Why do we enjoy tragedy? It may be thought that our best hope of answering this question lies in the psychology of Freud, who disclosed the dark side of the psyche. Behind this darkening of the mind, however, there lies another darkening, of our picture of the ancient sources of European literature. Antiquity, formerly given over to the Ego, becomes the province of the Id. Roughly speaking, a sunlit, rational, enlightened world – peopled as it were by marble figures in a state of tranquil felicity (think of Winckelmann) – was replaced, retrospectively, by an opposite world: blood guilt and sacrifice, dream and vision, orgiastic music, unreason. One way of expressing this change is to say that the pretence of Augustianism was dropped: instead of assuming that antiquity was somehow full of 18th-century rationalists having either no religion or a religion etiolated and simplified to the point of minimal Deism, it was at last noticed that the ancient world pullulates with spirits and deities, is crammed with unreasonable, alarming powers. This, by the way, is simply true.

Within this general darkening we find a particular thesis about Greek drama: namely, that Greek tragedy was fundamentally ritual, that it was religious in ways which moderns find difficult. There is an entry in one of Leonardo da Vinci’s notebooks which reads: ‘The sun has never seen any shadow.’ Leonardo is here thinking like a good Albertian perspective painter: it has suddenly struck him that for an eye which is itself the source of light, shadows will always be on the far side of any object. For such an eye, darkness is essentially and systematically suppressed. For the Greeks, the sun was Apollo (called ‘most powerful eye’ in Sophocles’s Trachiniae). But Apollo, god of unshadowed light, was not the only deity in the pantheon. There was also, for example, Dionysus, the god not only of intoxication but also of wildness, green nature, of the irrational. One way to represent the change of which I am speaking might be to say: ‘Apollo had been honoured for centuries; now Dionysus was given his due.’ And of course this antithesis is Nietzschean.

The effects of this transformation are all around us in Modernist writing. Joyce’s Ulysses is not just a 20th-century mock-heroic: Joyce used the Odyssey as a quarry, not of rational certainties to be subverted, but for magic and metamorphosis. Eliot in ‘The Waste Land’ deployed a Frazerian intuition: ancient European history and blood sacrifice linked to fertility. The movement appears in different forms in many places. Jane Harrison’s Themis and Gilbert Murray’s writings on the Dionysiac Year Spirit belong in this line.

Nor was Nietzsche the first to notice the murky side of the ancient world. Earlier rationalists were sometimes either too learned or too sharp to preserve the inherited blindness. Frazer’s The Golden Bough itself is in some ways a 19th-century rationalist work, finding the courage to be critical of antiquity for its very failures in enlightenment. Time and time again, Frazer can sound like Gibbon – for example in his account of ‘Saint Hippolytus’ or when he says that ‘the good taste and humanity of the Greeks must have recoiled from the more violent rites of the Magna Mater.’ In the pages immediately following, he explains how introversion, stemming from the East, corrupted the ideal of public service, until the Renaissance brought back ‘saner, manlier views of the world’. It will be said that the word ‘manly’ is richly Victorian, but in fact Gibbon speaks of ‘the manly pride of the Romans’. Mistrust of the Near East reappears, of course, in Nietzsche. There is a certain affinity between The Golden Bough and, say, the Pitt-Rivers Museum in Oxford, with its vast array of bloodcurdling fetishes, among which Christian votive offerings – by design – fail to stand out as different. Tantum religio potuit suadere malorum (‘So much evil could religion do’). Hume likewise, in his Natural History of Religion, written before 1757-130 years before Nietzsche’s Genealogy of Morals though not published until 1777, saw religion in naturalist, evolutionary terms rather than as a structure of revealed, unchanging verities. Hume has no doubt about the irrational character of ancient religion, I suppose we smell Modernism, not when antiquity is shown to be irrational, but when the irrationality is itself gratefully embraced.

Eliot would resist, but I feel this is palpably true of ‘The Waste Land’. There is a considerable irony in Eliot’s growing up in a time of world-weary scepticism and then finding something like religious belief only when he turned to a book designed to show the absurdity or wickedness of ritually-based religion. He disregarded the instructions on the bottle. Milton in his ‘Nativity Ode’, back in the 17th century, had beaten Frazer to some of his material (Milton’s ‘wounded Thammuz’ is Frazer’s Adonis), but for Milton, unlike Eliot, these sanguinary deities were enemies of true religion, to be hunted from their lairs at the birth of Christ. For Eliot, it is rather as if all the voices are somehow telling of one thing – resurrection, life out of death. The Eliot of ‘The Waste Land’ is not yet Christian, but I think one can smell it coming. I am suggesting, I suppose, that Eliot’s talk of tradition is, in fact, the masking of an almost solipsistic terror, and that his Christianity (which I take to be real and strong) was something reached in darkness by a route other than that of reason. He loved Dante but his thought (though never exactly Nietzschean) is much less close to Aquinas than it is to Schopenhauer – the first major European philosopher after Plato to turn to Indian religion. Schopenhauer had a bust of Buddha in his room and kept a poodle called Atma (‘World-Soul’). The Jesuit philosopher Coplestone has noticed, incidentally, that Bergson, whose influence on Eliot is undoubted, is himself remarkably like Schopenhauer. Sorry about all this rapid allusiveness. I am saying, in a way, that Eliot, unable to find a reason for his faith, found an unreason for it. Again, the cadence is Nietzschean: ‘Man would sooner will the Nothing, than not will at all.’ Nietzsche, indeed, is a theological nihilist and Eliot is not. Eliot’s God, like Nietzsche’s, died, but then, unlike Nietzsche’s, rose again.

Frazer did not fail to notice the analogy between Christian and pagan materials:

it appears from the testimony of an anonymous Christian, who wrote in the fourth century of our era, that Christians and pagans alike were struck by the remarkable coincidence between the death and resurrection of their respective deities, and that the coincidence formed the theme of bitter controversy between the adherents of the rival religions, the pagans contending that the resurrection of Christ was a spurious imitation of the resurrection of Attis, and the Christians asserting with equal warmth that the resurrection of Attis was a diabolical counterfeit of the resurrection of Christ. In these unseemly bickerings the heathen took what to a superficial observer might seem strong ground by arguing that their god was the older and therefore presumably the original, not the counterfeit, since as a general rule an original is older than its copy. This feeble argument the Christians easily rebutted. They admitted, indeed, that in point of time Christ was the junior deity, but they triumphantly demonstrated his real seniority by falling back on the subtlety of Satan, who on so important an occasion had surpassed himself by inverting the usual order of nature.

Frazer’s chosen stylistic posture is again one of Gibbonian contempt for all concerned – but especially for the Christians. Eliot conversely is like one of the benighted contestants (remember here Eliot’s defence of those who in the Middle Ages read Virgil’s Fourth Eclogue as a prophecy of Christ’s coming). If the Enlightenment had bleached the ancient world, then Nietzsche, more than anyone else, was the agent of its darkening. Yet Nietzsche’s own relation to the Enlightenment is more intimate than is commonly realised. His assault on Christianity preserves many features of the previous Enlightenment assault. The general picture here is indeed of a sunlit, healthful ancient society succeeded by a set of weak, envious, downtrodden persons, followers of a slave ethic which glorified torture and humiliation. This picture can be found in Gibbon, who, as he tells us in his Memoir, was first moved to write his great history by the sight of the barefooted friars going to sing vespers in (what had been) the temple of Jupiter Capitoline. There, as he watched in the evening September sunshine, a gust of indignation ran through him and the idea ‘started into’ his mind. He would tell the story of the decline and fall of the Roman Empire and ‘the triumph of barbarism and religion’. A foul priest-ridden religion of slaves, meeting furtively, displaced the noble civilisation of the Antonines. Gibbon even says that the Christians, finding their poverty inescapable, claimed it as a virtue. This passage immediately follows a quotation from Tertullian’s De Spectaculis – the very same paragraph which Nietzsche quotes at a parallel moment in The Genealogy of Morals.

Notice that Gibbon, like all great historians, is addressing a puzzle: how did the weak defeat the strong? How did this terrible thing happen? This is likewise the grand puzzle of The Genealogy of Morals. The answer there is that the noble blond beast – I have borrowed the blond beast from Twilight of the Idols – may be imagined strolling along, swinging his club, and then falling abruptly into a spiked pit, dug overnight by the runtish slave-men, working away with little spades, co-operating in their rassentiment of strength and beauty. The noble blond beast is unreflective and they are cunning (one is tempted to say, ‘intelligent’). It is of course a picture of the rise of Christianity which has taken quite a beating recently. Robin Lane Fox has shown that early Christianity was really a very middle-class affair.

Nietzsche dedicated Human All Too Human – admittedly in one of his spasmodic reactions against Schopenhauer – to Voltaire, saint of the Enlightenment. The connection between the two is, moreover, corroborated by a minor miracle. On 30 May 1878, Nietzsche received through the post a bust of Voltaire with a slip of paper which read: L’âme de Voltaire fait ses compliments à Frédéric Nietzsche.

The myth of decline into Christianity is an Enlightenment myth. Nietzsche’s kind of anti-semitism (indeed, less violent than that of many of his contemporaries) has affinities with Enlightenment anti-semitism, in that it is cultural rather than racial. Virginia Woolf’s father, Leslie Stephen, once observed that the high priests of 18th-century toleration are united in their hostility to the Jews. The 20th century is shocked by this, but there is a rationale, of sorts. They are against the Jews because they believe the Jews themselves to be enemies of toleration – the source of priestcraft, spiritual tyranny, the frightening of children through guilt. All these things are seen as Hebraic, as counter-Hellenic. It is almost as if, after centuries of being hated as the murderers of Christ, the Jews were now to be hated as the source of Christianity.

Nietzsche’s celebrated evolutionary account of ethics in The Genealogy of Morals is as much an outgrowth from Enlightenment historiography as it is a subversion of it. Indeed, it is interesting to ask, how ethically revolutionary is The Genealogy of Morals? When Nietzsche attacks Christianity on the ground that it is fuelled by hatred and envy, his thought is, it seems to me, ethically entirely conventional. As an account of Christianity it is indeed both disturbing and disruptive – that is quite another matter. He is saying: ‘Christians prate of love but really they are driven by hate.’ But notice how such passage can be read by Christians in a penitential spirit of grateful assent. The reason is simple: Nietzsche here presupposes that love is good and hatred bad. Were this not the case, his words would not constitute an attack. On other occasions Nietzsche really is ethically subversive. Such passages are instantly detectable because they cause the gorge to rise in a quite different way. For example: ‘Not very long ago a royal wedding or a great public celebration would have been incomplete without executions, tortures and autos da fé; a noble household without some person whose office it was to serve as a butt for everyone’s malice or cruel teasing ... To behold suffering gives pleasure, but to cause another to suffer affords an even greater pleasure!’

The surrounding prose is rather slippery but, quite inescapably, Nietzsche is commending this cruelty. At such points his attack on altruism becomes ethically fundamental. There is a shift from ‘Apparently altruistic behaviour is commonly self-seeking’ (which continues to privilege real altruism as a virtue) to ‘Altruism, pity etc are themselves contemptible, forms of weakness, a lowering of health’ (the notorious attack on pity in Antichrist is of this kind). Yet even in these shocking passages, the Enlightenment myth of descent is discernibly still operative.

Given that Nietzsche has chosen, as his set project, to give an account of the genetics of morality, we might have expected a work in which all morality would be explained away – shown to be historically determined by extra-ethical factors, to be a mere fluctuating epiphenomenon upon the struggle for life – or some such. But the story which Nietzsche tells – in which terms for ‘good’ begin as the self-description of the strong, noble people – is presented by him as a sort of ‘Fall-narrative’: an original ethical truth distorted by introversion. In a manner deeply congenial to the 18th century, he is, in effect, using ‘nature’ as a source of prescription: ‘This is what we are, basically, originally; therefore [logical leap] this is the way we ought to be; the first truth is the real truth, the rest lies.’ Nietzsche, it would seem, is profoundly the opposite of Aristotle. Aristotle as a working biologist tends to see things as most fully themselves when they reach the end of a complex development. Nietzsche, conversely, tends to be an eager victim of ‘the Genetic Fallacy’ (‘oak trees are disguised acorns’). It is one of many characteristics which he shares with Freud. Freud once wrote to his fiancée: ‘What we once were’ we ‘in part still remain’.

The Genealogy of Morals (1887) is Gibbonian in so far as it finds health and sunlit splendour in the ancient world, and introversion, priestcraft, asceticism and torture in the post-Classical. Thus far, one might say, Hellas is still bright, is as yet undarkened. Even in the Genealogy, however, we can also find a marked hostility to the objective element in Greek philosophy and this hostility Gibbon could not endorse. ‘Let us, from now on,’ Nietzsche writes, ‘be on our guard against the hallowed philosopher’s myth of a pure, will-less, painless, timeless knower ... All seeing is essentially perspective, and so is all knowing.’ It will be said that Nietzsche is here thinking of Kant, but he must have known that these are ideas having their roots in Plato and Aristotle. There is, in fact, a curious reticence in the Genealogy with regard to ancient philosophy. He is happy to talk of warriors, kings, gods and poets, but the thinkers are, one senses, an embarrassment. Therefore, from an Enlightenment point of view, even in the Genealogy one of the lamps – perhaps the most important one, the lamp of reason – has been turned off.

In the earlier Birth of Tragedy (1870-1) this darkening of antiquity is explicit. The Socratic spirit was fatal, says Nietzsche, to tragedy: the hated Euripides is the poet of ‘aesthetic Socratism’. Moreover, tragedy itself, a high point of culture for Nietzsche, is seen not as Aufklärung but rather as a chiaroscuro, the bright god, Apollo in tension with the dark god, Dionysus. In consequence, Sophocles’s Oedipus loses his marmoreal quality and assumes the lineaments, one can almost say, of Kurtz in Conrad’s Heart of Darkness: ‘It is as if ... the myth whispered to us that wisdom, and especially Dionysiac wisdom, is an unnatural crime, and that whoever in pride of knowledge hurls nature into the abyss of destruction must himself experience nature’s disintegration.’ Set against that passage from Nietzsche this from Conrad; ‘he had stepped over the edge, while I had been permitted to draw back my hesitating foot ... perhaps all the wisdom, and all truth, and all sincerity, are just compressed into that inappreciable moment of time in which we step over the threshold of the invisible ... it was an affirmation, a moral victory paid for by innumerable defeats, by abominable terrors, by abominable satisfactions.’ The ancient association of knowledge with light is here being prised apart. The deepest knowledge is now fused with a formless darkness. The effects of this prising apart upon modern literature are incalculable.

We have reached a point of profound subversion. Yet here also Nietzsche’s thought proves to be not only progressive but violently retrogressive. Plato is not just an inadvertent (retrospectively selected) enemy. Plato himself fought against the doctrines of Heraclitus and Cratylus, philosophers of flux and disintegration. Nietzsche contrariwise (with a little help from Schopenhauer) is fighting for flux and disintegration, against Kant, whom he sees as absolutist. It is worth remembering how Plato opposed Protagorean perspectivalism (see his critique of ‘man the measure of all things’ in the Theaetetus). The speech given to Callicles (despised by Plato) in the Gorgias can be read as a pre-emptive satire on The Genealogy of Morals. Callicles says:

How can a man be happy who is the servant of anything? ... He who would truly live ought to allow his desires to wax to the uttermost, and not to chastise them ... he should have the courage ... to satisfy all his longings. And this I affirm to be natural justice and nobility.

Think also of Thrasymachus in Book One of the Republic.

There is one element in the scheme of The Birth of Tragedy which is endlessly misreported in undergraduate essays and also in published writings. Dionysus, they say, is the god of the dream world. Nietzsche himself says the exact opposite: Apollo is the god of dream. Nietzsche contrasts dream with Dionysiac intoxication. All the rest is as we expect: Apollo is the god of ‘plastic powers’ – of form and of light – while Dionysus is associated with music, which is the expression of Schopenhauerian Will, in all its murky, pre-rational freedom (you will recall that Schopenhauer works with two principles – idea, a formalising agent, and will, which is a sort of existential darkness). I suspect that the tendency of undergraduates to align Dionysus with dream, against the instructions of the author, has something to do with Sigmund Freud, for in Freud the dream-world, though subject to modification, gives us some sense of the anarchic powers of the Unconscious.

You will realise that now, having striven to show an affinity between Nietzsche and the 18th-century Enlightenment, I am now making it my business to show that he is nevertheless more subversive than Freud, who in Civilisation and its Discontents was more than half willing to see the pleasure principle domesticated by a species of Utilitarian calculus. Similarly it is Nietzsche who is more powerfully linked to the disintegrative impulse of Modernism. Freud believes in the reality of the World investigated by science. Nietzsche, in this a true disciple of Schopenhauer, does not. The god of light is for him a god of unreality – of ‘fair illusion’. He equivocates, as many have done before and since, on the Greek word phainomena, ‘appearances’: for the empirical scientist, reality is made of phenomena, but, meanwhile, ‘appearance’ in ordinary usage is naturally opposed to reality. Thus Nietzsche contrasts the core of things with ‘the phenomenal world’. Similarly, he later uses the Kantian phrase Ding an sich ‘thing in itself’, to privilege music above ‘phenomenon’. Once again we must add that Schopenhauer stands between Kant and Nietzsche. Schopenhauer identified the mysterious ulterior reality of the Kantian ‘thing-in-itself’, first with pure will, and then, as we have seen, with music.

One reason why Nietzsche cannot adopt the usual characterisation of Apollo as god of knowledge is that, for him, Apollo is associated with the principle of individuation, itself a source of illusions. Most of us, unsubdued by Schopenhauer, tend to assume that the apprehension of finite existents might give knowledge, but all this is rejected by Nietzsche. The rejection comes very close to generating the weird conclusion: ‘Existents don’t exist; Unbeing alone Is.’ The notion of a potent Nothingness which by a converse energy becomes ontologically stronger than mere things is of course endemic in post-Niet-zschean Existentialism – especially in the (now despised) Sartre. Indeed the history of Structuralism and Deconstruction may be understood along these lines. Lévi-Strauss as a proto-structuralist admittedly discarded as un usable the néant of Sartrean philosophical psychology and chose as his material the constituted self which appears in discourse. But he never loses the sense that such public realities are in some way a series of fictions. Deconstruction, conversely but predictably, reasserts the néant by demonstrating the infinitely recessive character of purportedly rational discourse. Schopenhauer, so to speak, saw all this coming. He is careful to explain how he has a sense for the word ‘nothing’ which is ontologically positive. He writes: ‘We have recognised the inmost nature of the world as will, and all its phenomena as only the objectivity of will ... with the free denial, the surrender of will, all these phenomena are also abolished ... Before us there is certainly only nothingness.’ Hence the bizarre, irresponsible gaiety, founded upon nihilism, which characterises Nietzsche.

On the one hand, we have Nietzsche and his epigoni, Cornford, Jane Harrison, Gilbert Murray, E.R. Dodds, A.W.H. Adkins telling us as a matter of scholarship that the ancient world in general and ancient tragedy in particular were permeated by religion, and on the other hand we have Aristotle. The prime peculiarity for 20th-century readers (though not for any previous century) is this: Aristotle, who was there, appears not to know that Greek tragedy was religious.

In curiously uncritical fashion, we carry in our heads two perhaps incompatible pictures of our own century: first, that we are the first people to perceive the intractably fluid character of all reality, and secondly that we are the first genuinely to understand the past, as something firmly distinct from ourselves. Some readers of The Genealogy of Morals assume, without discomfort, that the demonstration of the perspectival character of all knowledge is there founded – properly and objectively(!) – in a great corrective movement of scholarship: Gibbon’s Romans are 18th-century gentlemen transposed; Tennyson’s Medievals are Victorians at heart, but Nietzsche’s Greeks are Greeks. It is possible, I would suggest, that, on the contrary, the scholarship is infected by the prior espousal of an inversely religious anti-rationalism. Cultural projection is insidious, It can insinuate itself into the most strenuous asseveration of objectivity, it can lurk in the most modest disclaimer. What if it should prove that even the gallant attempt to acknowledge the proper foreignness of ancient culture were really, at bottom, just an up-dated wish-projection? What if the dislodging of the Victorian Greeks of Verrall and Jebb were swiftly followed by a picture of Greek culture which is not pre-Romantic, but rather full of the willed primitivism of the post-Romantic? The 18th century, with good reason, thought itself a critical age. Today it is almost embarrassingly evident where that claim holds and where it fails. What if it were the special prerogative of the 20th-century critic to transform the troglodytic master of Altamira into Picasso?

What, then, of the ritual character of Greek tragedy? What was the ritual? Once again the main line of the modern answer seems to stem from Nietzsche. He says in The Birth of Tragedy that the real ritual subject of Greek tragedy is the sufferings of Dionysus; he anchors this by alleging that, for a long period, the dramas actually dealt with this explicitly as their principal subject.

There is no evidence whatever for this assertion. Moreover the consequences of the theory are, so to speak, internally peculiar. For example, Gilbert Murray found himself arguing that Pentheus in the Bacchae, the priggish, would-be oppressor of Dionysiac worship, ‘is only another form of Dionysus himself’. You will recall that the person who punishes and destroys Pentheus in this play is – Dionysus. On 3 January 1889, Nietzsche collapsed in the Piazza San Carlo, Turin, insane. In the following days he sent off a series of letters to friends, signed either ‘The Crucified’ or ‘Dionysus’. The real structure of Nietzsche’s projected myth of Greek tragedy is almost embarrassingly evident. The last two sentences of the autobiographical Ecce Homo read: ‘Have I been understood? Dionysus was the Crucified.’ To adopt the words of another prophet of the modern European mind, Albert Camus, Nietzsche finds the only Christ we deserve. Nietzsche, whose god, you will recall, died, needed a special resurrection; he needed, so to speak, a Christ with an altered physiognomy:

And what rough beast, its hour come round
                                   at last,
Slouches towards Bethlehem to be born?

From Nietzsche the torch passed to Gilbert Murray. Murray presents the central myth as the story of the Eniautos Daimon, or Year Spirit. The cardinal points of the myth are as follows: 1. Agon (Summer v. Winter). 2. Pathos of the Year Daimon – ritual or sacrificial death. 3. Messenger announces the pathos. 4. Threnos – clash of contrary emotions, the death of the old being also the triumph of the new. 5. Anagnorisis of the Slain Daimon 6. Epiphany and Theophany-in-glory of the Slain Daimon.

Murray seems to have derived these features of the archetypal ritual principally from a study of the extant, ectypal plays. So it is not really surprising that the whole picture is rather suggestive of a Greek tragedy. What is surprising (in the circumstances) is that no extant play contains an epiphany of the protagonist, as Pickard-Cambridge pointed out in his complete (though largely ignored) demolition of Murray’s theory. Pickard-Cambridge allows, however, that certain features of Murray’s postulated ritual were performed ceremonially at different points in the year. But there is, in fact, no sign of any unified ritual. As for the phrase Eniautos Daimon (Year Spirit), we can go a little further. It is not just that it occurs nowhere in the source material: it could not occur, because it is not possible Greek.

I am saying that it is not only when we assume likeness to ourselves that we distort the past. The very assertion of unlikeness – especially a weirdly symmetrical unlikeness – can prove, at a deeper level, to be more in accord with our wishes than with external fact. The 20th century has been marked by a kind of inverse Narcissism – or by the confounding of the parallel Ovidian myths of Narcissus and the Cyclops. Gazing into the pool of history, it sees, or chooses to see, not the fair face of civil humanity, but the shaggy features of the Satyr.

It may be thought that, even if my argument has cast doubt on the detailed historical argument of The Genealogy of Morals, the effect in the long run is not to subvert but rather to confirm Nietzsche’s major thesis, which is precisely a denial of objectivity. That has not been my intention. I have sought to direct attention not to some great, pulsating Schopenhauerian negation, but rather to the difficulties of ordinary, piecemeal ignorance. Spectacular cultural subjectivism could never be detected had we no access at all to an object. Not for us, indeed, the unshadowed panoramic clarity of Leonardo’s sun-centred Eye. Nor are we offered the converse intoxication of a radical nihilism. We blunder, then correct; we deceive ourselves, but sometimes detect the imposture. It may sound a lot less exciting than the metaphysical extremes, but I firmly believe that, in the long run, it is much more interesting.

Freud hedged his bets. As we saw, with half his mind he went for controlling form, but with the other half – I think we may say, with the excited half – he went for real poetic force linked to a potentially destructive libidinal drive. Artists are distinguished from neurotic dreamers by their ability to impose an acceptable form; but, meanwhile, Hamlet is great because of its latent libinal content. Contrary to popular belief, the Freudian Death Instinct is not primarily a wish for one’s own death but a desire to inflict death on others. Here the rough beast, Apollo’s enemy, is given its head. But when we watch a tragedy, it seems clear, first, that we identify with the protagonist, and second, that the pleasure of the tragedy is inextricably bound up with a consequent pity, rather than simple exultation over a fallen victim. This fact of identification with the protagonist seems to belong more with the central energy-charge of the drama than with the mitigating formal aspect. The Freudian may say that the pity is itself a disguise-mechanism, to render the pleasure (born in an original, hidden cruelty) morally acceptable to the conscious subject, but at this point in the discussion the ordinary playgoer may begin to rebel. How do we know that the psychoanalyst is not simply constructing in the conveniently uninspectable area of the Unconscious a solution which is a mere serviceable fiction, its form dictated not by available evidence but simply by the structure of the original problem? ‘Pleasure in tragedy looks like cruelty? Why, then, we are cruel, but unconsciously so.’ Meanwhile, those who attend to the oikeia hedone, that strange sweetness of grief and fear, will wish to look further.

Nietzsche, somewhat differently from Freud, suggested that tragic pleasure might spring from our sense of re-immersion in pre-individuated consciousness – in the orgiastic unity of the multitude – as the hero dies. This idea has the merit of being founded on a close and intelligent engagement with the actual poetic impact of Greek tragedy. It is strikingly true, for example, that the choruses of Greek tragedy, with their Doric dialect, their orgiastic, heavily syncopated rhythms, exist in a kind of tension with the very different verse uttered by the individual persons of the play. But Nietzsche also says that the dying protagonist is always a form of Dionysus himself and that Dionysus represents the principle of anti-individuation. Surely, on Nietzsche’s theory, we cannot rejoice at the demise of anti-individuation. Nietzsche ought to have argued that tragedy dramatised the death of Apollo. There is in fact no evidence for either thesis, of course.

Once more, one suspects that the Zeitgeist is dictating the terms. Nietzsche, like a dozen other anguished, over-introspective post-Romantics, feels a nostalgia for a supposed primal unity. But Greek tragedy, I would venture to suggest, was itself actually pressing in the opposite direction – becoming more and more absorbed by the isolated self standing apart, in contradistinction to the group. We are driven back to our original problem: back to Aristotle.

I suspect that one thing which makes the Nietzschean and Freudian accounts of the pleasure of tragedy persistently interesting – and I say this in spite of the recent surge of formalism in literary studies – is that they offer a solution rooted not in form but in underlying substance, in human nature. The Aristotelian account, on the other hand, seemed to place its main emphasis on form, and on the palpable unreality of tragedy. Is there a way of reconstruing – or reconstructing – Aristotle’s theory in order to give some weight to human substance? I would suggest that for ‘catharsis’, ‘purgation’, we substitute ‘exercise’. ‘Catharsis’ implies a passive experience, a mere loss of dangerous emotion; ‘exercise’ implies an active use of emotion. It has been suggested that in frightening dreams the subject can, as it were, experience disaster without actually experiencing it (I am exploiting an old ambiguity in the term ‘experience’). In this way, perhaps, we are able to practise for crises. When the real emergency arrives, it is not, so to speak, our first time. I do not know whether this general theory is true, but I do want to suggest that something of the sort may happen with tragedy. For the process to work, two things need to be the case: first, the situation must be hypothetical rather than categorical (as football is hypothetical warfare, not actual), and second, that it should nevertheless involve a probable relation to real danger; if there is no probability to face, the exercise will not be testing, will not energise. The simultaneous presence of hypothesis and probable mimesis means that our theory is still Aristotelian, but I have now substituted an active term for the passive ‘catharsis’. The human capacity to think provisionally, to do thought experiments, to form hypotheses, to imagine what may happen before it happens, is fundamental to our nature and to our spectacular biological success (so far). I think the cleverest thing Sir Karl Popper ever said was his remark that our hypotheses ‘die in our stead’. The human race has found a way, if not to abolish, then to defer and diminish the Darwinian treadmill of death. We send our hypotheses ahead, an expendable army, and watch them fall. It is easy to see how the human imagination might begin to exhibit a need, in art, for a death-game, a game in which the muscles of psychic response, fear and pity, are exercised and made ready, through a facing of the worst, which is not yet the real worst. ‘The worst is not so long as we can say “This is the worst!” ’ (King Lear)

When hypothesis lapses into actuality one has indeed a corruption of tragedy, and the element of self-trial is replaced by simple cruelty. There is a horrible poem by Martial, about Prometheus chained to his rock (as he is at the beginning of the Greek tragedy). The poem explains how a criminal called Laureolus was actually crucified, in a parallel dramatic presentation: Non falsa pendens in cruce Laureolus (On no pretended cross he hung).

The mythological dramas of the Roman arena abolished the space between hypothetical and actual suffering: Ixion on a real burning wheel, one playing the part of Mucius Scaevola burning his hand off in a brazier, the shirt of Nessus by which Hercules meets his agonising death in Sophocles’s Trachiniae – ‘the intolerable shirt of flame’ – realised in the unspeakable tunica molesta of the circus. A snuff tragedy is not a tragedy.

But what about our own horror videos? This, I confess, I find difficult. Setting aside alleged cases in which actual death or mutilation is exhibited (these, I take it, would be directly analogous to performances in the Roman arena), we are left with a very large number of people enjoying the spectacle of simulated horror. I do not believe that this pleasure need be sadistic, any more than that the tragic pleasure need be so. Indeed, I strongly suspect that my notion of energies quickened by a kind of psychic exercise applies here also, but in an arrested form: I suspect, that is, that in the horror video the hypothetical experience is not carried through to its fully human conclusion, in pity and fear. The notion of a facing of the worst, the cognitive element, is only half-present.

In his poem ‘Lapis Lazuli’ Yeats baited the moralising figures of his day with the pleasure of tragedy. ‘Hamlet and Lear are gay,’ he wrote, deliberately choosing the most irresponsible word (though little suspecting the meaning it would carry in 1992). And in ‘An Irish Airman Foresees his Death’ – also against the grain of high moral sentiment – he wrote:

Nor law, nor duty bade me fight,
Nor public men, nor cheering crowds,
A lonely impulse of delight
Drove to this tumult in the clouds.

It is an attitude sometimes described as aesthetic, but it does not circumscribe art, in separation from life. Rather, it links art to a very serious, incorrigibly pleasurable game of death. Many games are exciting. Tragedy differs from most in that it requires a peculiar stillness in the watcher together with strenuous activity in that watcher’s sympathetic imagination. Above all, it leads to a conclusion (which the airman, as long as he is ‘dicing with death’, must avoid). Of all the literary genres, tragedy is the one which lays the heaviest emphasis on ending, and the ending is a mimesis of a death. In so far as we sympathise, we experience the dying, but of course we do not die. The hypothesis dies instead.

We should remember that Aristotle’s catharsis, a term we have rejected, was coined in part to account for the difference between the behaviour of a sporting crowd and that of a theatre audience. But perhaps the notion of an accomplished conclusion will help here also. Football matches end only because an allotted time runs out, and the excited spectator may well come away with a sense of unfinished business.

I have argued for a basis in human nature for tragic pleasure, but of course form has its part to play. It may be that the special impact of tragedy lies in an eerie coincidence of depicted story and psychic effect. I have said that the spectator achieves a moment of recognition, faces a truth known to be necessary for all. Meanwhile, within the fiction, the protagonist is commonly brought to a point of crucial recognition and insight. Sophoclean tragedy is like this. Shakespeare, on the other hand, might prove troublesome.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.


Vol. 14 No. 12 · 25 June 1992

A.D. Nuttall (LRB, 11 June) argues that Greek tragedy was non-religious, by which he appears to mean that the myths that provided its plots had lost all religious efficacy. Here he opposes the view of Gilbert Murray and Jane Harrison and others that the tragedies still functioned as enactments of the sacrificial death of Dionysus. Nuttall says that Pentheus in the Bacchae cannot be intended as a Dionysus surrogate, being an opponent of Dionysus worship. Apparently Nuttall is impervious to the irony of the sceptic suffering the hallowed death that he refused to acknowledge as salvific. But such irony is not uncommon in myths: Neoptolemus, for example, having insulted Apollo in his shrine, dies a sacrificial death and becomes a tutelary spirit of the very same shrine. Such ‘guilty victim’ sacrifice must comprise one of the earliest devices for making sacrifice morally bearable (note, for example, how the animal-victim in the Bouphonia sacrifice was enticed to ‘sin’).

Nuttall uses the lack of a happy ending as an argument against the religious character of Greek tragedy, urging that, if Dionysus were the inspiration, the plays would end with a resurrection. But this is to ignore the independence of each stage in the death-and-resurrection cycle: as Morton Smith has pointed out, while the mourning of the god is taking place, it would spoil the effect to say or think anything about his subsequent resurrection. Something of this emotional and dramatic dislocation can still be seen in the separation of Good Friday from Easter Sunday. The Gospel of Mark, in its original version, contained no account of the resurrection. It was a Good Friday gospel.

Nuttall’s clinching argument is: ‘Aristotle, who was there, appears not to know that Greek tragedy was religious.’ But in his Poetics, Aristotle was temporarily unconcerned with the religious aspect while focusing on aesthetics. It is from Aristotle that we know that tragedy was derived from the Dionysiac dithyramb, and comedy from phallic fertility songs. His concern is not to deny this religious dimension but to show how the drama became aesthetically refined. After all, it is difficult to believe that Aristotle could totally discount the fact that the tragedies formed part of a religious festival. He does indeed use language elsewhere that suggests an awareness of a deep connection between religion and aesthetics: in writing about the mysteries, he says that ‘the initiates were not required to learn anything but to experience certain emotions and to be put in a certain disposition’. This throws light on his famous saying that tragedy effects ‘a purgation’ – or purification – ‘of the emotions of terror and pity’. The initiates experienced terror and pity, and so did the audience at the tragedy; both underwent a ‘purification’ leading to a kind of rebirth. The Medieval Passion Plays provided a similar religious experience and sense of participation. The theory that tragedy originates in, and remains informed by, religious notions and rituals of sacrifice has a great deal to support it. It will hardly be shaken by Nuttall’s somewhat pedestrian argument.

Hyam Maccoby
Leo Baeck College,

Vol. 14 No. 14 · 23 July 1992

In reply to Hyam Maccoby (Letters, 25 June): 1. I did not say – and do not think – that the myths which provided the plots of Greek tragedy had lost all religious force. 2. I do indeed reject the view that the tragedies we possess were ritual re-enactments of the death of Dionysus. As E.R. Dodds used to say, the essence of ritual is repeatability. The Mass, variously performed in various places, remains one thing (compare ‘When you’ve seen one ritual murder, you’ve seen the lot’). The Dionysia were clearly not a ritual of this kind: year after year, different structures were presented, often to the amazement of the spectators (yes, ‘spectators’, not ‘participants’; the Greek word which gives ‘theatre’ is a place where people watch). Indeed, the fact that both the festival and the theatre bear the name of Dionysus makes it the more striking, in my view, that the tragedies range freely over the available myths. This is odd, if you like, but not mind-numbingly odd. If I may offer an admittedly frail analogy, it is as if we had in London a St George’s Theatre and a St George’s Day Festival at which all sorts of plays were performed. Of course it is possible, by metaphor, to interpret all Greek tragedies as accounts of the death and resurrection of Dionysus, but it would be equally possible to show, by like means, that all Greek tragedies are about Heracles (or Apollo, or Orestes, or Athene …). My thought was doggedly elementary: which is the more appropriate term to apply to these texts, ‘ritual’ or ‘play’? I go for ‘play’. That said, Maccoby is right to say that Greek religious feeling is powerfully present. Eliot’s Murder in the Cathedral is brimful of Christianity. But it isn’t a ritual.

A.D. Nuttall
New College, Oxford

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN


Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences