Judgement and emotion

Aristotelian virtue ethics invests emotions and feelings with much moral significance: indeed, the Aristotelian view would seem to be that virtues are more or less equivalent to states of emotion, feeling or appetite ordered in accordance with some deliberative ideal of practical wisdom. On this view, courageous persons are those who avoid excesses and defects of fear; the temperate are those who avoid extremes of appetite; the generous avoid those of extravagance and meanness; and so on. However, despite the valuable work of contemporary virtue (and other) ethicists on emotion in general and/or this or that emotion in particular, certain aspects of human affective experience would appear to have problematic implications for serious reflection on virtue in respects that have not been widely noticed.Footnote ³

In this light, it may be appropriate to begin – despite the abundance of modern philosophical literature already amassed on this topic – with some general remarks about the logical status of emotions, passions, sentiments and other forms of human affect. First, following the pioneering work of Bedford,Footnote ⁴ Peters and Mace,Footnote ⁵ KennyFootnote ⁶ and other post-war analytical philosophers, it has been widely agreed that emotions are ‘intentional’, that they have ‘cognitive cores’ or that they in some (variously specified) way entail or embody judgements. Indeed, on stronger ‘cognitivist’ versions of this view emotions are little more than forms of appraisal or evaluation.Footnote ⁷ On such views, feeling fear or anger largely amounts to judging that things are thus and so, and being disposed to take this or that action in the light of such evaluation. It would also seem not only that such strong cognitivism has often invoked the authority of Aristotle, but that there is also some Aristotelian warrant for such a broadly cognitivist view. Thus, in Book 3 of the Nicomachean Ethics, Aristotle cites – with little apparent reservation – a popular ‘definition’ of fear as the ‘expectation of evil’, and he does also sometimes speak as though an emotion or feeling is primarily some kind of judgement or verdict on this or that state of affairs.Footnote ⁸

To be sure, such thoroughgoing cognitivism has attractive economy: to regard a fear as some (true or false) judgement that this or that state of affairs is threatening or dangerous, offers to avoid awkward commerce with psychologically indeterminate states whose role in explaining conduct may seem no less anomalous. On the other hand, such strict cognitivism seems at odds with both common sense and the equally familiar Aristotelian terminology of emotional excess and defect. Thus, for example, there are problems about conceiving the fear (of the coward) as simply a high expectation – perhaps as some kind of overestimate – of evil or danger. Indeed, one trouble may here turn on some ambiguity about the notion of ‘high expectation’. If a high expectation of danger is simply an inaccurate estimate of the amount of danger, it is in that sense an excessive estimate or an overestimate. But it seems less plausible to regard any such ‘high expectation’ as a stronger or more intense (psychological) feeling of expectation – in the way, say, that this fear might be stronger or more intense than that one. On the contrary, in so far as this (high) expectation could differ from that (low) one only in terms of what is (propositionally) expected, it is less easy to construe the terms ‘high’ and ‘low’ as referring to degrees of intensity in the manner suggested by talk of excess or defect of emotion.

Moreover, in addition to its evident common-sense or ‘phenomenological’ plausibility, the idea that there is an affective as well as a cognitive dimension to emotions and sentiments seems fairly well upheld by what Aristotle elsewhere says about the character of emotions as feelings, and by his tendency to characterise the defects of particular emotions as states of insensibility: Aristotle's account of the role of emotion in virtue indicates that the fearless, for example, are not just those who make errors of judgement, but those who actually lack certain non-cognitive – or not purely cognitive – sensitivities or sensibilities. This also seems in line with the efforts of some recent philosophers to ‘put the feeling back into emotion’.Footnote ⁹ On the one hand, such attempts to reclaim the place of feeling remind us that value judgements may not be sufficient for emotion: that, in short, there might well be value judgement in the absence of what would appear to be emotionally characteristic affect. On the other hand, such reinstatement of feeling has often been prompted by doubts about whether value judgements are even necessary. In this respect, it has been noticed that too close identification of emotion with judgements of value (or other ‘intentional’ content) would preclude the attribution of emotion to pre-linguistic human infants or other non-linguistic creatures. It is also worth noting some reservations about the widespread contemporary view that emotions are necessarily (or fully) intentional, much encouraged by Anthony Kenny's influential work,Footnote ¹⁰ on the part of one of Kenny's own key mentors and influences.

Intentionality and emotion

In some now apparently neglected work, Peter GeachFootnote ¹¹ argued that reports of emotion and passion do not seem to be clearly or fully ‘intentional’ according to the Brentano criteria from which many modern discussions of intentionality have taken their lead.Footnote ¹² To be sure, as with beliefs, emotions can be directed towards false propositions or non-existent (or ‘inexistent’) objects. Again, neither reports of belief nor those of emotion allow ‘salva veritate’ inter-substitution of equally true propositions: thus, if A believes or fears that Paris is the capital of France, it does not follow that he or she believes or fears some other true proposition such as ‘Dublin is the capital of Eire’. But, according to Geach, it is not clear that reports of emotion – as with reports of belief – are (at least always) referentially opaque. If Oedipus believes that the arrogant stranger on the road to Thebes has insulted him, and that arrogant stranger is his father, it does not follow that he believes (lacking knowledge of the relevant identity) his father has insulted him. On the other hand, if Oedipus is angry that the arrogant stranger has insulted him, and the stranger is his father, we might still say that he is angry that his father has insulted him – irrespective of his ignorance of the identity of arrogant stranger with father. To be sure, it may be a more natural way of putting this point – bringing out more clearly the extent to which emotions are focused more directly on objects or states of affairs than upon propositions – to say that Oedipus is angry with the stranger for insulting him (irrespective, as it were, of any mediating description). In this regard, finding the rose sweet, I may find it no less so under some other name: indeed, respecting the familiar source of this allusion, Geach refers to such ‘half-way extensional’ reports of emotion or feeling as ‘Shakespearean’.

That said, it is also reasonable to suppose that in the typical human case – at least in cases involving some level of linguistic competence – emotions or sentiments are likely to be fully intentional in light of some measure of cognition. Moreover, this dimension of emotion and passion would seem to be of utmost importance for the education of emotions – not least in relation to the cultivation or education of virtuous states of character. Might it not therefore be appropriate to regard emotions and passions as – at least potentially – rational? It may have been on some such assumption that one widely influential work in the field of aesthetic and arts education appeared some years ago under the title The Rationality of Feeling.Footnote ¹³ However, if the story so far is correct, this title seems variously questionable. First, if it makes sense to speak of at least some affective states – those of animals and infants – as genuine emotions, then feelings need not always be cognitive and would not in that case be inevitably liable to rational assessment. But secondly, even if we were to concede that all affective states are cognitive or intentional, it might yet be questioned whether emotions or other feelings are always either rational (whatever this exactly means) or apt for rational evaluation.

In short, can we infer from the claim that emotions are in some sense cognitive or intentional that they are also forms of rational judgement? Leaving aside (for present purposes) evidence conditions of rationality, the most common criteria of rational engagement would appear to be coherence and consistency. There is a fairly strong epistemic requirement for beliefs and some other forms of judgement to be consistent if they are also to be rational. So, from a logical (though perhaps not from a psychological) viewpoint, one cannot both believe and not believe that p is the case, and it would seem hardly possible (even psychologically) to believe both p and not p (at least in the space of a single thought). But while it might look no less contradictory to say of the same individual that he or she (both) does and does not love (or hate) p, such seeming contradiction may be more apparent than real. Thus, to the extent that (as Geach observes) one can both love and hate the same objects or states of affairs, love and hate are not – unlike believing and disbelieving – directly contradictory states. One might say that one both loves and hates fish; but once it is appreciated that fish has properties F (appealing taste) and G (annoying bones) and that one loves fish in respect of F but hates it in respect of G, any and all appearance of formal contradiction is dispelled. However, in so far as I might well (albeit in some more or less phobic sense) fear cheese, even though I know that there are no rational grounds for such fear, it is far from clear that one has to provide rational justifications for (at least some of) one's emotions and feelings anyway.

The Ximene problem

All the same, it could be that what has drawn philosophers to think or speak in terms of the rationality or otherwise of emotions is appreciation of the moral or practical significance of some emotional tensions or conflicts. Consider, for example, the story of Ximene in the medieval romance of El Cid. In this narrative, Ximene is deeply in love with Cid until he accidentally slays her equally beloved father. From that point on, Ximene is fired by a passionate hatred and resentment of Cid and much of the subsequent narrative is devoted to exploring the heroine's moral dilemma of both loving and hating the same man. Moreover, this is not clearly a tension that might be resolved in the manner of loving and hating fish. One could not easily imagine Ximene confiding in her best friend Xantelle that while she loves Cid in respect of property F (qua man of her dreams), he also exhibits the unfortunate property G of having murdered her father. In this respect, it is not one separable part or aspect of Cid that is loved by Ximene and another that is hated by her: rather, it is one and the same Cid, the whole Cid and nothing but the Cid who is both loved by Ximene as the man of her dreams – and also hated by her under the aspect of ‘slayer of my father’.

This is evidently a case of emotional ambivalence with no small moral implications. Even so, it is far from clear that Ximene's ambivalence stands to be evaluated in terms of rationality or irrationality – or, for that matter, what any rational resolution of such ambivalence might look like. For a start, were we to insist that some kind of rational resolution of this ‘dilemma’ is required, it is far from obvious whether we should say that Ximene should seek this by suppressing her loathing of Cid as the slayer of her father in the name of forgiveness (which may make her look a bit shallow), or by hardening her heart against Cid in the name of revenge (which may seem not much more morally exemplary). Indeed, what one might wish to say here is that the only moral or virtuous course for Ximene is that in which (perhaps as suggested by the quote from Lessing at the start of this paper) she honestly accepts if not embraces the conflict or ambivalence. We might therefore think that there is something morally untoward about any Ximene who sought psychiatric ‘healing’ of the conflicts between her love for Cid, the pain of losing her father and her resentment of the man who took his life. In any case, any attempt – from some ‘value-neutral’ therapeutic perspective – to resolve Ximene's conflict by removing the psychological obstacles to what is in her best interest, is prone to wreck on the grounds that we cannot be sure (if she is genuinely conflicted) where her best interests might lie.

Still, there can be little doubt that such emotional conflicts, tensions and ambivalences have significant implications for the cultivation of the qualities of character, deliberation and action with which virtue ethics is centrally concerned – not least in so far as such theory evidently aims for some degree of integrity or consistency of moral character and conduct. Indeed, virtue ethics turns crucially on a distinction – rather less clearly acknowledged (if at all) in other moral theories – between the virtuous and the merely continent or self-controlled. In this light, though the continent are distinguished from the incontinent precisely in terms of their resolute pursuit of clear moral ends, they also differ from the virtuous in view of a certain wholehearted commitment to such ends. In short, whereas the feelings and motives of the virtuous are held to be in general line with their moral ideals and convictions, those of the continent – however well they may ultimately perform – may be at some odds with such convictions.

In this light, however, it may be asked whether an emotionally conflicted Ximene could ever be fully virtuous, not just in the general sense that full cultivation of virtue must lie beyond final human achievement, but in the more presently particular sense that there can be no virtue in the face of emotional conflict or disunity. For in such terms, if Ximene is not to succumb to complete psychological paralysis (for which therapy may be the only way out) she will need to adopt some course of action that moves her beyond her emotional impasse. In this regard, she might remain a continent moral agent either by repressing her resentful or vengeful feelings in marital commitment to Cid or (in honour of her father's memory) by renouncing her love for him and taking the veil. But if conflicted emotion is not merely a matter of mistaken judgement but of mixed feelings, and if such feelings are not to be eliminated without some loss of personal integrity, then it might appear that Ximene could never be a virtuous agent construed as one in whom feelings are in full alignment with adopted moral goals. She may eschew violent retribution on the grounds that such violence is morally unacceptable, but she may not be able to resolve her ‘inner’ conflict without also giving up the resentment towards her father's murderer that emotional honesty would seem to require.

Virtue and continence

In a paper close to present concerns, Susan Stark is also anxious to rebut what she takes to be an implausible conception of Aristotle's virtuous (as opposed to continent) agents as those whose right conduct is entirely free of any conflicted, contrary or mixed feelings or emotions.Footnote ¹⁴ Indeed, Stark takes the fairly plausible line that it is hard to make sense of (virtuously) brave or temperate conduct apart from supposing that the brave have to face fear or the temperate may (at least sometimes) have to control intemperate appetites or inclinations. All the same, Stark detects a tendency to a more austere view of the virtuous as emotionally free from such conflict in some contemporary versions of virtue ethics – in particular in the work of her main target John McDowell.Footnote ¹⁵ Drawing on work by Michael WoodsFootnote ¹⁶ and Michael Smith,Footnote ¹⁷ Stark distinguishes between two types of reasons identified as normative and motivating reasons. On Stark's account, the difference between virtuous and merely continent agents is not that the latter experience contrary emotions whereas the former do not, but that whereas the latter continue to experience such emotions as motivating, the former are driven only by some overriding vision of what it is right to do.

Stark maintains that McDowell fails to give sufficient weight to contrary emotion in virtue, because he is in the grip of an ‘internalism’ about reasons which insists that any and all evaluations – including the evaluative components of emotion – must provide reasons for action. In particular, since conduct that involves contrary-to-virtue emotions must provide motives for less than virtuous action, there can be no room in virtue for such emotions on pain of collapse of any sustainable distinction between the virtuous and the merely continent. However, Stark holds that if reasons internalism is rejected – at least with regard to the evaluative aspects of emotion – then there is no reason to deny that contrary-to-virtue emotions or emotional evaluations may find a non-motivating place in the moral psychology of virtuous agents. On this view, the virtuous differ from the continent to the extent that the contrary-to-virtue appetites, feelings and emotions that continue to sway the latter are now motivationally defused in the former. The virtuous may continue to experience feelings of fear, anger or (illicit) sexual desire, but their commitment to a larger overriding vision of the good has effectively deprived such feelings of their motivational power. In the case of the virtuous, according to Stark, whereas contrary-to-virtue evaluations may provide reasons for emotions, they do not provide reasons for action.

Despite the neatness of Stark's solution, however, it seems doubtful whether it does justice to the complexities of virtuous emotion in general or is effective in sustaining the distinction between virtue and continence in particular. For one thing, irrespective of these and other problems of reasons internalism, it is difficult to see why there should be any less difficulty for virtue ethics about non-motivating contrary-to-virtue emotions than there is about motivationally live ones – given that, as Stark concedes, Aristotelian virtue is as much (if not more) about integrity of character as consistency of conduct. But, more seriously, it is far from clear how her account successfully preserves any distinction between the virtuous and the continent – since contrary-to-virtue emotions would appear no less inert in the latter case than in the former. Indeed, what the virtuous and the continent precisely have in common is that both successfully pursue what they take to be the correct course of action in the face of contrary feelings. To be sure, what the virtuous, the continent and the akratic all seem have in common – as distinct from wanton agents – is that they are equally capable of recognising the fine and noble conduct that ought for them to override any and all contrary impulses or inclinations. Of these, however, it is only the contrary-to-virtue inclinations of the akratic agent that are sometimes effective in deflecting the agent from the virtuous path. In short, any agent whose contrary-to-virtue feelings or appetites remain motivationally active would have to be considered akratic rather than continent.

The virtuous and the continent reconsidered

Moreover, if it is part of Stark's case that the virtuous do not experience moral ambivalence – not merely that they are not inclined to less than virtuous conduct – then this is not so clear either. To see this, we might first revisit the distinction between the continent and the virtuous. No less than others (not excluding the virtuous) the continent are vulnerable to a range of contrary-to-virtue passions, emotions and inclinations to violent anger, lust, cowardice and so on. Moreover, like the virtuous (and unlike the akratic), the continent will generally regard acting upon such inclinations as morally untoward and therefore resist such action by efforts of will. According to the standard Aristotelian view, however, the virtuous differ from the continent – and are therefore morally superior – by virtue of not requiring the same effort of will. To be sure, continent agents – or at least some of them – may actually view heroically resisted opportunities for adultery with some regret: they may even wish (albeit unconsciously) that they were not cursed with the strong moral conscience (the product, perhaps, of a strict puritan upbringing) that has denied them the opportunities for pleasure available to the less scrupulous. The virtuous, on the other hand, are said to be incapable of such ambivalence. Although they may feel strong sexual attraction to the spouses of others or feel fear in circumstance that demand a principled stand, they are unable to view adultery or cowardice as anything other than dishonourable or shameful.Footnote ¹⁸ The virtuous person is therefore blind to the attractions or non-moral benefits of backsliding or cheating.

But would it thereby be true to say that virtuous (or, for that matter, continent) agents always act virtuously – or as a virtuous person should act? It would seem that the answer to this must be both yes and no. In one sense, to be sure, virtuous agents must (by definition) act virtuously – with, that is, virtuous judgement and intent – for otherwise they could not be considered virtuous. Indeed, on one common interpretation of Aristotelian virtue ethics, virtuous action is more or less defined as what the virtuous would do. That said, in the light of our consideration of Ximene type dilemmas, it seems that even the virtuous cannot avoid less than morally correct action. Indeed, though our recent discussion of the distinction between continence and virtue has proceeded as though the key moral question is that of how to do what we know is right, it is time to recognise that dilemmas of the Ximene variety (though not necessarily, of course, this particular one) are a familiar feature of the human moral condition. In a significant sense, to be sure, there is no moral problem in those cases where it is clear what we have to do and all that we require is the will to do it. In this respect, real moral problems are those in which there are conflicting moral principles and imperatives such that whatever one does is likely to incur some moral loss. Thus, people will often find themselves in (more or less weighty) circumstances in which they cannot prevent a large (or larger) amount of real human harm without some ‘economy with the truth’, or compromise of principle.Footnote ¹⁹ Moreover, to whatever extent such dilemmas might yield to the cost-benefit resolution of utilitarian analysis they cannot but remain virtue ethically problematic, since – whatever the actual scale of dissembling or irresolution – failure to observe strict honesty or integrity must always involve some compromise of moral character for the would-be virtuous.

At all events, in so far as the virtuous cannot – any more than the rest of us – avoid the pitfalls of moral dilemma, they can get things wrong and also do wrong.Footnote ²⁰ To be sure, they may be expected – by virtue of practical wisdom – to do a bit better than the rest of us; but since they are no more blessed than others with prescience, practical wisdom is no sure proof against error and miscalculation. Thus, returning to the Ximene dilemma, let us first suppose that she is a candidate virtuous agent and that her decision – made in the best possible faith and knowledge at the time – is to bear the pain of Cid's slaughter of her father and reconcile herself to him in a spirit of love and forgiveness. However, having married Cid and borne his children, it turns out that his violence towards her father – far from being the out-of-character episode she was led to believe – is actually symptomatic of a deeply ingrained brutality that now threatens her safety and that of her children. Or, alternatively let us suppose that Ximene had wrongly believed that Cid was a cruel and brutal man and therefore decided to renounce both him and the world and take the veil. In the event, it turns out that the killing was an accident and that Cid was a deeply sensitive person whose remorse and (as he believed) unrequited love led to his breakdown and suicide. In the first case, Ximene might have been better guided by her negative attitude to Cid; in the second, by her positive feelings.

How, in such cases, might Ximene cope with such eventualities? To be sure, there is probably not much that she might do in the second case – though the added misery caused by such revelation may well impose further psychological and moral burdens and challenges. In the former case, however, Ximene clearly has practical choices. One of these might well be to stick with her original decision to suppress her negative feelings of anger and resentment in light of more positive emphasis on the virtues of love and forgiveness. To be sure, if (as Stark appears to hold) it is the mark of the virtuous (as opposed to the continent), to have ruled out – in the interests of some virtuous unity of character and conduct – contrary-to-virtue evaluations as possible reasons for action, it might for Ximene be the only conceivable course of action to honour her original pro-Cid commitment come hell or high water. But this rather assumes that Ximene's original dilemma was a straight choice between virtuous and non-virtuous action – which of course, being a moral dilemma, it was not. Indeed, in so far as virtue ethics recognises – even requires – righteous or justified anger, and Ximene's anger at the murder of her father was clearly justified, it would be misleading to suggest that anger and resentment are necessarily negative feelings. Moreover, as Stark herself insists, no coherent conception of virtuous agency should exclude the kind of emotional complexity in which both pleasant and painful emotions are mixed in the usual human proportions.

However, if Ximene is to have other (virtuous) options besides remaining loyal to Cid, the emotional grounds of her negative post-murder assessment of him may need to retain their motivational power. In fact, it is hard to see how such emotions could lack such potency and still continue to be genuine emotions – as distinct from (say) recollected emotions. Here, indeed, it is worth noting that a recalled or remembered emotion would fall well short of the genuine article: I may remember having had emotions that I no longer have (and that I may despise myself for having had) and I may also presently have emotions (as psychoanalysis teaches) that I am not now clearly aware of having. Still, it is certainly hard to see how Ximene might act on any other apparent moral options open to her (of perhaps leaving Cid for the sake of herself and her children) unless the negative evaluations of Cid grounded in her resentment at the slaying of her father provide reasons for action as well as emotion – though of course she might also act on these without either ceasing to love or even forgive Cid. But is this not precisely what we might expect of the virtuous: that they would possess the reflective and motivational capacities to recognise and act on the widest possible range of moral options? On this view, the virtuous or practically wise would be precisely those equipped with the richly complex – albeit conflicted – psychological life through which alternative possibilities of (virtuous) action remain available.

Virtue, emotional conflict and repression

To the extent that Stark's account of the difference between the virtuous and the continent reflects features of a fairly common view of Aristotle's distinction, we should recognise the element of truth that it contains. Thus, the virtuous are those in whom certain common springs of human motivation are no longer operative: virtuous agents are those who are no longer in the least inclined to certain squalid or dishonourable courses of action. In this light, for example, it may be that although the virtuous may still feel sexual attraction to unlawful partners (for example, the husbands or wives of others), they could not be credited with lust or adulterous feelings – in so far as the values that serve to define such feelings or passions would not be those of a genuinely virtuous agent.Footnote ²¹ On the other hand, the continent may remain prey to some motivational struggle or ambivalence involving such base sentiments: thus, although the continent must (by definition) resist adulterous temptations, they may well do so with some reluctance – perhaps to the point of resenting the moral conscience that denies them the pleasures of less scrupulous others.

Such points aside, however, it is arguable that Stark's account has things more generally the wrong way around. First, as we have seen, while the continent may experience emotional ambivalence and struggle, they are also defined as having unswerving devotion to what they have come to regard as the morally right path – for if they were not so inclined they would not be continent but akratic agents. On the other hand, though there may be squalid or dishonourable courses of human action to which they are no longer drawn, the virtuous are no less prey than the rest of us to the emotional conflicts and moral dilemmas to which human flesh is heir, so that the absence of emotional or moral uncertainty or conflict would not in and of itself distinguish the virtuous from the continent. But, more strongly, the standard Aristotelian story might lead us to expect that the virtuous are those who have the wisdom, imagination and ability to address such conflicts in ways that are not available to the merely continent. On this view, it takes a virtuous Ximene to appreciate that her conflicted simultaneous love and resentment of Cid have objective grounds that cannot be wished away and that require honest and courageous confrontation. Such honest confrontation with the moral and emotional complexities of her relationship with Cid may also be needed to enable the virtuous Ximene to change her mind and revise her conduct in the event that (say) she realises her loyalty to him has been misplaced.

On the other hand, it could be a more single-mindedly continent Ximene who lacks not only the self understanding but also the moral imagination – the capacity to judge how to feel and act with the right motive, at the right time and with respect to the right persons or objects and so on – that the wisdom of virtue demands. It may also be that such points have large implications for some of the deepest problems of moral psychology. For while Aristotle's concept of continence might be considered a morally required developmental stage on the way to virtue – through which perfectly decent human beings may have to pass to genuine love of the good for its own sake – it may also be the psychological source and cause of some of the worst human excesses and vices of self-delusion, hypocrisy and fanaticism. One trouble with continence construed as a mechanism for moral resolution of conflicts between instinctual human drives and the demands of civilized propriety and decency is that it can sail dangerously close to the process – brought most vividly to modern attention by psychoanalysis – of repression. Indeed, the idea that much excessive moral zeal is often fuelled by projected self-hatred was clearly recognised long before psychoanalysis. Thus, for example, Shakespeare: ‘Thou rascal beadle, hold thy bloody hand! Why dost thou lash that whore? Strip thine own back; Thou hotly lust'st to use her in that kind for which thou whipp'st her.’Footnote ²²

Hence, the problem with much (though perhaps not all) continence may be an inability to face the emotional and moral complexity to which a variously divided human nature is all too often heir. The virtuous may feel attraction to the partners of their best friends but have the wisdom and self-understanding to put this in its proper human place. It is a brute fact of life that one is humanly prone to such feelings: however, the morally mature recognise that they cannot have everything they desire and that the satisfactions of acting on untoward attraction could not possibly outweigh the personal and interpersonal damage (to integrity, relationships and so on) that such conduct is likely to incur. It may be that some merely self-controlled agents are incapable of coming to mature terms with the prospect of such inevitably ungratified desire. That said, the self-controlled do not seem divided in the way that Stark suggests: on the contrary, they may be single-minded to a fault. In cases of severely repressive self-control, anything at odds with the agent's adopted moral or religious imperatives will be the object of monomaniacal hostility – whether it is located within or beyond of the self: the lustful beadle whips the whore; penitent anchorites scourge themselves. Unsurprisingly, the wild and hysterical responses of the repressed are also prey to absurd and telling incoherence. Thus, fundamentalists who observe the strictest and most abstemious requirements of holy writ in preparing acts of terrorism against the decadent liberal west, may nevertheless aspire to heavenly rewards of debauching virgins – with little inkling of what this reveals about their own motivationally and morally divided selves.

All of this, moreover, fits more or less well with the Lessing quotation at the start of this paper – as well as with the general spirit of Aristotelian virtue ethics. To be sure, on a superficial reading of the Lessing quote, it might seem to be saying: (i) that the essence of neurosis is conflict; and (ii) that the essence of life is conflict – from which one might also (incautiously) conclude that life is essentially neurotic. However, in a context in which the main character appears to be questioning a certain view of psychoanalysis as concerned with the clinical ‘cure’ of neurosis, the point would seem to be rather that the concern of good psychoanalysis should be the moral one of helping ‘patients’ to progress from an unhealthy pathological or neurotic state of conflict to a rather more healthy or normal appreciation of conflict as an unavoidable aspect of the human condition. In this light, the positive mission of psychoanalysis might be summed up in the well known prayer of Reinhold Neibuhr: ‘God, grant me serenity to accept the things I cannot change, courage to change the things I can, and wisdom to know the difference’. Thus seen, the trouble with neurotic agents – or the more seriously repressed continent – is that they lack the wisdom or self-knowledge needed for both honest recognition of what they cannot change and the courage to bear the ungratified desire that is the inevitable consequence of such recognition. Their posture is rather one of quixotic self-denying commitment to second hand ideals that they are – developmentally and educationally – ill-equipped to sustain.

But the role of psychoanalysis so construed would also seem consistent with the cultivation of moral virtue as conceived by Aristotle. One trouble here may have been a certain contemporary trend towards regarding Aristotelian ethics (perhaps by over-sharp contrast with more rationalist ethics of duty or utility) as an ethics of disposition rather than principle – and therefore towards conceiving the growth of virtue in terms of unquestioning or unswerving commitment to some set of unproblematically unified moral goals (a conception which may have echoes in Stark's account). However, as the other introductory quote to this essay from Reeve warns, it may be a mistake to ask too much of the unity of the self – if, for example, this means complete psychic wholeness – and, indeed, the unity of the virtues is probably better routed (as ancient philosophers held) through the mediating offices of practical wisdom.Footnote ²³ The virtuous need practical wisdom precisely because no genuine moral experience or engagement can possibly avoid the slings and arrows of emotional ambivalence and moral uncertainty. Aristotle himself (perhaps in some contrast with Plato) ascribed significant moral developmental value to great literature and the arts precisely because such literature explored the deep psychological complexities of what he took to be moral issues and concerns of universal human significance.Footnote ²⁴ Hence, rightly seen, far from impeding the development of virtue, emotional ambivalence and personal conflict may provide the rich psychological and moral soil, that – in the light of practical wisdom – is actually required to enhance the possibilities and prospects of virtuous character and conduct.