped by unjust conditions are unreliable guides.

1 Morality and conceptions of emotion

Emotions are usually considered to be distinct both from bodily appetites, such as hunger and thirst, and from objectless moods, such as irritation and endogenous depression (see Emotions, nature of). Major members of the class have traditionally been love, anger, grief, fear, envy, jealousy, guilt and pity (compassion) (see Love; Moral sentiments; Rectification and remainders). Some philosophers have adopted an account of these experiences that makes them rather close to appetites after all: emotions are surges of affect or energy in the personality, unreasoning movements that push people into acting without being very much connected to their thoughts about the world. This view is frequently connected to the idea that emotions derive from an animal part of our nature, often by thinkers who do not have a high regard for animal intelligence.

Seen this way, emotions figure in morality only as forces that either advance or impede the purposes of moral judgment or will. They can be trained and to some extent conditioned, but in a relatively mechanical manner. Thus they will never be what makes a virtuous character virtuous, though a virtuous person can take pride in having exercised an appropriate degree of discipline over them (see Self-control; Virtues and vices §3).

Other philosophical theories about emotions hold that they involve interpretation and belief. An emotion such as grief is not simply a mindless surge of painful affect: it involves a way of seeing an object, an appraisal of that object as important, and the belief that the object is lost. Fear involves the belief that bad events are impending, and that one is not fully in control of warding them off. Such theories hold that changes in the relevant beliefs entail changes in emotion: one who learns that danger is not really at hand will cease to fear. When we scrutinize the beliefs involved in the emotions, it emerges that they are of two sorts: beliefs about what is happening in the world (Is the person really dead? Is the enemy really at hand?) and evaluative beliefs (Is the person really worth getting upset about? Is the danger serious?). Many of the evaluative beliefs derive from social teaching, some of it moral in nature. We learn what insults are worth getting upset about, what losses are serious, what damages are to be avoided.

Seen this way, emotions become intelligent parts of the moral personality, which can be cultivated through a process of moral education (see Moral education §3). Such a process will aim at producing adults who not only control their anger and fear, but experience anger and fear appropriately, towards the appropriate objects at the appropriate time in the appropriate degree. Merely self-controlled persons look to these theorists like those whose moral development is incomplete or imperfect. If we find them hating foreigners, but controlling their behaviour towards them, we will judge that there is some further moral work they should be doing before they can claim to be fully virtuous.

2 Emotions and character-based theories

All moral theories have some occasion to talk about emotions, but some focus on their cultivation more than others. Theories that focus on the calculation of utility will see emotions as forces that either promote or eclipse such calculations and the appropriate acts (see Utilitarianism). Theories that focus on duty can have a substantial place for an account of emotion, since emotions will be seen as motives that either promote or impede action in accordance with duty (see Deontological ethics; Duty). But if the account of duty is not supplemented with an account of virtue, it will not be clear how these motives should be integrated into the personality. Kant’s theory of virtue (1797) shows us that a duty-based view need not neglect emotions as elements in virtue (see Kant, I. §10; Emotions, philosophy of §3). Since this theory has a noncognitive conception of emotion, and also a rather negative view of the contributions made by emotion, it gives emotions a rather restricted role; but there is nothing to prevent a virtue-theory such as Kant’s from giving more positive attention to the cultivation of emotion.

Theories of character based on Aristotle and other Greek thinkers, however, have been Western philosophy’s most substantial sources for thought about the proper cultivation of emotion within morality (see Emotions, philosophy of §2). Aristotle’s norm of a reasonable person is one whose character is infused completely by the correct reasons for action, which have shaped all their motives and attitudes. Because he aims to describe the cultivation of a whole person and way of life, rather than simply to prescribe a list of duties, he has ample scope for discussing emotional self-shaping.

3 Ancient Graeco-Roman theories of emotion: Plato, Aristotle, Stoics

Plato argues in the Republic that the soul contains three distinct parts: the calculative part, the appetitive part, and the ‘spirited’ part (see Plato §14). This last part seems to be where Plato locates emotions such as anger, grief and fear. Although his account is cryptic, he suggests that the emotions differ from appetites in their responsiveness to changes in belief, and also in the fact that they are less brutish and more discriminating. But he refuses to construe them as belonging to the ‘part’ that performs evaluative reasoning. In normative terms, the Republic argues that all pity, fear and grief should be prevented from developing by a process of moral education that teaches young people that there is nothing for a good person to fear or grieve over; anger is apparently allowed to remain, but is channelled for military purposes. In the Phaedrus, Plato’s account of the emotional ‘part’ seems more positive: focusing on emotions of reverence and awe, he shows how this element in the personality makes a crucial contribution to virtue and understanding.

For Aristotle, emotions are combinations of a feeling of pleasure or pain with a belief (or perhaps a more rudimentary cognitive attitude, a seeing x as y). Fear, for example, combines painful feeling with the thought that there are bad events impending. The combination is not casual: the pain is pain at the thought of that impending danger. Therefore, changes in belief will change emotions. (Aristotle makes these arguments in Rhetoric, showing how an orator can manipulate the passions of the audience.) In normative terms, Aristotle argues that the virtuous person is one who has attained balance and appropriateness in emotion as well as action. A person who completely lacked anger at an insult to loved ones, for example, would be culpably deficient; but excessive anger is strongly criticized, and the condition to aim for is ‘mildness of temper’ (see Aristotle §§22-4).

Epicurus and his school (including the Roman poet Lucretius) presented impressive accounts of several emotions, in particular the fear of death (see Epicureanism §13). They argued that this fear poisons individuals’ lives and causes social distress. Since they agreed with Aristotle about its cognitive basis, they argued that it could be completely removed by teaching people that death is not a bad thing for the person who has died.

The Greek and Roman Stoics were the great passion theorists of antiquity. They produced impressive analyses and taxonomies of the passions, arguing powerfully in favour of the view that they are essentially evaluative judgments that ascribe to things or persons outside our control great importance for our flourishing (see Stoicism §19). This analysis is continuous with Aristotle’s, but it goes a step further, denying that there is any bodily feeling over and above the cognition that is essential to the identity of a particular emotion. This apparently counter-intuitive view is rendered plausible by extensive consideration of the power of thought to transform the personality.

One might accept this analysis of emotion and still hold, with Aristotle, that many of the judgments involved in emotions are true and appropriate: when one’s child dies, for example, it is right to think that something of enormous importance has been lost. The Stoics, however, held that all the judgments involved in all the passions are false, because all ascribe too much importance to external things and persons, making people dependent on the world for their happiness. To this argument they added several others: emotions weaken the personality, robbing it of force and integrity; they make one prone to excessive and violent acts; and, finally, they are not necessary to motivate good acts, which can be chosen out of duty alone.

Their conclusion is thus that the emotions should be extirpated from human life. Moved especially by the damage done by anger in social life, and convinced that there is no getting rid of anger without getting rid of the attachments to externals that are also involved in love and grief and fear, they came to the radical conclusion that we can stop cruelty and violence only by cultivating utter detachment from everything that used to matter to us. They then strove valiantly to show that we can motivate an active concern for humanity without relying on emotion. Removing emotions is not expected to be like removing false beliefs about more trivial matters: because the evaluations involved are transmitted early through social and parental teaching, they have become deeply habitual, and can be changed, if at all, only through a lifetime of patient effort. Their hope was that as this effort is exercised with greater success, the personality as a whole would become enlightened.

Because their analyses are so compelling and humanly rich, and because they focus with convincing examples on the depredations of emotion in politics, writers such as Seneca and Marcus Aurelius continue to make this radical thesis compelling.

4 Modern supporters of the Stoic normative thesis

The Stoics set the agenda for early modern moral theories of emotion. Most thinkers, though not all, accept some version of the cognitive Stoic analysis of passion; they then differ sharply over the normative thesis. Descartes (1649) modified the Stoic analysis to fit into his divided account of mind and body, but his definitions of the passions are related closely to those of the Stoics; his normative judgments, while not quite as severe, are similar (see Descartes, R. §10).

Spinoza accepted the Stoic analysis and the radical normative thesis, holding, with Seneca, that philosophical therapy can free us from bondage to our emotions (see Spinoza, B. de §9). He gave the normative programme new urgency by insisting that all strong attachments are essentially ambivalent: we love an external person or thing because we find that it assists our efforts to flourish. But anything that can assist us, so long as it is separate from us, can also frustrate us; so all love is mixed with hate. Like the Stoics, Spinoza held that there is a type of joy that is not an emotion in the sense that it is not based on a overvaluation of the significance of externals; it is the contemplative joy with which one regards the deterministic system of the universe as a whole. Like the Stoics, Spinoza identified that order with god; thus the good state is designated the ‘intellectual love of god’.

Kant did not accept the Stoic analysis of the passions. Oddly enough, he presented no arguments against it, and no substantial analysis of his own; but he plainly conceived of passions as pre-rational, impulsive and undiscriminating. Thus, the only role he saw passions playing in virtue was a rather mechanical one, as forces that either aid or impede duty. In general, he thought they impede it, and he therefore conceived of virtue as a kind of strength of will, in which the will maintains its control over the potentially disrupting passions.

Kant was well aware of the Stoic normative view. To a great extent he shared it, praising the Stoics for cultivating detachment from passion, and for promoting active beneficence rather than relying on compassionate emotion. But he showed considerable ambivalence, and was unwilling to dismiss compassionate emotion utterly. He understood that it may be difficult to motivate beneficence without this emotion, that motives of duty by themselves may not suffice. He therefore urged people to seek out experiences in which they will naturally be moved to compassion, so that their beneficence will have the strength of passion behind it.

A surprising ally of the Kantian position was Nietzsche (1881), who objected to pity as an emotion that insults the dignity of the suffering person by implying that this is a person who really needs the things of this world. Equipped with a rather romantic picture of strength and self-sufficiency, Nietzsche believed that the truly strong can rise above life’s ills through will, and therefore do not need compassion, grief or fear. Citing the Stoics, Spinoza and Kant as his predecessors, Nietzsche denounced the Christian/democratic tradition of praising compassion as a basic social motive.

5 Modern anti-Stoics: the defence of sympathy

Opposition to the Stoic normative thesis began in the ancient world, with Augustine, who insisted that the goal of self-sufficiency was an inappropriate one for a Christian to pursue. Characterizing the ideal Stoic wise man as obtuse in his detachment from longing and pain, he concluded: ‘The fact that something is tough does not make it right; and the fact that something is inert does not make it healthy.’ Subsequent Christian views have agreed in characterizing moral error as due to wrongly placed love, and in making love itself an essential ingredient of the cure. Dante’s Divina Commedia (The Divine Comedy) (1313-21) famously ends with the visionary experience of a harmony of emotion and will. Because emotional error is understood (in an anti-Stoic fashion) to be rooted in the body itself and its sexuality, it is expected that this perfected state will arrive only in paradise, after the purifications of purgatory. Most important Christian thinkers make pity or compassion central parts of Christian life, arguing that these emotions are appropriate expressions of and responses to our vulnerable and imperfect earthly condition.

The eighteenth century saw the flowering of a number of distinct approaches to the sympathetic emotions. Although Christian views did not cease to exert an influence, the notion of original evil was contested increasingly in favour of a view of natural goodness in which basic emotional equipment played a salient role. Hutcheson and Hume made the capacity for sympathy a basic part of human nature (see Hutcheson, F. §2; Hume, D. §3). Hume (1739/40) made many valuable observations about the role of sympathy in motivating moral conduct (see Moral motivation §6). But his conviction that desires and passions are basically noncognitive, and that reasoning is capable only of devising means to ends set by desire, led him to short-change many aspects of the passions that even his own concrete analyses at times acknowledge. Although he connected passions with a characteristic object, he seems not to have treated the connection as essential to the identity of the passion: a passion is simply a particular type of impression caused by the object. Hume’s enormous influence has led to the sharp split between passion and cognition that characterizes much modern Anglo-American thought (see Emotions, philosophy of §§3-4).

Rousseau gave the emotion of pitié, compassion, a central role in social morality (see Rousseau, J.-J. §4). Describing the education of young Émile (1762), he shows that the experience of being astonished and pained at the pain of another is an essential foundation for all society. Compassion leads us to recognize the vulnerability that all humans share, and to appreciate the pain that disasters of various types cause others. Thus, the emotion leads us to be sceptical of distinctions that situate some people as high above others, their fortunes as vastly more secure. Rousseau evidently thought that emotions, involving complex imagining and thought, convey moral information that it would be difficult to obtain in any other way. Accepting a version of the Stoic idea of original human innocence, he seems to have held that excesses in self-love derive from experience of unequal social conditions.

Adam Smith’s The Theory of Moral Sentiments (1759) provides one of the modern tradition’s most detailed analyses of the moral contributions of emotion. Heavily indebted to the Stoics, Smith develops a richly detailed account of the cognitive content of passions such as anger and sympathy. He differs with the Stoics on the normative question, taking up a more Aristotelian view about attachments to family and loved ones. His device of the ‘judicious spectator’, modelled on Stoic conceptions of self-scrutiny and conscience, gives the moral agent a way of estimating the point of propriety in passion: we are to ask what a concerned spectator, not personally involved, would feel at the events, and this will help us identify bias and irrationality deriving from our own personal immersion. Despite his positive view of sympathy and other emotions, Smith thought them inconstant and unreliable as guides to social choice, since he observed that we are most easily moved by events closest to ourselves: an earthquake in China means less to a person than an injury to his own finger (see Smith, A. §§2-3).

Another philosopher who defended the fundamental role of compassion in morality was Schopenhauer (1840). Criticizing Kant, he argued that all genuinely moral action must be grounded in other-directed emotion (see Schopenhauer, A. §6). Imagining compassion as involving a mysterious union of the self with the other, he leaves the reader unclear as to precisely why it is not, therefore, a form of self-concern. Schopenhauer’s account had, nevertheless, tremendous influence in a culture dissatisfied with duty-based views.

6 Contemporary views: influences from psychology and anthropology

The middle of the twentieth century saw the rise of noncognitive views of emotion in both behaviourist psychology and in Freudian psychoanalysis (see Behaviourism, methodological and scientific; Freud, S.; Psychoanalysis, post-Freudian). Philosophers such as Anthony Kenny, Robert Solomon, Robert Gordon, William Lyons, Ronald de Sousa and Michael Stocker have by now demonstrated the poverty of such accounts and our need for some type of cognitive view (see Emotions,nature of §2). (Kenny also in the process offered cogent criticisms of Hume’s elusive and influential account.) During this period, cognitive psychology was itself evolving. By now, psychologists no longer expect to be able to give purely behavioural accounts of emotion in terms of stimulus and response, and most of the dominant accounts hold that emotions are a form of intelligent interpretation in which an animal takes in news of how things are in the world with respect to its most important goals and projects. Richard Lazarus (1991) has argued that emotion’s evolutionary contribution is best explained this way: emotions give animals information that is essential to survival. Psychologists generally agree that we have reason to credit many animals at least with the capacity for complex appraisals suited to this account of emotion. Their work has enriched the philosophical debate.

At the same time, anthropologists have produced fascinating accounts of the role played by social norms in shaping the emotion-categories of different societies. These accounts raise questions about the extent to which an emotional repertoire is malleable; at least with respect to specific sub-types and internal demarcations within categories, there appears to be considerable cross-cultural variation in emotion-types, and also, presumably, at least some variation in emotional experience (see Moral sentiments §3; Morality and ethics §4).

This returns us to the moral problem with which the Stoics grapple: for if we see that emotions are learned with the learning of social norms, and we are convinced that our society is not perfect, then it might be unwise to trust the emotions too much as guides to conduct. One society described by anthropologist Jean Briggs (1971), the Utku Eskimos, actually embodies a relatively successful Stoic programme for the elimination of anger, thus prompting us to ask how we should think about our own goals and moral projects.

7 Feminist accounts of morality and emotion

Women have frequently been devalued on the grounds of being emotional and making decisions by consulting emotion. This has led many feminist thinkers to defend the contribution that emotions might play in moral reasoning, arguing that it is the (male) denigration of emotion that is mistaken. Philosophers such as Annette Baier, Lawrence Blum, Sara Ruddick, Virginia Held and Nel Noddings have defended a normative ethical view in which care for others plays a central role, and have seen this as a way of promoting feminist goals (see Feminist ethics).

On the other hand, feminists who consider the social origins of the appraisals involved in emotion have reasons for doubt. If women have a great propensity to care for others, is this always a good thing? Is it not the case that men have standardly urged women to see themselves as care-givers, rather than as sources of worth and agency in their own right? J.S. Mill (1869) argued that even women’s desire to please men has a social origin and is a legacy of women’s subordination. This line of argument has been continued recently in the work of Catharine MacKinnon (1989), who argues that women’s emotions and desires are in significant ways created by inequality and injustice, and that we should therefore be highly sceptical of the claim that women’s instincts of care are always trustworthy and morally valuable.

It is not necessary to choose between these extremes. We can hold that many emotions are valuable parts of the moral life without giving implicit trust to any that have been shaped in imperfect and unjust social conditions - judging, with Aristotle and Adam Smith, that emotions can be good guides as elements in a life organized and examined by critical reflection.

This article originally appeared in Routledge Encyclopedia of Philosophy.

References and further reading

Aristotle (c. mid 4th century BC) Nicomachean Ethics, trans. with notes by T. Irwin, Indianapolis, IN: Hackett Publishing Company, 1985, books II-IV.(On the role of passions in virtue.)

Aristotle (c. mid 4th century BC) Rhetoric, trans. G. Kennedy, Aristotle on Rhetoric: A Theory of Civic Discourse, New York: Oxford University Press.(Illustrates Aristotle’s view that, since emotions are combinations of a feeling of pleasure or pain with a belief, changes in belief will change emotions.)

Baier, A.C. (1994) Moral Prejudices: Essays on Ethics, Cambridge, MA: Harvard University Press.(Discusses the role of care in moral theory.)

Blum, L. (1980) Friendship, Altruism, and Morality, London: Routledge.(Defends the role of compassion in moral judgment.)

Briggs, J. (1971) Never in Anger, Cambridge, MA: Harvard University Press.(Anthropological study of a society that removes anger.)

Dante (1313-21) Divina Commedia, trans. J. Ciardi, The Divine Comedy, New York: E.P. Dutton, 3 vols, 1989. (Highly influential medieval Christian account of love, which ends with the visionary experience of a harmony of emotion and will.)

Descartes, R. (1649) Les passions de l’âme, trans. The Passions of the Soul, in The Philosophical Writings of Descartes, ed. and trans. J. Cottingham, R. Stoothoff, D. Murdoch and A. Kenny, Cambridge: Cambridge University Press, vol. 1, 1984.(Analysis of passions in accordance with mind-body dualism.)

De Sousa, R. (1987) The Rationality of Emotion, Cambridge, MA: MIT Press.(Excellent account of the cognitive content of emotions and its evolutionary role.)

Gordon, R. (1987) The Structure of Emotions, Cambridge: Cambridge University Press.(Valuable philosophical analysis of emotion/cognition relationship.)

Held, V. (1993) Feminist Morality: Transforming Culture, Society, and Politics, Chicago, IL: University of Chicago Press.(Care ethics and social justice.)

Held, V. (ed.) (1995) Justice and Care: Essential Readings in Feminist Ethics, Boulder, CO: Westview Press.

(Collection of different feminist viewpoints on care ethics.)

Hume, D. (1739/40) A Treatise of Human Nature, ed. L.A. Selby-Bigge, revised by P.H. Nidditch, Oxford: Clarendon Press, 2nd edn, 1978.(Influential account of the relationship between emotion and action.)

Kant, I. (1797) Metaphysische Anfangsgründe der Tugendlehre, trans. J.W. Ellington, Metaphysical Principles of Virtue, Indianapolis, IN: Hackett Publishing Company, 1964; repr. in Immanuel Kant: Ethical Philosophy, Indianapolis, IN: Hackett Publishing Company, 1983.(Discussion of role of emotions in virtue; critique of pity as moral emotion.)

Kenny, A. (1963) Action, Emotion and Will, London: Routledge.(Influential and devastating critique of behaviourist and Humean conceptions of desire.)

Klein, M. (1921-45) Love, Guilt, and Reparation and Other Works, 1921-45, London: Tavistock, 1985.(Central psychoanalytic account of emotion and morality.)

Lazarus, R. (1991) Emotion and Adaptation, Oxford: Clarendon Press.(Critique of behaviourism and account of the field’s return to cognitive conceptions of emotion by leading cognitive psychologist.)

Long, A.A. and Sedley, D.N. (1987) The Hellenistic Philosophers, Cambridge: Cambridge University Press, 2 vols.(Basic sources for Hellenistic views of morality and emotion, with fine commentaries.)

Lutz, C. (1988) Unnatural Emotions: Everyday Sentiments on a Micronesian Atoll, Chicago, IL: University of Chicago Press.(Valuable anthropological study of role of social norms in emotion.)

Lyons, W. (1980) Emotion, Cambridge: Cambridge University Press.(Well-argued defence of a cognitive-Aristotelian account.)

MacKinnon, C. (1989) Feminism Unmodified, Cambridge, MA: Harvard University Press.(Argues that women’s subordination to men has shaped their emotions and desires.)

Mill, J.S. (1869) The Subjection of Women, in J. Gray (ed.) On Liberty and Other Essays, Oxford: Oxford University Press, 1991.(Argues that women’s emotions are shaped by unjust social conditions.)

Nietzsche, F. (1881) Morgenröte, trans. R. Hollingdale as Daybreak, ed. M. Clark and B. Leiter, Cambridge: Cambridge University Press, 1997.(The central locus for Nietzsche’s attacks on pity.)

Noddings, N. (1984) Caring: A Feminine Approach to Ethics and Moral Education, Stanford, CA: Stanford University Press.(Argues that women’s experience of maternal care should be the basis for ethics; opposes abstract principles.)

Nussbaum, M. (1994) The Therapy of Desire: Theory and Practice in Hellenistic Ethics, Princeton, NJ: Princeton University Press.(Discussion of ancient Graeco-Roman views of passion and morality.)

Oatley, K. (1992) Best-Laid Schemes: The Psychology of Emotions, Cambridge: Cambridge University Press.(Valuable account by a cognitive psychologist.)

Plato (c.386-380 BC) Symposium, trans. A. Nehamas and P. Woodruff, Indianapolis, IN: Hackett Publishing Company, 1989.(Describes the ascent of love to contemplation of the immortal form of beauty.)

Plato (c.380-367 BC) Republic, trans. G.M. Grube, Indianapolis, IN: Hackett Publishing Company, 1992, book IV. (Account of the structure of the soul and the role of the passions in virtue.)

Plato (c.366-360 BC) Phaedrus, trans. A. Nehamas and P. Woodruff, Indianapolis, IN: Hackett Publishing Company, 1995.(Illustrates how emotions play a positive role in connecting the person to beauty and truth.)

Rousseau, J.-J. (1762) Émile: ou, de l’éducation, trans. A. Bloom, Emile: or, On Education, Harmondsworth: Penguin, 1991, book IV.(Contains a fundamental account of compassion’s role in social morality.)

Ruddick, S. (1989) Maternal Thinking: Toward a Politics of Peace, Boston, MA: Beacon Press.(Balanced account of the moral importance of maternal care.)

Schopenhauer, A. (1840) Über die Grundlage der Moral, trans. E.F.J. Payne, On the Basis of Morality, Providence, RI and Oxford: Berhahn Books, 1995.(Argues that all proper moral action must be grounded in other-directed emotion.)

Seneca, L.A. (before AD52) On Anger, in J.M. Cooper and J. Procopé (eds) Seneca: Moral and Political Essays, Cambridge: Cambridge University Press, 1995.(Influential account of the removal of anger as social goal.)

Sherman, N. (1989) The Fabric of Character: Aristotle’s Theory of Virtue, Oxford: Clarendon Press.(Excellent account of role of emotion in virtue.)

Smith, A. (1759) The Theory of Moral Sentiments, ed. D.D. Raphael and A.L. Macfie, Oxford: Clarendon Press, 1976. (Rich account of balance in passion, and of role of sympathy in morality.)

Solomon, R. (1976) The Passions: Emotions and the Meaning of Life, Indianapolis, IN: Hackett Publishing Company; repr. 1993.(Influential philosophical account of emotion.)

Spinoza, B. (1677) Ethica Ordine Geometrico Demonstrata (Ethics Demonstrated in a Geometrical Manner), trans. E.

Curley, Ethics, Harmondsworth: Penguin, 1996.(Analyses our ‘bondage’ to passions as cause of strife, urges a therapeutic programme to remove passions.)

Stocker, M. (1996) Valuing Emotions, Cambridge: Cambridge University Press. (Valuable analysis of evaluative dimensions of emotion.)

---