I leave the performance of Uncle Vanya with a peculiar mixture of emotions. Undoubtedly, they are imprecise, fleeting and indefinite. But I may be able to say something about them, without fear of gross inaccuracy. I pity Vanya and Sonya; they will have a hard life ahead of them – they know it and we know it – and if they do find peace, then it will only be the peace of the grave. But I am thinking not just about the characters I have seen, but about myself and some of the people I know. On stage, we saw Serebryakov: an academic who writes about art, who is respected (to some degree) by his family, but about whom there is some doubt whether he really knows very much or has contributed anything to the academy or to the world beyond. Vanya sums it up: ‘all he ever does is write nonsense, grumble and feel jealous.’1 An academic spectator who hears this description and does not wonder, at least for a moment, if it applies to her is probably not an academic whose writings I would bother to read. There are other characters whose concerns do not directly relate to my own, but who remind me of other people I care about – be they bored, depressed, hopelessly idealistic, ill or elderly. I am also filled with admiration and respect for some of the actors and, perhaps, I am mildly frustrated with others. And there is a familiar feeling after seeing a Chekhov play: I wonder in a vague and general way whether humans can make progress, whether they can achieve anything meaningful or be fulfilled by their work and by their relationships with each other. Undoubtedly, these are sad thoughts, even if I do not draw exclusively negative conclusions (and I don't normally draw any conclusions at all); and it would not, I don't think, be completely unjust to say that there's something unsatisfyingly pleasurable in having the occasion to wallow in them. I am also aware of a kind of general gloominess or moroseness, which may be linked with all my other, various feelings, but which is certainly distinct from all of them. At the same time, I wonder what I have missed in the translation and what it would have seemed like to an audience in 1897, and perhaps I am a little nervous that we may have missed the last bus home.
This description is hardly complete; perhaps some of it rings a bell, and perhaps not. But I hope it is not completely incomprehensible, at least to those who have not immersed themselves in too much philosophical discussion of art and emotion. I mention the various kinds of emotions produced by the play in order to emphasise that a single theatrical performance does not produce a single emotion, any more than, say, a funeral or an ordinary day at work produces a single emotion. Our emotions during and after the performance may well relate to characters, to actors, to themes; to ourselves, to our friends, to others in general; some seem more like moods, lenses onto the world that don't have any particular object or focus.2 Some of them may be broadly characterised as positive emotions, others negative, others not obviously either. It is true that we are not, in general, very good at describing and expressing our emotions. Nor are we very good at defining them, understanding what they are and how they work. There are occasions when we know what we feel when we feel it; but there are plenty of occasions when we don't know until later, and plenty when we need others to help us to understand or when we never know at all. But in as much as we do feel, it seems evident that theatre can produce such feelings, often to a surprising and powerful degree.
So theatre moves us. Not always, of course; but often. Most philosophers who have written about theatre have agreed about that. But they have disagreed about pretty much everything else. Some, for example, have held that the production of certain feelings is the definitive feature of art in general. This was Tolstoy's view: ‘Art is that human activity which consists in one man's consciously conveying to others, by certain external signs, the feelings he has experienced, and in others being infected by those feelings and also experiencing them.3 If, say, a performance does not permit the communication of feeling from the artist to the audience, then that performance simply is not art. The result of this strong definition is that ‘art which does not move us’ is a kind of contradiction in terms – the ‘art’ in question is not really art at all, but rather something else, perhaps a kind of intellectual confidence trick. This result was hardly accidental: Tolstoy's definition is offered en route to claiming that new so-called ‘art’ (by which he means late nineteenth-century art, which includes, for example, late Ibsen plays) should not really qualify as art at all, precisely because it leaves audiences completely unmoved.
Aristotle's definition of tragedy, as we have seen, names not the production of feeling in general, but that of two feelings in particular (and their catharsis) as the goal of tragedy:
Tragedy is an imitation of an action that is admirable, complete and possesses magnitude; in language made pleasurable, each of its species separated in different parts; performed by actors, not through narration; effecting through pity and fear the catharsis of such emotions.4
For both Aristotle and Tolstoy, in rather different ways, the production of feeling must lie at the heart of theatre. However, it is possible to admit that art produces emotion, but to deny that this is central either to its definition or to its success. If one thought, for example, that the role of theatre lay in the transmission of certain truths, then the effect that theatre has on the emotions might seem relatively insignificant. Hegel wrote that feelings are ‘the indefinite dull region of the mind’; hence, he thought, a study of art that places emotion at its core inevitably ‘becomes tedious from its indefiniteness and vacancy, and repulsive from its attentiveness to little subjective peculiarities’.5
As we have already seen, a similar spectrum of views is represented in accounts of the relationship between theatre, morality and emotion. For some, the production of emotion at a theatrical performance is (at least) an extremely useful contribution to moral development; for others, the ability of the theatre to ‘feed and enflame’ the emotions is the subject of intense suspicion.6 Here again, of course, the fact that theatre moves us is not in question – it's just a matter of whether the effects it has are positive or negative in moral terms. So emotions may be central to the definition and value of theatre; or they may be peripheral, even immoral. But just the fact that we feel strongly, or that we can feel strongly in response to theatre, has been the subject of much philosophical debate – regardless of what the value or significance of this emotional response is taken to be.
This chapter treats three problems in relation to the emotions that, as we’ve seen, all can agree are produced by theatre (at least some of the time). First, the problem of emotional responses to fictions; second, the problem of tragic pleasure; third, the problem of catharsis. These problems may well be related and we will have a chance to think about their connection; but they are distinct and therefore deserve independent attention.
Put as simply as possible, the first problem is this: why do we care about the characters in the play, when we know that they don't exist? I know that Vanya and Sonya are Chekhov's creations, played by actors; the actors may or may not have similar general life concerns, but that is completely irrelevant, and in any case they are hardly stuck in a country estate, condemned to slave away in misery for the rest of their lives. And yet it came naturally to me to say that I pitied them. And not just to me: recall the centrality of pity to Aristotle's definition of tragedy.
This is an old problem: Plato was aware of it; Hamlet comments on it.7 It continues to bother us. It has become conventional to set it out in the form of three claims, each of which looks intuitive, but that cannot all be true together:
1 We are moved by the fate of the characters in the play.
2 We know that the characters in the play do not exist.
3 We are moved only by that which we believe to exist.
To say that these three cannot be true together is to say that they cannot be true in such a way that we are perfectly consistent or rational. One response, then, is simply to accept all three claims and accept, therefore, that we aren't rational: sometimes we do things that we know to be irrational; one of these things might be weeping over the fate of fictional characters. Hence Radford concludes, from his discussion, that the way we respond emotionally to artworks ‘involves us in inconsistency and so incoherence’ – but no more so than fearing death when we believe it to be a ‘dreamless sleep’.8 Radford's claim is appealing in many ways. But we can probably admit that if there's a simple explanation of my pity for Vanya, which doesn't make me fundamentally irrational, then that's the explanation I would prefer to go for. Responses that seek to avoid irrationality may be broadly divided into those that aim to deny or modify the first, second or third claim, such that all three sit nicely together (and also do justice to the phenomena). We shall look at each in turn.
Looking back to the description of my response to Uncle Vanya, it's clear that plenty of other things move us, which have nothing to do with the characters. One thing (amongst many others) that Chekhov is doing with his play is drawing our attention to a set of concerns that, quite simply, bother us: can we get meaning or fulfilment from our lives? If we feel emotional about these concerns, it is nothing to do with Vanya or Sonya in particular; perhaps we simply see them as media, as messengers who pass these concerns from Chekhov to his audiences. We don't care about them as such; we care about the questions they raise. And because the questions they raise are all too real, there is no problem reconciling this with the second and third claims, above. But this won't quite do. Nobody, I think, would deny that these general kinds of question can stir us as part of watching a play. Furthermore, the distinction between what I feel specifically for Vanya and what I feel (say) for myself is always going to be blurry. But, in order for this to be a successful resolution of the problem, it's not enough to say that the emotions stirred up by the play are often connected with real-life concerns. To resolve the problem, one must also deny that we ever feel for the characters. After all, I might well care about Vanya's particular concerns and also care about how not to waste my life. The one does not exclude the other.
The question would be, then, whether we can explain all of what we feel in relation to the performance, without making reference to feelings about the characters. This would involve a kind of translation between claims about the characters and claims about something else. Thus, when I say that I pity Vanya, I am really saying that I pity (say) people who reach a certain age having dedicated their lives to something they no longer believe in and who feel it is too late to make anything meaningful out of their existence. Because the latter undoubtedly exist, there is no problem in meeting the third criterion. Hence, although saying I pity Vanya (the character) sounds a lot like saying that I care about my (real) niece, Anya, my two claims are actually completely different in kind. One is about a person – the other refers to general concerns.
But could we really translate every claim about the character into a general claim? This seems doubtful. One thing I might care about, for example, is whether or not Helen (the professor's wife) and Michael Astrov (the doctor) will have their affair or whether they will part without seeing each other again. A ‘translation’ of this might then go as follows: that I care, in general, about affairs between professors’ wives and doctors. But that is false – I don't. Or perhaps: I care about affairs between beautiful, bored, unhappy women and aging, overworked, proto-environmentalists with ridiculous moustaches. But, again, I don't. Now it is true, of course, that many people have general feelings of various kinds about the role of infidelity in their lives and the lives of others. Some theatrical performances may well offer us the opportunity to work through our own thoughts and feelings – a platform for understanding ourselves. But, watching Uncle Vanya, I care about the specific outcome of the conversations between Helen and Michael; I want to know whether or not they’ll meet in secret in the forest reservation (which, as it happens, I also know to be non-existent). And these specific concerns don't seem connected to any general concern I have about infidelity. To put it bluntly: whether or not Helen and Michael get together won't tell me very much (say) about whether my partner is cheating on me, or whether my own infidelity is justifiable, and so on. Yet, still, I want to know what happens to them. And, because I know that they don't exist, we haven't resolved our problem.
A different way of denying that we are moved (denying 1, above) would be to deny that what we understand to be an emotion in response to a character is really an emotion at all. We have already said something about Kendall Walton's theory of make-believe in the context of the discussion of mimesis. For Walton, once we view plays (and others works of art) as games of make-believe, it turns out that apparent emotions in response to artworks are not real emotions, but make-believe emotions. Walton holds that genuine emotions are connected to existential beliefs (as stipulated in 3, above). Thus, to fear X is to believe that X exists.9 If I do not believe that X exists, then by definition I cannot fear X. But Walton does not deny that we have certain psychological and physiological responses to fictions. In such cases, it's not that we believe that X exists (and then have certain psychological and physiological responses to X) it's that we make-believe that X exists (and then have certain psychological and physiological responses to X). Whatever the response to our make-believing, it cannot be a genuine emotion; instead, it is a make-believe emotion. There are examples (notably, it must be said, examples of uncontroversial make-believe10) in which this seems plausible. When children play a game in which they are being chased by the ‘vampire’ (someone wearing fake vampire teeth), and they scream as the vampire chases after them, it might make sense to say that they are not really afraid, they are make-believing that they are afraid of a make-believe vampire.11 Walton extends this to all (supposed) emotional responses to artworks. Fictions license us to respond to them in a certain way; thus, just as it is fictional that Willy Loman is a failed salesman or that Iago tricks Othello, so ‘it is only fictional, not true, that we feel for Willy Loman or detest Iago’. Hence, there is no ‘loss of touch with reality’ and we should reject the first claim – namely, the claim that we are moved by theatre.12 Indeed, Walton goes further, claiming that reports of emotional responses to fictions are themselves a kind of make-believe in accordance with the rules of the game; so when I say ‘I pity Vanya’, it would be somewhat equivalent to the child who says ‘I see a vampire’.
In one sense, then, Walton places all my feelings about Vanya on the same level as the child's ‘fear’ of the vampire-teacher: a kind of playful playing-along to a game, rather than a genuine, emotional experience. Given the strength and depth of emotion that theatre can produce, it is certainly tempting for any theatre-lover to throw Walton's book out of the window at just this point. But we need to be careful here. It's not that my subjective state – the one that I intuitively call my ‘pity’ for Vanya (although Walton would not call it ‘pity’) – is weaker, or less troubling than my pity for a real person with a similar biography. It is merely a claim about the cause of the two instances of so-called ‘pity’ (make-belief versus belief). It has nothing to do with their strength, or depth, or about how they seem to me subjectively. Thus one and the same subjective experience (my heart beating faster, my fists clenched, a feeling of heightened agitation and so on) would be one of genuine fear or one of fictional fear, depending on whether I believe in or make-believe in the existence of the threat. In both cases, I am experiencing the subjective state associated with fear (what Walton terms ‘quasi-fear’); sometimes quasi-fear is brought about by belief, but sometimes it isn't. So it is no problem (in principle) for Walton that fictional fear is very distressing, seems just like real fear, has many or all of its physiological accompaniments and is ‘as intense as anything you might feel outside the theatre’.13
Nonetheless, Walton's solution to the problem leaves a great deal to be desired from an account of my emotional response to Uncle Vanya. For one thing, his account leads to some very peculiar results. I mentioned not just my pity for Vanya, but also my general feeling of gloominess. The kind of moods we find ourselves in during and after performances are integral features of theatrical experience, but they don't attach themselves to particular characters or events in the play, nor do they obviously have any specific beliefs associated with them. Here, Walton effectively claims that my gloominess may be both real and fictional.14 This looks to be unnecessarily complicated. Isn't it more natural just to say that I feel gloomy and to forget about the hunt for clearly distinguished fictional and non-fictional emotions? More importantly, one has to wonder whether all that much has been explained. After all, isn't it still odd that I experience what Walton would call ‘quasi-pity’ (i.e. all the subjective responses associated with that emotion) in response to the story of Vanya? Even accepting that my pity doesn't count as real pity, you might think that full knowledge of Vanya's non-existence would suffice to rule out the kind of subjective feelings associated with pity: being sorrowful, weeping and so on. So ideally we would want to have a better account of what Walton takes to be our peculiar ability (say) to make our hearts beat faster, to clench our fists, to sweat, to feel all that we would associate with fear in relation to a non-existent object. The simple explanation (which he has to deny) is that I’m experiencing the symptoms of fear because I’m afraid; likewise for pity. Make-believing that I am a millionaire does not make me rich; why should make-believing that there is a ghost make me actually tremble? Trembling and experiencing all the symptoms of terror in response to a make-believe object is still odd. Walton has hardly offered a solution if the curious phenomenon he sought out to explain is simply pushed back to the level of make-belief, rather than belief.
To emphasise some of the problems here, it might be helpful to look beyond the Vanya example, to some slightly different cases. In the Vanya example, I have watched (what I take to be) a successful performance of a great work. But, sadly, this is not always so. Sometimes plays completely fail to move us: I know I’m supposed to feel pity for Elektra, let's say, but the performance has been so unutterably dreadful that all I feel is frustration and regret. One could imagine similar scenarios for other emotions: a supposedly terrifying scene that makes me laugh instead of making me afraid. In such cases, it is not clear how the make-believe plus quasi-emotion structure could help me to understand what is happening. Naturally, one would say that it wasn't scary, or that I just didn't feel for any of the characters. But for Walton that was never a possibility. Why is it that sometimes we experience quasi-fear and sometimes we don't, given that we are just as willing to make-believe in each case? This is no less mysterious than the problem we began with.
Second, just as there are times when I do not feel anything in response to the performance, so there are times when I do feel something, but I very strongly wish that I hadn't. At a grotesquely sentimental play for children, I might weep as the naïve but war-stricken boy is finally reunited with his long-lost horse. But I hate myself for being so susceptible to this trash. Naturally, I would say what Walton cannot: that I pitied the boy and his horse, even though I didn't want to. And there is nothing strange or irrational about feeling something and wishing I didn't. Walton, however, would have to say that I make-believed not only the boy, the horse and the reunion, but also that I make-believed my own pity – the same pity that I find myself so disappointing for being unable to control. The idea that I am make-believing both a story and an emotional response to that story that both has the force of bringing me to tears and also disgusts me is, to put it mildly, not a very intuitive account of what is going on. It's much simpler to forget about the make-believe and say that the boy–horse reunion made me feel a genuine pity that I wish I hadn't felt. The position that we are seeking to reject is the irrational one of holding all three inconsistent claims, above. But it's not clear that the picture offered by Walton of me make-believing emotions that I detest is in any way more rational that the picture of me feeling sorry for characters I know not to exist.15
Denials of the second claim are relatively rare, but we have come across one of them already. In our discussion of theatre and illusion, we noted Stendhal's view that, for brief moments during a performance, the spectator really does believe that what she is seeing is real (moments of ‘perfect illusion’). Thus, one can explain why I ‘cry so copiously’ at the theatre: for brief moments, I believe I am seeing Vanya, not the actor playing him; and I believe that Vanya exists. Needless to say, my belief is false; but many people are scared of plenty of things that they falsely believe to exist (hell, for example) and there is nothing strange about that.
I am very sympathetic to the idea that, for brief moments during the performance, we really are experiencing perfect illusion. We can get absorbed in plays relatively easily and it's hard work to keep reminding ourselves that it's not really happening.16 We cannot simply argue that, because we believe it was all a fiction when the curtain goes down, we believe this at every moment during the course of the performance. Emotions arising from perfect illusion may perhaps account for some of what we feel in relation to some performance.
But it cannot be the whole story. For Stendhal admits, as he must, that perfect illusion, if it happens at all, happens only very briefly – ‘a half-second or a quarter second’.17 Well, perhaps it might last a little longer than that. But even so, the question must be whether such brief moments, be they minutes or seconds, can suffice to account for all of the emotions that we have in relation to fictional characters and events. Thus, for example, I may pity Sonya just at the moment that she learns of her rejection, because I am absorbed in perfect illusion and I believe her to be real. But if I pity her in any way after that – say, as I leave the theatre after the show is done – then we still have a problem. Intuitively, it seems as though I do pity her once the show is over. In fact, I still pity her now, as I write. If emotions arising out of perfect illusion account for all that I feel about Sonya, then either Stendhal must claim that I continue to believe that Sonya exists once the performance is over (which I don't); or he must claim that somehow the memory of the genuine emotion (but not the emotion itself) carries over, in which case I do not pity Sonya once the show is over, I merely remember pitying her (this begins to look like a denial of 1, not 2). So a denial of 2 doesn't seem to give us the full picture, even if we accept the notion of perfect illusion.18
Could we deny that we care about things only when we believe that they exist? In some clear cases, we can. First, there is a sense in which the events of next year do not currently exist, but still I care about some of them a great deal. But, of course, this is not a helpful model for caring about Vanya.19 The events of next year certainly will take place: assuming I am still alive, they will have a direct effect on me and on my life. Not so for Vanya. Second, there are emotions that do not appear to require any kind of belief at all. Suppose you are alone at home, and for some reason you begin to feel afraid. There has been no evident cause for this fear – no strange noises or unexplained phenomena. You cannot say exactly what it is that you're afraid of. This general kind of fear doesn't seem to require any kind of belief about the existence of something to fear; it's just a mood. So the question of belief just seems irrelevant to our understanding of this kind of emotion. As discussed, general moods caused by performances can be powerful and significant. But this can't be all there is to it. For example, the parallel with Vanya is not helpful. My pity for Vanya is not a general mood; I know exactly what the object of pity is: it is Vanya, wasting his life away working for a thoughtless, vain academic whose wife he is desperately in love with. Third, suppose you tell me a story about a girl who suffers a series of terrible misfortunes. Suppose, too, that I don't know whether or not the story is true. In such a case, before knowing whether or not the story is true, I might be justified in pitying the girl. It doesn't seem as though I need to believe in her existence in order to pity her. But here, still, we don't quite have what we're after. In the case of the story of the girl, I don't know whether she exists or not. In the case of Vanya, I know full well he doesn't exist, and still I pity him. So even if we changed the third claim such that we feel emotions for things, the existence of which is unknown to us, we haven't come any closer to understanding my pity for Vanya.
It is perfectly true that, although they help us to modify 3, none of these kinds of emotion helps us to explain how I pity Vanya, knowing that he does not exist. But they do suggest that we ought to be more liberal about the kinds of emotions we have (and their causes) than 3 would suggest. They point to the possibility of ways of experiencing emotion which are not as simple as, say, fearing the tiger that is directly before my eyes. And if all of these various ways of experiencing emotion are possible, then why is it impossible to experience emotion in relation to a fictional character? We feel things based on certain triggers. Sometimes we understand what those triggers are; sometimes we don't. Of course, sometimes when I feel fear, it's the fear that this particular tiger is going to eat me immediately. And sometimes I feel pity for a real person facing a difficult situation. But I can experience fear without there being any particular thing that I'm afraid of; and I can also feel pity for a little girl you tell me about, before I know whether or not the story is true. So why assume that I can't feel for Vanya, whose story is so pitiable, who suffers as I watch him? What we're effectively suggesting here is that my feelings for Vanya simply prove that we can feel for things that we believe not to exist. So it's just not true that we feel for things only when we believe they exist (i.e. 3 is false), just because I can experience emotions for fictional characters. Of course, sometimes when I pity a person, I believe that they exist; but sometimes I don't. As it happens, in Vanya's case, I don't. Now, it is clear that for some philosophers the fact that I know that Vanya doesn't exist means I can't pity him; but one might respond that, for others, the fact that I pity Vanya means that I can pity things when I know they don't exist. This is something of an impasse, and it's not clear where we can go from here. But we might note that, of the three claims we have considered, this third claim looks the least firmly grounded.
What we have set out here are some basic strategies for responding to the problem of how I might pity Vanya. Given the enormous amount of material written about this problem, these arguments have hardly been given full exposure in this chapter. But I'll end the discussion of this problem simply by pointing out that there is absolutely no reason to think that every apparent instance of an emotion directed towards a fictional character must be explained or analysed in exactly the same way. Some of these strategies are incompatible, but many are not. Perhaps sometimes, as Stendhal suggests, we get so carried away by the performance that we really believe that a character exists and we respond accordingly. Perhaps, on other occasions, our emotional response is a kind of make-believe, a pretend fright that we act out to play along with the action. Or what we first take to be a concern for a character disguises a feeling that we have about a theme or problem that they represent, or a real person whom they resemble. And perhaps, sometimes, we find ourselves genuinely feeling sorry for someone whom we know not to be real – whether we see this as an anomalous instance of sheer irrationality, or whether it is a legitimate case of sympathy for non-existents. Outside the theatre, our emotional lives are complex, varied and difficult to describe with any great accuracy; the same is probably true when we watch a play.
The second problem related to theatre and emotions is, in a sense, dependent on there being some kind of solution to the first. When I wrote, above, about my feelings after the performance of Uncle Vanya, it is notable that much (although not all) of what I described was far from positive. I feel sorry for some, I feel sad for others. Often, when we go to plays – tragedies, but not only tragedies – we watch people whom we admire or respect going through terrible misfortune: Desdemona is strangled; Hippolytus is killed; Oedipus is blinded. Put simply, the so-called ‘paradox of tragedy’ is this: under normal circumstances, we don't like crying, we don't like feeling sad, we don't like watching people whom we care for suffering horribly and we don't like being made to feel sorry for them. But when we watch tragedies, we experience many if not all of these things; and, in some way, we seem to like it, enjoy it, recommend it to our friends. So, if we don't like these things in everyday life, then why should we like them at the theatre?20
As with the other problems with theatre and emotion, the paradox of tragedy has a long history. Gorgias, Plato, Aristotle and Augustine show some awareness of it; Hume devotes an essay to it. We shall begin with Hume, although it's clear from his essay that by his time it was already an established topic for philosophical discussion. He begins:
It seems an unaccountable pleasure, which the spectators of a well-wrote tragedy receive from sorrow, terror, anxiety, and other passions, which are in themselves disagreeable and uneasy. The more they are touched and affected, the more they are delighted with the spectacle, and as soon as the uneasy passions cease to operate, the piece is at an end.21
This quotation gives us the basics of the problem, although it's worth noting straight away that Hume seems wrong to say that we are more pleased, the more we are affected.22 After all, some tragedies may well be too painful to watch. More generally, plenty of people don't like tragedies precisely because they are so miserable and, judging by box office sales for long-running shows, comedies of various kinds continue to be far more popular than tragedies. Indeed, Hume's essay is remarkable mostly for its lack of interest in tragedy (not to mention its lack of insight), and also for its peculiar solution, which has been given far too much philosophical attention, perhaps because its author was writing in English.23 So rather than spending too much time on Hume's particular account, let us begin with two claims, both of which require our assent if the paradox of tragedy is even to get off the ground:
1 Tragic events, when real, are not pleasing to us.
2 Tragic events, when theatrical, are pleasing to us.
There is more to the paradox than these two claims. But it is clear that a denial of either would render the problem immaterial. The paradox rests on there being a difference in our response to tragic events, when those events are on and off the stage. A denial of 1 would suggest that tragic events, whether real or theatrical, are pleasing; a denial of 2 would suggest that tragic events, whether real or theatrical, are not pleasing. In each case, there would be nothing special about theatre. Broadly speaking, we may divide responses to the paradox of tragedy into three categories: those that deny 1; those that deny 2; those that deny neither 1 nor 2, but seek to resolve the problem in another way. We shall look at examples of each of these approaches.
It has become almost a custom in recent writings on the paradox of tragedy to open by saying that, obviously, we simply don't take pleasure in watching suffering. However, given the kinds of entertainment humans have indulged in over the years – gladiator battles, public executions, jousting, hunting, bullfights, boxing and so on – it seems that the possibility of taking pleasure in suffering (on stage and off stage) should not be ruled out without discussion.24 Indeed, plenty of philosophers, among them Lucretius, Hobbes, Burke and Nietzsche, have thought it relatively uncontroversial that we enjoy watching people suffer, especially (as in theatre) when we know ourselves to be safe and sound.25
I have placed responses of these kinds under the heading of ‘sadism’. But it might be helpful to make a distinction between a strong and a weak version of sadism. Both versions maintain that we enjoy watching the suffering of others. According to the stronger version, when we watch other people suffer, we feel no negative emotions at all; on the weaker version, there is some negative, perhaps painful response to watching the sufferings of others, but there is also a pleasurable response, which is inextricably linked to the suffering and which outweighs the pain. In both cases, the point is that our experience of tragedy is not especially unique and does not require a special explanation.
Burke appeals to a strong version of sadism, by way of an explanation of tragic emotion.26 His starting point seems to be the observation (uncontroversial in his day) that the public take great delight in executions – greater, in fact, than in the best tragedies:
[Choose] a day on which to represent the most sublime and affecting tragedy we have; appoint the most favourite actors; spare no cost upon the scenes and decorations; unite the greatest efforts of poetry, painting and music; and when you have collected your audience, just at the moment when their minds are erect with expectation, let it be reported that a state criminal of high rank is on the point of being executed in the adjoining square; in a moment the emptiness of the theatre would demonstrate the comparative weakness of the imitative arts, and proclaim the triumph of the real sympathy.27
As Burke realises, most people aren't going to be thrilled at the suggestion that we all want to hurt each other. After all, it's not as if we go around happily mutilating strangers. But he has a response. The key distinction, he maintains, is between, first, wanting some horrible event to happen and, second, given that it is happening, wanting to watch. Whereas the first desire would be peculiar, perhaps pathological, the second, he suggests, is almost universal. Most people don't want a horrible accident to happen on the road; but, driving past one, most people crane their necks to see. This allows for a general desire to see other people suffer without the extra (and at least prima facie improbable) claim that we actively seek to make them suffer whenever we can. Because enjoying tragedy doesn't involve bringing about horrible events, we can set about doing what we always do, namely enjoying watching them unfold. Note, then, that to deny 1 we do not need to claim that we like making each other suffer, nor that we like watching our friends and relatives suffer. All Burke claims is that, given the suffering of others, we often seem to like watching: and that is all that theatre requires.
Although Burke avoids one problem (the unintuitive claim that we actively try to hurt each other), he is still left with another. For as a description of the psychology of tragedy, this looks unintuitive (at least for some tragedies, in which we take a certain pleasure). We don't crane our necks to watch the characters meet their sticky ends: often the deaths do cause us some discomfort and we do find ourselves experiencing negative emotions such as sorrow, pity and so on. It's not clear where Burke finds space in his discussion to admit that some of the emotions that we feel in response to tragedy might really be unpleasant.
According to the weaker version, though, I feel some sorrow at watching Hamlet meet his end; but I also take pleasure in his suffering, perhaps for other reasons. Hobbes, for example, writes:
As there is novelty and remembrance of own security present, which is delight, so there is also pity, which is grief. But the delight is so far predominant, that men usually are content in such a case to be spectators of the misery of their friends.28
Note that Hobbes isn't denying the significance of feelings of pity and sorrow, nor that the experience of such feelings is negative. It's just that pleasure from the fact of our own security trumps whatever pity we feel, and thus enables us, on balance, to enjoy the sufferings of others, including those of our friends.29 Thus he is not faced with the same problem as Burke, for he does not deny that we feel a genuine grief at the sufferings of others.
One problem for Hobbes is that he seems to assume a psychology, according to which we normally forget about our security and are reminded of it by being shown the misfortunes of others. But there is no obvious reason why (at least for some people) it shouldn't work exactly the other way around. Why assume that my reaction would be one of relief at being reminded that some misfortune is not happening to me right now, as opposed to terror at being reminded that it might? Perhaps Hobbes (or the defender of a solution arising from his remarks) would claim that we take more pleasure, the more certain we are that the misfortunes depicted couldn't harm us. Certainly, his examples from the same section – following Lucretius, the pleasure of watching from land as a ship is tossed in a storm – suggest that we take pleasure in threats to others that aren't presently threats to ourselves. But in that case, we would expect to enjoy tragedies that show events unlikely to threaten us more than those that show real, extant threats. Thus, a Chekhov play that depicts bored, spoilt, miserable people whittling away their lives would be less pleasurable to a modern, Western audience than Oedipus Tyrannus, whose particular misfortunes, whatever else one may say about them, do not pose an imminent danger to most of us. However, this just doesn't seem to be true: some people prefer Chekhov; others prefer Sophocles. And, what's more, it's not always so easy to treat misfortunes at such a particular level. Thus, I might be happily going about my day, when, upon being taken to Oedipus Tyrannus, I am suddenly reminded (say) that human lives are frail, and that even the best and most noble individuals have little or no chance at fending off the horror that confronts them at every turn. This is made more pressing, because it's at least a reasonable interpretation of Oedipus Tyrannus that misfortune tends to strike just when you feel most secure and prosperous – exactly when, for Hobbes, the pleasurable response to the misfortune of others is meant to kick in. So even though some of the specifics of Oedipus' plight are unlikely to concern each audience member, it's fair to say that the general menace of deep instability and insecurity might be enough to cause them some bother. Again, why assume that pleasure at not being Oedipus would override displeasure at being reminded that the same (sort of) thing might happen to me? Of course, we haven't ruled out some instances of just the response that Hobbes outlines. It's just that, for all we know, reminding people of the dangers facing others could at least go either way.
If we can't solve our problem by appealing to pleasure that we take in real suffering, then perhaps we could deny that the portrayal of tragic events on stage is pleasurable. There are two ways to deny this: (1) tragedies are not pleasurable at all; (2) tragedies, although they may be pleasurable, do not please us in virtue of the tragic events that they portray.
A strict denial of the claim that tragedies please is relatively hard to find. Some philosophers have argued (with justification) that ‘pleasure’ doesn't seem to be a good description of what it is that we value about tragedies, or of why it is that we watch them. Certainly, it seems right that the best tragedies are not necessarily the tragedies that give us the most pleasure.30 If you had asked me how Uncle Vanya was, I would probably have replied that it was one of the best performances I had seen in a long time; I might not have said, unprompted, that ‘I really enjoyed it’. There's nothing particularly surprising about this: the best philosophy, the best political systems and the best computers (however one were to measure such things) would not obviously be the most pleasurable in any straightforward sense. So it may be that what we like about tragedy should be separated from what we value about it, why we accord it a high status among art forms, or what we think it, in particular, has to offer us.31 Indeed, some discussions of the paradox of tragedy appear to assume that we only do things if we enjoy them; this assumption – sometimes called ‘the hedonistic theory of motivation’ – has rightly been called into question.32 But although this is an important qualification, it does not actually amount to a denial that we find tragedy pleasurable: it's just that taking pleasure isn't the most important thing.
To see why there is a problem with the full-strength claim that tragedy does not please us, it may help to recall a distinction made in Chapter 5 between pleasure as a kind of physical sensation (the opposite of pain) and a more general notion of something being pleasing to us. Some activities – such as eating, when we are hungry – bring us a physical sensation of pleasure. But there are plenty of things we do that please us, but that do not obviously give us a physically pleasurable sensation: learning, gossiping, travelling and so on. Then, of course, there are things that we take part in, which do not please us at all: household chores, perhaps certain kinds of paid labour, or unpleasant medical examinations. In such cases, we are doing something unpleasant for the sake of something else that we value (hygiene, money, health). It is unlikely that anyone would claim that tragedy is pleasurable in the first of these senses. It doesn't give us a physically pleasurable sensation akin to that of eating when hungry or warming up when too cold. The claim is, rather, that tragedy pleases us in a more general sense. Of course, there may be all sorts of benefits to watching a play that do not reduce to pleasure: but, if one wants to claim that tragedy doesn't please us at all, then (as with household chores or medical appointments), one had better explain why it is that we go. Normally, when I pay some money to do something that I don't find at all pleasurable, I am able to offer some plausible account of why.
It's not hard to think of possible candidates for benefits from tragedy that are independent from pleasure: truth, beauty, moral edification. It's just that, if tragedy is not at all pleasing, then we would expect going to the tragedy to be something that we perhaps felt neutral about, disliked, dreaded, or simply were bored by. This doesn't seem like a good description of those who frequent tragedies. What's more, if tragedies were not pleasurable at all, but useful for the sake of some other benefit, then, just as we praise a doctor who performs some procedure less painfully than the rest, so we would expect spectators to praise a tragedian who conveys more truth with less distress.33 This seems all wrong: we often praise tragedies for being moving in a ‘tragic’ way.
Going to the tragedy does not fit the model of a non-pleasing activity undertaken for the sake of some other benefit. But we can still deny (2) while accepting that tragedy pleases. We engage in plenty of activities that please us overall, but that have unpleasant elements. One could make this claim for tragedy: that it produces unpleasant, negative emotions that we find displeasing, but it also offers other compensating factors that make the experience, as a whole, a positive one. This would neatly dissolve the problem, because there's nothing all that strange about the benefits or pleasures of an activity outweighing the costs: so, just as one might enjoy visiting a friend despite having to make the unpleasant journey to see her, so one might go to the theatre to enjoy certain features (the beautiful lines, the performances, the complex plots, the universal themes), despite the negative emotions one will be forced to endure. But although the independence of the pleasure from the pain is an analytic possibility, it doesn't quite do justice to our experience. So, for example, it would be difficult to imagine a successful tragedy – a tragedy that we appreciated and enjoyed as a tragedy, rather than as a farce or as a play starring someone we know – that did not make us feel in any way sad or compassionate (and so on). As James Shelley points out, if we enjoy the tragedy despite the negative emotions it produces, then one wonders why we couldn't watch plays that retained the pleasing features, but that didn't make us feel sad at all.34 What seems more likely is that there is something about the negative emotions themselves that is pleasing; but, with that, the problem of tragic pleasure has returned.
Could we solve the paradox of tragedy, while accepting both that tragic events are pleasurable in the theatre and that tragic events are unpleasant when not in the theatre? A number of solutions have been proposed along these lines. A prominent solution appeals to the pleasure taken in emotional catharsis. Because we shall be treating this in a separate section, I shan't discuss it here. But note that catharsis, as a concept, is just as problematic as tragic pleasure itself, so catharsis certainly can't be wheeled out, without discussion, to settle what it is we enjoy about tragedies. Nor can it be offered, with any certainty, as Aristotle's ‘solution’ to the paradox of tragedy: although Aristotle thinks that tragedy involves taking pleasure in pity and fear, he neither formulates the paradox, nor makes it sufficiently clear that catharsis could function in that role. Setting catharsis aside, I shall consider three proposals which appeal to truth, morality, and the nature of the emotions, respectively.
Perhaps one reason we go to tragedies is that they present us with truths. In Chapter 3, we looked at some problems with this claim; but, setting those aside, would this perhaps provide a solution? For one thing, truth can't be all there is to it. If tragedy tells us truths, but does so via negative emotional responses, then one wonders why we couldn't do without the truth altogether, saving ourselves an unpleasant evening. Or, if we are set on truth, wouldn't we be better off getting our truth via a less distressing medium? One would have to provide an explanation as to why we discover truths via tragedy (with its attendant pains) rather than by neutral or even pleasurable means (or just not at all). James Shelley's proposed solution attempts such an explanation, using the notion of repression. The truths presented by tragedy are truths that we seek to repress in our everyday lives, because they are disturbing and they conflict with how we wish the world to be. But repressing them is (somehow) unpleasant or difficult. Hence, when they are revealed to us at the tragedy, we experience the pleasure of no longer having to repress such difficult truths: ‘tragedy relieves us from the pressures of thwarted truths.’35 Shelley doesn't deny that tragedy is pleasurable, nor that watching other people suffer is unpleasant. The claim is as follows: that we repress all sorts of unpleasant truths about the world; that repressing them is unpleasant; and that releasing them (through watching a tragedy) brings the pleasure of no longer having to repress.
We have already seen that deriving truths from theatre is not straightforward. The first problem with Shelley's argument is the lack of useful examples. At one point he suggests the following, by way of a candidate: ‘a good man, acting in accordance with his best judgement, acts in way that leads, unforeseeably though with astonishing ease, to his own destruction.’36 This seems a plausible (though not unproblematic) reading of Oedipus Tyrannus. But plenty of people, myself included, would happily agree that this is true, without feeling the need to repress it at all. Yet, contrary to what Shelley's account would predict, I can still enjoy a performance of the play. Second, supposing the truths in question are in fact so nasty that one has to repress them: why is the joy at no longer having to repress them not outweighed by the horror at their release? If I am sitting on a box that is full to bursting with venomous snakes, then the effort of keeping them shut away may be troubling. But the pleasure at releasing them (i.e. the pleasure of no longer having to exert myself to keep them in) would surely pale compared with the discomfort of being surrounded by angry, venomous snakes; and if letting them out of the box isn't all that bad, then why did I exert myself so much to keep them shut up there in the first place? Finally, if we repress these truths outside the theatre, why don't we repress them inside it, too? Returning to Shelley's ‘good man using best judgement’ truth: if the everyday world is the kind of place where that holds, and Sophocles' Thebes is also a place where that holds, then why can't I repress it at the theatre just as in real life? I might, for example, deny that Oedipus is good or that he acts in accordance with his best judgement – both of which are plausible interpretations.
The next major candidate for a solution to the problem of tragic pleasure is that tragedy gives us a kind of pleasure that is related to morality. As we saw in our discussion of theatre and ethics, the case for theatre being moral in relation to the emotions has been made frequently and in different ways. Susan Feagin has argued that tragedy provides a kind of test of our sympathetic, moral responses. If we witness real, tragic events, we are overcome by sorrow at what we have seen. At the tragedy, in contrast, we know that it isn't real; nonetheless, we can see that we do have the correct kinds of sympathetic responses to tragic events and so we can take pleasure in the confirmation that we are suitably moral beings, connected with our fellow moral beings in the right kinds of ways: ‘we find ourselves to be the kind of people who respond negatively to villainy, treachery and injustice.’37 Of course, taking pleasure in being sympathetic would be completely inappropriate when faced by real, tragic events; but when the events are taking place in the theatre, we are permitted to feel ‘satisfaction’ at our sympathetic, moral responses.38
Feagin's proposed solution has the advantage that it can explain why the negative emotions (pity, fear, and so on) are necessary in order for us to experience tragedy as pleasurable: pleasure comes from appropriate sympathetic responses to the hardships of others; such responses occur in the face of troubling situations; therefore, the pleasurable experience of tragedy requires the depiction of troubling situations. However, Feagin's account still leaves a great deal to be desired. For one thing, feeling pleasurably satisfied with our moral responses to the pains of others – even to the pains of characters in play – doesn't necessarily look that moral. If the audience at Uncle Vanya is composed of hundreds of individuals, who have attended with the express purpose of delighting in their ability to pity plain, hardworking, lovelorn girls and their weary, hopeless uncles, then I'm not sure it reflects very well on any of them, morally or otherwise.39 We might also recall Rousseau's point, that it's hardly a good moral test to be made to feel a certain way in relation to a situation, but not be required in any way to act upon it. More importantly, though, Feagin's understanding of tragedy seems unnecessarily restricted. For her, tragedies would appear to have plots, the sympathetic responses to which should be relatively straightforward, thus enabling a certain satisfaction at getting them right. We would predict, on her account, that morally complex tragedies – in which it is not clear how to interpret the ethics of those involved and to sympathise accordingly – would be less pleasurable, because they would not allow us to take satisfaction in our responses (we wouldn't know what ‘getting it right’ looked like). Clear-cut moral plots, on the other hand, would afford good tests for whether we respond in the right way to the good characters, and so on. But this is hopelessly wrong. If we praise tragedies at all in relation to their morality, it is hardly because they present us with characters and plots, the ethics of which are self-evidently available to all spectators.40 Often, following fairly standard interpretations, tragedies offer worlds that are radically incompatible with the kind of satisfaction that Feagin describes: irreconcilable opposition between conflicting but legitimate moralities (Antigone); the inability of human beings to make moral sense of the world and their place within it (Oedipus Tyrannus); the failure both of social convention and of non-conformism to provide the basis for a good life (Ghosts, Rosmersholm, or pretty much any of the late Ibsen plays). I repeat, here, my scepticism (see Chapter 5) about summing up the moralities of plays in any convincing or exhaustive way – these are merely proposed examples, which the reader can take or leave; my point is only that if these tragedies set our moral gears in motion, they do not do so in such a way that obviously leads to satisfaction in the way we respond. Put another way: it seems to me that tragedies might have the function of challenging our responses or revealing various moral commitments to be incompatible with one another, rather than simply functioning as triggers for the production of (and satisfaction with) whatever moral outlook we took with us into the theatre.
A last attempt at explaining the pleasure of tragedy re-examines the understanding of the emotions that is presupposed by the discussion so far. Up to now, we have spoken of negative emotions, meaning sorrow, pity, fear and so on; and we have assumed that experiencing such emotions is an intrinsically negative experience. In his discussion of the paradox of tragedy, Kendall Walton challenges just this assumption.41 Sorrow, he claims, is not intrinsically negative. But sorrow often arises in relation to very negative events – a bereavement, for example. By calling sorrow negative, we confuse the experience of the emotion with the cause of the emotion. The death of a loved one is obviously a terrible thing; but the feeling that one has in relation to that event is not necessarily in itself good or bad. At a funeral, one would not exactly take pleasure in being made to feel sad; but one might conceivably feel better as a result. Thus, there may well be circumstances in which we could say, without obvious contradiction, that it feels good to feel sad. So, for Walton, there's nothing peculiar about enjoying the feeling of pity or sorrow or fear, whether at the theatre or not. Sometimes, such emotions are pleasurable; sometimes, they aren't.
There is something appealingly simple about this solution to the problem and it has a lot going for it – not least because it is helpful to highlight the difference between the causes of our emotions and what it feels like when they are expressed. But, even acknowledging that it might sometimes feel good to feel sad, it still seems a general rule that people, on the whole, would prefer not to feel sad and would prefer to go out of their way to avoid it. The opposite is true of feeling happy: one can imagine feeling happy and not enjoying it, but this sounds like the symptom of some kind of medical condition or the side-effect of a drug, rather than a standard, everyday occurrence. Tragedy regularly produces emotions that are typically thought to be unpleasant; and yet we seem to enjoy such emotions when at the tragedy. Walton does not offer an explanation of just what it is about tragedy that enables us to enjoy the emotions that typically (if not always) are unpleasant. A more satisfying answer might be able to fill in more of the details.
I'll end my discussion of tragic pleasure in much the same way that I ended my discussion of feeling for fictional characters. We should be wary of looking for one solution that completely solves the problem; the answer may well be different for different kinds of theatrical experience, and there's no reason why different spectators or different performances shouldn't fit different patterns. Thus some tragedies might appeal to a kind of sadistic voyeurism, whereas others provoke a pleasurable or stimulating moral reflection; some tragedies may not be all that pleasurable, but worth sitting through for other reasons; the feeling that you are empathising with your fellow man may on occasions be a pleasurable one, and so might finally acknowledging that the world works in a way that you had always known but tried to deny. I have given reasons why these proposals, in some of their guises, do not give a satisfactory, single solution – and some of them seem significantly less appealing than others. But they may help explain instances of tragic pleasure and it may be that explaining certain instances is all we can do. Understanding the pleasure that you get from a particular tragedy may require self-examination, or perhaps literary criticism and analysis of the tragedy in question.42
What I termed ‘the problem of catharsis’ at the start of this chapter is, it must be said, a different kind of problem from the other two addressed so far. In the other two cases, we were investigating concerns that are relatively easy to grasp, independently of the writings of particular philosophers. We feel sorry for people who don't exist: how strange! We seem to enjoy feeling bad: why? However, in the case of catharsis, the main question is: what did Aristotle mean by ‘catharsis’? To state the obvious: we are not talking about an independent problem, which would have arisen had Aristotle's Poetics never been written; we are talking about a problem of interpretation.
To make matters worse, we are talking about a problem of interpretation which, most writers agree, can never be resolved to anyone's satisfaction. The reasons for this are relatively simple. First, although the word ‘catharsis’ appears in Aristotle's definition of tragedy (see above), that is the only place it appears, when used in this way, in the whole of the Poetics.43 Not only is it not defined or explained; it isn't even mentioned, except in that one lonely sub-clause. And although it is standard to translate the relevant clause as saying that tragedy effects ‘through pity and fear the catharsis of such emotions’ – thus implying that catharsis is something that affects the spectators – a number of commentators have either challenged this translation or interpretation, arguing that the catharsis is something that happens not to the spectators at all, but rather something that applies to the tragic events or action.44 The remainder of my discussion assumes that it is the spectators who experience catharsis, but the alternative views show the range of interpretations available.
Second, although Aristotle writes about catharsis in other works, he doesn't treat it in any detail and it's an open question whether the kind of catharsis that he is writing about in other places is the same as the one that he associates with tragedy. How helpful, for example, is his brief discussion, in the Politics, of the catharsis brought about by music? Tragedy, after all, had musical elements; but, in his discussion of catharsis and music, Aristotle looks to be talking about specific kinds of ritualistic singing in response to cripplingly ecstatic emotional outbursts.45 Indeed, in the Politics he writes that he won't say anything more about catharsis, because it will be given a fuller treatment ‘when we speak of poetry’.46 If ‘when we speak of poetry’ indicates the text we call the Poetics, then either the fuller discussion of catharsis didn't survive, or he never wrote it (or, given that the Poetics is thought to be teaching material, perhaps he never wrote it down); if it indicates a completely different text on poetry, then that other text is lost. An answer to this riddle is as distant as an answer to the riddle of catharsis.
Third, there are a number of related interpretative problems, which, given the sparse language of the Poetics, we are unlikely to solve with any satisfaction. As things stand, then, we lack a solid foundation on which to build an account of catharsis. I shall give some brief examples. First, Aristotle tells us that tragedy produces the catharsis of ‘pity and fear’. Some have taken it as obvious that ‘pity and fear’ stand in for a range of emotions, including perhaps sorrow, shame, anger and so on.47 Others argue that, if Aristotle had meant to include those emotions, he would have done so.48 If catharsis is something that happens only to pity and fear (but not, say, to sorrow or anger), then it's highly specific and our account of it will follow suit. Thus, it would be worth analysing what Aristotle thought about pity and fear (including what he writes about them in the Rhetoric).49 But if ‘pity and fear’ just head up a long list, then an investigation into Aristotle on pity and fear (only) would be of limited help to us.50 Second, there is the question of how catharsis relates to pleasure. Aristotle speaks of the characteristic pleasure of tragedy (without saying what it is);51 and he (elsewhere, not in the Poetics) associates catharsis with pleasure.52 Does this mean that tragic catharsis is the characteristic pleasure of tragedy? Most assume that it is – hence its central place in the definition of tragedy; others have argued that it is not, because Aristotle mentions several kinds of pleasure that may be related to tragedy.53 A further and related problem, which we shall come to, is the question of who is meant to experience catharsis. Is it (1) all spectators (2) only the most virtuous spectators (3) not all and not the most virtuous? Finally, there is a question that we touched on in relation to mimesis: to what extent is the Poetics in general, and catharsis in particular, intended to be Aristotle's response to Plato's complaints about the effect that theatre has on the emotions?54 How we answer any of these interpretative questions reflects on how we answer the others, leaving us with no safe starting point. For example, if catharsis is just for pity and fear, then it looks less convincing as a response to Plato, who was concerned with the effects of theatre on all the emotions.55
All of which might be thought sufficient to forget about the problem of catharsis altogether. Why should we devote an enormous amount of time and energy to a problem that, even if solved, would tell us what one Macedonian thought was the function of one kind of (now virtually unknowable) theatre approximately two and a half millennia ago and that will, in any case, never actually be solved? When put this way, it's hardly surprising that plenty of people think that we shouldn't and that the extensive critical interest in catharsis has been dismissed by one critic as ‘a grotesque monument of sterility’.56 Nor is it surprising that the word ‘catharsis’, like ‘freedom’ and ‘justice’, has been used to support pretty much any prevailing view, no matter how ludicrous or far-fetched.57
Nonetheless, we shall look at catharsis in a little more detail, for two reasons. First, because I suspect that part of the reason why this one word has attracted so much attention is that we think that there's something plausible about the idea that something like catharsis – whatever Aristotle might have meant by it – can be a feature of theatre, and especially of tragedy. It has a certain intuitive appeal, despite the critical haze. Second, because the notion of catharsis has such a long and established place in the history of philosophy and theatre, for this reason alone, we shouldn't pass it by altogether. My aim, then, is to combine a discussion of various interpretative options with our intuitive sense that something like catharsis may be a feature of our emotional experience of theatre.
The Greek word, ‘catharsis’, had a variety of meanings. While it could be used in a relatively straightforward way to refer to everyday cleaning, modern scholars have tended to focus on two different umbrella terms as the starting point for their discussions. The first, ‘purging’, is primarily a medical notion. The second, ‘purification’, is primarily religious. Before we look at each term, two qualifications are important. First, both purging and purification relate to Greek institutions and practices – medicine and religion – which were by no means standardised, which are completely foreign to a general, modern readership and which (in many cases) remain relatively obscure even to scholars.58 In any case (and this is the second qualification), the suggestion is not that Aristotle thought that tragic catharsis was actually medical or religious: very likely, he was using the term metaphorically, or in a technical sense. Both the medical and religious terms lend themselves to metaphors of various kinds and had already done so by the time Aristotle wrote his Poetics. For both these reasons, therefore, even settling on one of purging or purification won't help us enormously. Nonetheless, because they would take us in different directions, let us say something about each.59
For the Greeks, catharsis as ‘purging’ covered all sorts of different physiological processes – among them, basic bodily functions such as menstruation or the emptying of the bowels (the latter was an accepted meaning of ‘catharsis’ in English well into the nineteenth century) and medical procedures such as the draining of pus. Put crudely and literally, it is ‘getting something out of your system’, which, just as in English, could easily develop into metaphor. If vomiting is ‘getting something out of my system’ in a literal sense, then an intense bout of weeping may be ‘getting something out of my system’ in a metaphorical sense – not the tears, but the emotion. Note that getting something out of your system could be getting rid of it completely; or it could be draining off the excess, where only the excess was unwanted. So, in relation to the emotions, it might suggest getting rid of my fear altogether or getting rid of the extra bit of fear that was in some sense too much.
The modern, English use of the term ‘cathartic’, which probably derives from a certain interpretation of Aristotle's Poetics, looks to express this kind of metaphorical notion of purging. It is used to describe an experience that was emotionally difficult, but that led to a certain kind of satisfactory or even pleasurable resolution. The difficult emotions will be ‘discharged’ (to keep the medical metaphor alive). A ‘cathartic’ conversation with a friend might then be one in which certain conflicts, brewing for a long time, were finally discussed in the open and were resolved, at least for the foreseeable future. It is a term that might be used to describe the experience (or even the function) of a funeral: an opportunity to express intense emotions, which were in some sense there all along (since the death), in such a way that brings a kind of relief or satisfaction. The medical metaphor may also lend intuitive support to the sense in which catharsis might be connected with pleasure: vomiting, one supposes, is not in itself something one particularly enjoys; but, when it happens, it is often followed by a kind of pleasurable relief.
Initially, then, the notion of purging, of getting the fear and the pity out of our system, looks to be a promising one. Hence, for a long time, the notion of purging was given as the standard and relatively uncontroversial interpretation for catharsis.60 It has also found favour with some theatre theorists, among them Artaud, who took up the idea that theatre was designed to ‘drain abscesses’.61 Indeed, ‘purging’ is often presented as Aristotle's intended meaning, without much further discussion.62 We go to the tragedy; our pity and our fear are expressed; we feel a pleasurable relief. It is, in the modern, English sense of the term, a cathartic experience.
However, for various reasons, purging turns out to be problematic. As I have said, our interest here is both in what Aristotle might have meant and, regardless of what he meant, whether the interpretation in question is intuitively helpful for our understanding of theatre. To begin with, compare purging at the theatre with the other kinds of purging under discussion: in the bodily cases, there is something unwanted that is being secreted, excreted and so on; in the metaphorical cases, there is also something unwanted or uncomfortable, which is being expunged. Thus, when I go to the funeral, my weeping is ‘cathartic’ because my grief was always present. The cathartic conversation with the friend was such, because there was always trouble brewing. However, in the case of theatre, there doesn't seem to be an obvious candidate for what was there all along. To be sure, my pity for Vanya and Sonya is expressed during the course of Uncle Vanya. But, before I saw the play, I had absolutely no idea who they were and I certainly didn't have feelings about them that I needed to get out of my system.
To maintain the ‘purging’ metaphor in relation to tragedy, we would need to claim that either (1) the emotion that gets purged by theatre is just the emotion that theatre itself produces or (2) before we even step into the theatre, there are emotions that we need to get out of our system, which theatre helps us to purge. If we opt for (1), then the significance of catharsis is questionable: instead of getting something out of our system, theatre puts something in and then gets it out of our system. In medical terms, this would be like taking a healthy person, making them feel terribly nauseous and then helping them to vomit. Doubtless, the latter would produce a certain kind of relief, so there's nothing contradictory about tragedy performing this kind of role for certain negative emotions. It's just that it doesn't seem a terribly noble or significant role; and, in fact, it makes the spectators sound more like emotional junkies or Roman banqueters than anything else. Perhaps this accounts for why this first option doesn't seem to have been taken seriously by Aristotle critics, who generally seek something profound and significant for the effects of catharsis.63
Suppose then that we opt for (2) – that there is something, all along, that we want to purge. But, if so, then we seem to be saying that, for audience members who experience catharsis, there's some kind of unwanted or troublesome emotional element that they bring with them into the theatre, and that theatre helps them to expunge. In other words, if we choose this option, we need a much broader account of human psychology. (We have already discussed a version of this, in relation to why we might have feelings for fictional characters.) I don't doubt that this is sometimes what happens when we go to the theatre: we are feeling sad about something else, and the play helps us to express that feeling: one reason I was so moved by Uncle Vanya is that Chekhov puts his finger so exactly on some of the things that worry me. But it's unlikely that everyone has that experience or goes to theatre with that in mind and, as we have seen, it's not exactly clear why watching a play in which what I worry about happens to fictional others should bring me any particular relief or pleasure. What's more, the proposed model seems more intelligible for some emotions than for others. Thus, I might be feeling sad and find relief in expressing this at the theatre; but does it really make sense to say that I might be carrying around too much pity?64
These were objections to the intuitive appeal of (2) as an explanation of what happens to our emotions at the theatre. But as an interpretation of Aristotle, the purging claim faces all of this and more. For one thing, Aristotle has a particular view about what it is to be a virtuous person: namely, a virtuous person has the right sort of emotional responses at the right time. He will be afraid when it is appropriate to be afraid, and so on. He is not somebody who doesn't have any emotional responses.65 Thus, a virtuous person sounds like exactly the kind of person who doesn't need a purge at all; a purge might do him harm, by getting rid of the very emotions in virtue of which he is virtuous. But (2), above, suggests that the people who experience the benefits of catharsis are those who need a purge. Thus, either we say that only non-virtuous people experience catharsis, which most (although not all) Aristotle scholars wish to avoid; or we say that purging can't be the right metaphor for catharsis.66 A second objection relates to Aristotle's claims (elsewhere) about the emotions. Fear, he suggests, is not just a feeling; it is attached to particular (fearsome) objects at particular times.67 I don't just ‘feel fear’ (he thinks); I feel fear because of (say) the tiger that is coming towards me. So it's not clear how fear, in general, could be purged. I ought to be afraid of that which is fearful; because whatever was fearful before the tragedy is fearful afterwards, Aristotle probably doesn't think that purging would (or should) take place.68
In a religious context, ‘purification’ seems to have meant something like a ritual cleansing of a person or a place. We can compare this with the thought of ‘washing away’ sins, or purifying a place of worship with holy water or incense. In any case, the intuitive connection is between everyday hygiene and religious or spiritual hygiene: making things clean. Whereas it's fairly obvious what a purging of the emotions would involve (even if it isn't obvious that this is what Aristotle had in mind), it's not at all obvious what a purification of pity, fear or any other emotion would be. So just what exactly does Gerald mean when he tells Anabel that ‘we shall hate ourselves clean at last, I suppose’?69 The purging metaphor is one of quantity. We have a pretty good idea of what having more or less of an emotion involves: I can be more or less afraid, angry and so on. But the purification metaphor is one of quality. So what does it mean to have purer fear or purer pity? The religious context is of little use to us, because there's no obvious analogy, in the spectator's emotion, for the ritual purification of a person, place or object. More so than accepting the purge metaphor, accepting the purification metaphor in itself gives us nothing much to work with.
One thing it might mean, of course, is that the fear or the pity is somehow better. Thus, a more speculative account of catharsis has been suggested, along the lines that it offers a kind of moral training or education, which makes your emotions – and therefore you, as a whole – ‘purer’ in that sense. This is given some plausibility by the fact that, for Aristotle, virtue consists (at least to some degree) in having the right kinds of emotional responses to the world: one shouldn't be afraid of everything; one shouldn't be afraid of nothing, and so on.70 Perhaps, along these lines, we could imagine the spectator ‘getting clearer’ about pity and fear, about the sufferings of others and so on.71
However, what's missing is an account (in Aristotle) of why tragedy should in any way make our pity and our fear more appropriate, or in what way it would morally educate. Because we have already discussed the common but problematic view that theatre can train our moral sentiments, I shall not rehearse the general arguments here. Suffice to say, it's not clear just how it does so, nor why theatrical ‘training’ would fit with the world of everyday experience – which they would have to, if the ‘education’ is to be of any use.72 So to those general concerns with theatre as a school for the sentiments, I shall add some problems for this as an interpretation of Aristotle. First, as with purification, there is the question of whether virtuous people could experience catharsis. If so, and if the virtuous man's emotions are already in harmony, then he has no need of catharsis in this sense. You don't need to be taught what you already know.73 (As before, one can jettison the notion that the virtuous man experiences catharsis. But then theatre – as moral education – becomes something for the morally needy, not for the morally accomplished.) Second, one of Aristotle's other remarks on catharsis seems to tell against this interpretation. When he discusses the effects of music, Aristotle appears to distinguish clearly between music that is useful in ethical training and music that causes catharsis. This is the same text in which Aristotle tells us that catharsis is to be explained in relation to poetry.74 If, in the case of music, catharsis and moral education are distinct, then they are probably distinct in the case of poetry. To repeat: the word ‘catharsis’ in no obvious way means ‘moral education’. Given this, and the clear distinction that Aristotle makes between catharsis and moral education elsewhere, it seems to me that this interpretation is on tricky ground.75
We have had the opportunity to investigate three problems, each of which asks questions about how theatre moves us. It should be clear that these problems may well be related. Some choose to explain the second problem with reference to the third (tragic pleasure is the pleasure of catharsis). More generally, how one approaches the problem of tragic pleasure may depend on how one has answered the problem of how, in general, we respond emotionally to fictional characters. And, before explaining how tragedy purges or purifies our fear, we might think it reasonable to get a better sense of what ‘fear’, in relation to a piece of theatre, actually means. The defenders of theatre, and its detractors, have often appealed to its effects on the emotions. Perhaps this chapter has helped us to understand some of the challenges that confront both sides.
For a sample of pieces on emotional responses to fiction, see Radford (1975), Walton (1990: 195–204), Neill (1993), Zemach (1996), Suits (2006). On the paradox of tragedy, Hume (1965) is often used as a starting point, although he evidently did not invent the problem; see Neill (1999) for a critical analysis of Hume. For a sample of contemporary discussions, see Feagin (1983), Budd (1995), Shelley (2003), Friend (2007). On catharsis: Aristotle's Poetics is, of course, the source of the dispute; interpreters make frequent reference to his Politics (for its reference to catharsis), Rhetoric (for the analysis of emotions like pity and fear) and Nicomachean Ethics (for its discussion of ethics and the emotions) to support their various claims. For a sense of the variety of interpretations, see Halliwell (1986: 184–201) (which also features a helpful discussion of the uses of the Greek term), Schaper (1968), Nehamas (1992), Golden (1973) and Lear (1992). On the rich history of interpretations of catharsis, see Halliwell (1986: 350–6).
1 Chekhov (1980: 144).
2 Having seen a performance, it is perfectly possible to feel one emotion in relation to the actor and another in relation to the character that the actor was playing. See, e.g., Brock (2007: 217–9) for discussion.
3 Tolstoy (1995: 40).
4 Poetics 49b.
5 Hegel (1993: 37–8).
6 See, e.g., The Republic 605b–606d.
7 Hamlet, II.ii 229.
8 Radford (1975: 78–9).
9 Walton (1990: 249).
10 As I have mentioned in my discussion of Walton in Chapter 2, his notion of ‘make-believe’ is much broader than the everyday use of the term.
11 This example is developed from Woodruff (2008).
12 Walton (1990: 241).
13 The quotation is from Woodruff (2008: 164), who appears to miss this feature of Walton in arguing against him.
14 Walton (1990: 252–3).
15 For further critical discussion of Walton's view, see Neill (1991).
16 Pirandello's Six Characters in Search of an Author plays on exactly this phenomenon.
17 Stendhal (1962: 24).
18 For a contemporary defence of a version of this response, see Suits (2006), who argues that, given a more flexible notion of ‘belief’, one can believe that what one watches is real while simultaneously believing that what one watches is fictional.
19 See Eldridge (2003: 195).
20 The ‘paradox of tragedy’ is one of the few problems in the philosophy of art that, by name at least, is connected with theatre. But writers on the problem often argue about whether it just applies to tragedy, or whether it can be applied to other works of art. It seems clear that it also applies, say, to films and novels. Other candidates include paintings, documentary films, roller coasters, fine oratory and so on. Because our concern is with theatre, I won't spend time discussing how far the problem stretches.
21 Hume (1965: 185).
22 See Friend (2007); Feagin (1983).
23 Roughly, Hume thinks that the artistic elements of a performance ‘convert’ the negative emotions to positive ones. For an excellent critical analysis, see Neill (1999).
24 Lennard and Luckhurst (2002: 134) even suggest that public slaughtering of humans and animals in Rome accounts (in part) for the relative lack of appreciation and popularity of theatre.
25 Nietzsche has his prophet, Zarathustra, say that man is the cruellest animal because he enjoys tragedies, bullfights and crucifixions (in Thus Spoke Zarathustra, Part 3, ‘The Convalescent’); a more nuanced discussion of our pleasure in the sufferings of others (and ourselves) can be found in the second essay of The Genealogy of Morality.
26 Burke (1998: 43–4).
27 Burke (1998: 43). Long before Burke's time, the public execution and the theatre were often compared and contrasted as public shows. Thus, in the middle of the seventeenth century: ‘the association of public executions and theatre was a commonplace. Executions were staged for the edification of the audience.’ (Thomson 1995: 203)
28 Hobbes (1994: 58); as Carlson (1993: 129) notes, these remarks are not explicitly about theatre, although he does speak of the ‘spectator’ and the ‘spectacle’.
29 The view that our security is the source of our pleasure at the suffering of others is one that, as it happens, is explicitly rejected by Burke.
30 Thus I agree with Ridley that ‘successful tragic drama – think of Lear, think of Oedipus – is simply not all that pleasing’ (2003a: 413); and with Woodruff that ‘the best theatre is not the theatre that gives us the most pleasure’ (2008: 186). See also Budd (1995).
31 See e.g., Gardner (2003: 236).
32 See e.g. Neill (2003).
33 The medical example is developed from Shelley (2003).
34 See Shelley (2003: 178).
35 Shelley (2003: 183).
36 Shelley (2003: 185).
37 Feagin (1983: 98).
38 Feagin (1983: 98).
39 Friend (2007) makes a similar point.
40 We discussed this in Chapter 5.
41 Walton (1990). As he acknowledges, Walton's solution to the paradox is independent of the claims he makes about emotions and make-believe, which we discussed earlier in the chapter; I have chosen to present it without using the rest of his theoretical apparatus.
42 A further question, unexplored here, concerns the unity not of the answer but of the problem or ‘paradox’ that we began with: are we in fact dealing with a single ‘paradox’ or are we looking at a series of overlapping concerns that deserve isolated treatment? Some have suggested, for example, that the psychological question of what motivates us in going to the tragedy should be kept apart from the moral question of whether it is permissible to enjoy depictions of the suffering of others. I have treated them in a unified manner here, partly for simplicity, partly because the answers are evidently related to one another (are our psychological motivations morally permissible?). But see e.g. Neill (2003: 207–8) for elaboration.
43 The only other mention of the word describes the ‘purification’ of Orestes' madness as part of a tragic plot. See Poetics 55b.
44 Nehamas (1992) and Else (1957) give different versions of this alternative reading.
45 Compare Golden (1973), Nehamas (1992) and Janko (1992) on this question.
46 Sometimes translated as ‘in the Poetics’, for obvious reasons; see Politics VIII.7, 1431b, pp. 37–9. See Halliwell (1986: 190) for discussion.
47 Compare e.g. Taplin (1995: 23) and Halliwell (1986: 200f).
48 A further option, endorsed by some of the French tragedians, was that either pity or fear would do; Lessing argues strongly against this position. See Lessing (1962: Sections 75–6, p. 182).
49 For a flavour of this, see Lessing (1962: Section 74–75, pp. 175–82) and (1962: Section 78, pp. 191–2). Halliwell (2002: ch. 7) gives a detailed reading of Aristotle's account of pity.
50 Recent scholarship, it is fair to say, favours a ‘pity and fear only’ solution; but see Schaper (1968: 136–7) and Janko (1992: 349–350), for an argument that ‘a wider range of feelings’ is invoked. Lessing (1962: Sections 74–75, pp. 175–82) tries to reconcile both positions, by suggesting that pity and fear accompany all the emotions that we feel for the characters; pity, in the sense of feeling what the character feels, makes it possible to get (say) angry and our pity is related to the fear that what happens to the character might happen to us.
51 Poetics 53b.
52 Politics VIII.7, 1431b.
53 See Heath (1996).
54 See Chapter 2 for general discussion. On catharsis in particular, compare Halliwell (1986) with Nehamas (1992).
55 See Nehamas (1992: 306).
56 Quoted in Halliwell (1986: 184).
57 See Halliwell (1986: 350–6).
58 See, e.g. the discussion of Aristotle on ‘homeopathic’ treatment in Halliwell (1986: 192–3). The claim is that curing pity and fear using pity and fear (following the medical ‘purging’ metaphor) wouldn't have made sense in the light of Aristotle's own medical theory, even although others frequently made use of it.
59 It is true, of course, that the metaphor of catharsis might somehow have been intended to combine purging and purification: getting rid of some and purifying the rest. In the absence of a compelling account of such a combination, I won't pursue this here.
60 It is associated with the pioneering work of J. Bernays, discussed in Janko (1992).
61 Quoted in Balme (2008: 76).
62 E.g. Lennard and Luckhurst (2001: 62); Balme (2008: 72) (although Balme considers other interpretations elsewhere).
63 Exceptions include Nehamas (1992) and Heath (1996).
64 Nietzsche, to be sure, thought that nineteenth-century Europe was, in general, suffering from too much pity; he takes Aristotle to be offering a ‘purging’ account of catharsis and prescribing tragedies precisely in order to curb the dangerous excesses of pity. See, e.g. Antichrist, section 7.
65 See Aristotle's Nicomachean Ethics II.6 1106b, pp. 40–1.
66 Compare Heath (1996) with Lear (1992: 316–7) and Halliwell (1986: 191).
67 See Rhetoric, 1382a–1383b.
68 Lear (1992: 317); for more general criticism of the ‘purging’ view, see Golden (1973).
69 D. H. Lawrence's Touch and Go (III. i. 72). I feel as though I know exactly what he means, but it's not easy to explain.
70 See Aristotle's Nicomachean Ethics II.6 1106b, pp. 40–1.
71 Nussbaum (1992: 280–3). One could present ‘moral education’ interpretations, not as a species of ‘purification’, but under a heading of their own – and indeed, some do. But because ‘catharsis’ doesn't mean ‘moral education’, interpreters who favour the latter as a kind of interpretation tend to develop it out of the notion of ‘purification’.
72 See Nehamas (1992: 303–4) for more on this problem, specifically in relation to Aristotle.
73 Nussbaum (1992) rejects the idea that the virtuous person doesn't need training – a virtuous person, she supposes, would be open to correction, and she provides some evidence from Aristotle to support this intuitive idea.
74 Politics VIII.7, 1431b.
75 Halliwell (1986: 195) tries to answer this concern; but see Lear (1992: 319) for critical discussion.