Art and Mimesis in Plato’s ‘Republic’

M.F. Burnyeat

Plato is famous for having banished poetry and poets from the ideal city of the Republic. But he did no such thing. On the contrary, poetry – the right sort of poetry – will be a pervasive presence in the society he describes. Yes, he did banish Homer, Aeschylus, Sophocles, Euripides, Aristophanes – the greatest names of Greek literature. But not because they were poets. He banished them because they produced the wrong sort of poetry. To rebut Plato’s critique of poetry, what is needed is not a defence of poetry, but a defence of the freedom of poets to write as, and what, they wish.

No big problem, you may think. But suppose poetry was not the minority pursuit it has become in Britain today. Suppose it was the most popular form of entertainment available, the nearest equivalent to our mass media. That is not far from the truth about the world in which Plato wrote the Republic. The Athenian democracy, audience for much of the poetry Plato objected to, accepted that it was their responsibility to ensure the quality of the poetry funded by the state. In modern terms, they thought that democracy should care about whether the mass media encourage the right sorts of values. Do we want Rupert Murdoch to determine the overall quality of the culture? Should money decide everything? If not, what can we do about it?

Plato was no democrat, and had no qualms about proposing Soviet-style control from above, by those who know best. But democrats who reject such authoritarian solutions may still learn from Plato’s disturbing presentation of the problem. What he is chiefly talking about is the words and music by which the culture is transmitted from one generation to the next. Tragedy and comedy were performed before a crowd of 14,000 people at the Great Dionysia and other civic festivals. We hear of 20,000 people attending a recital of Homer. Then there are hymns sung at religious ceremonies and songs at feasts or private symposia. Forget about reading T.S. Eliot to yourself in bed. Our subject is the words and music you hear at social gatherings, large and small. Think pubs and cafés, karaoke, football matches, the last night of the Proms. Think Morning Service at the village church, carols from King’s College Cambridge, Elton John singing to the nation from Westminster Abbey. Think popular music in general and, when Plato brings in a parallel from the visual arts, forget the Tate Gallery and recall the advertisements that surround us everywhere. Above all, think about the way all this is distributed to us by television, the omnipresent medium at work in every home. What Plato is discussing in the Republic when he talks about poetry is how to control the influences that shape the culture in which the young grow up. How to ensure that what he calls the ethos of society is as ideal as possible. Even as adults, none of us is immune.

Books II-III of the Republic present Plato’s proposals for reforming the culture in a carefully arranged sequence of stages. The first stage concentrates on the content of musical poetry, the last on its material and social setting – with special reference to the symposium or drinking party. In between come various other elements of poetic performance. This sequence of stages is not a sequence of independent topics. Each should be thought of as one layer among others in the analysis of a single cultural phenomenon: the performance of poetry with music (and sometimes dance as well).

From time to time the discussion touches on a non-musical topic, be it nursery tales or the content of the visual arts. But the central thread is the performance of musical poetry at a social gathering. This for Plato is the main vehicle of cultural transmission. This is what he is trying to get right when he designs a musical education for the warrior class in the ideal city – the Guards, as they are called, from whose ranks a select few will go on to become philosopher rulers. All else is subordinate.

One further preliminary. Plato is well aware that what he has to say will shock and appal his readers, then as now. His proposals for the ideal city amount to a complete reconstruction of Greek culture as it existed in his day. What motivates the proposals is his profound understanding of the many subtle ways in which the ethos of a society forms the souls who grow up in it. If you shudder at the authoritarianism of his programme, remember that shudder when the newspapers next debate whether bad behaviour in schools is the fault of parents or teachers. As if parents and teachers were anything but a tiny facet of the total culture of our time. Either grasp the nettle of devising democratic alternatives to Plato’s authoritarianism, or stop bleating.

Plato’s first charge, and perhaps the most shocking to ancient readers, is that, from Homer onwards, poetry has been full of lies about the gods. The entire religious and mythological tradition stands condemned for blasphemy. It is like someone today proposing to ban the Bible and all reference to Biblical stories, because the Bible presents a wrong picture of divinity. None of the stories of God’s dealings with humankind can be true; and even if some of them were true, they are morally unsuitable for young ears.

And what is unsuitable for ears is unsuitable for eyes as well. Stories it is wrong to sing, like the battle of gods and giants, must not be represented in embroidery. This is no joke. Plato’s readers would think at once of the colossal embroidered robe (πέπλος) carried in procession at the festival of the Panathenaea. The robe showed the battle of gods and giants, spotlighting the victory of Athena over the giant Enceladus. A ban on such embroidery is a stake through the heart of Athenian religion and Athenian civic identity. Though Socrates does not stop to mention it, the censorship of embroidery will inevitably extend to painting and sculpture. The battle of gods and giants will be removed from the carved metopes of the Parthenon (currently on display in the British Museum). In the ideal city, the religious content of the visual arts will be as restricted as that of poetry and music.

How much Greek literature would survive enforcement of the following norms? (1) Divinity, being good, is not responsible for everything that occurs, only for the good. So gods never lead mortals into crime. (2) Divinity is simple, unchanging, and hates falsehood and deception. So gods never appear in disguise to mortals, never send misleading dreams or signs. (3) Hades is not the dreadful place the poets describe. So a good man finds no great cause for grief in the death of himself, his friend, or his son. (4) Heroes are admirable role-models for the young. So they never indulge in lamentation, mirth, or lying (save for high purposes of state), impertinence to their commanders or arrogance towards gods and men, sexual passion or rape, longing for food and drink, or greed for wealth; nor, mutatis mutandis, should any such thing be attributed to the gods. Finally, (5) the moral argument of the Republic itself, when completed, will prove that it is justice, not injustice, that makes one happy. So no poet may depict a happy villain or a virtuous person in misery. Under this regime very little of the Greek literature we know would remain intact, and much of the art would disappear.

Nearly all the poetry cited in the Republic so far will be banned. Many of the themes of the earlier discussion came from poetry, because poetry articulates the values and beliefs of the culture. In Book I, Cephalus recounts how, when old age comes and death is near, one begins to take seriously the stories about Hades and the terrors it holds for wrongdoers. In a society with no Bible or canonical sacred text, the chief source for these stories is poetry. Conversely, it is poets like Pindar who hold out the hope of a pleasant afterlife for those who have lived in justice and piety. On the other hand, a major theme of Adeimantus’ speech at the beginning of Book II is the way the poets instill in the young a wrong attitude towards justice, because they praise it for its contingent consequences rather than its intrinsic value. Justice, the poets say, is a real sweat in this life, much harder and less pleasant than injustice (provided you can get away with it). It is only in the very long run that justice pays: the poet Musaeus, for example, promises the righteous that their afterlife will be an unending symposium, as if the ultimate reward for virtue was eternal intoxication. But at the same time his teaching is that the wicked can always bribe the gods with sacrifices and festivals to let them off. None of this is compatible with the norms that Socrates has now put before us.

To begin with, however, Socrates speaks as if he is merely purging the culture of certain objectionable features. He asks Homer and the other poets not to be angry if he and Adeimantus expunge all the passages that breach the norms. He takes the scissors to Aeschylus, but implies that tragedy (cleaned up by himself) will still be performed. At this stage, Plato is concerned only with the content of the arts, especially their religious content. Like many later (and earlier) religious reformers, he will have his new orthodoxy, utterly different from traditional Greek religion, rigorously enforced throughout the society. The next stage of the discussion, concerned with the manner of poetic performance, will justify banishing Homer and the tragedians altogether.

But already it is clear that the norms for art in the ideal city will reshape the whole culture. Students of Plato are sometimes told they need not be shocked by the censorship advocated in Republic II-III, because its target is the education of young Guards, and any responsible parent today keeps watch on the entertainment and reading-matter of young children. The proposals are made for the sake of the young. But Plato’s insight is that if you are concerned about the souls of the young, it is no good simply laying down rules for parents and teachers, or agreeing to keep sex and violence off the TV screen until after 9 p.m. His conclusion: for the sake of the young, the entire culture must be purged.

The text makes this quite plain. The stories which must not be told to very young children by nurses and mothers should not be heard anywhere in the city – or if at some ritual they have to be told, the audience should be kept as small as possible. Conversely, once we have the right kind of stories for the very young, we will compel the poets to tell them the same kind when they grow older. The norms about the representation of divinity apply to all poetry, whether epic, lyric or tragic: epic and tragic meters are primarily used for public occasions, while lyric is for smaller group gatherings like the symposium. And things that must not be said in verse must not be said in prose either, must not be said or heard by anyone in the city, young or old. They are not fit for the ears of boys or men. Such things are not merely false, but impious, and therefore harmful for anyone to hear. The one mention of schoolteachers is a sharp passage at the very end of Book II, referring to some objectionable lines of Aeschylus: ‘When anyone says such things about the gods, we shall be angry with him, we will refuse him a chorus, and we will not allow teachers to use him for the education of the young.’ Nothing is to be put on in the theatre unless it is fit for classroom use afterwards. The Greek word παιδεíα means both culture and education. Plato’s message is that culture should be taken seriously for what it is: education.

Yet telling false, blasphemous, immoral and passionate stories is not the worst thing a poet can do, in Plato’s opinion. Such stories corrupt the young by filling their minds with dangerously wrong ideas about matters of great moment. But a more enlightened, grown-up mind, with the aid of philosophy, may come to reject the community’s religious narratives, as Socrates does in the Euthyphro. Stories as such are something a rational mind can resist, question and reject. With visual images and likenesses in sound and music, resistance is not so easy. The manner of poetic performance is more insidious than the content. Even the best philosophical minds are at risk.

The advanced industrial countries of the West have fewer occasions for community singing than more traditional societies, but one that survives is Christmas:

Once in Royal David’s city
Stood a lowly cattle shed,
Where a mother laid her baby,
In a manger for his bed.
Mary was that mother mild,
Jesus Christ the little child.

This carol is a third-person narrative, all the way through. Listeners hear about the birth of Jesus. But when someone reads the Lesson from the Gospel and their voice modulates to express kindness or anger in words that Jesus speaks in the first person, or when in Bach’s St Matthew Passion Jesus sings those words in recitative – then it is mimesis. We do not merely hear about the Son of God. In a certain sense, we hear him. We hear him in the same sense as we see him on the Cross in a picture of the Crucifixion.

In Book X of the Republic, painting is the paradigm Plato uses to explain the meaning, and the menace, of poetic mimesis. His example is a painted couch, and the point he emphasises is that the picture shows only how the couch appears when viewed from a particular angle – from the side, the front, or some other perspective. Christ on the Cross is also seen in a fixed perspective. But that does not stop us saying we see him there. Just so, the painting of a symposium reproduced here (by the Brygos painter, c.490 BC) is not particularly naturalistic, but anyone will say, if asked what it represents, ‘I see couches and tables, a lyre and pipes for the music, people enjoying a party.’

It is the task of the philosophy of art to explain what grounds this way of speaking, why it is not only possible but the correct thing to say in the presence of a wide range of representational painting. My interest here is in what happens when the same language is applied to the likenesses of poetry and music.

Back to the St Matthew Passion. As in a rhapsode’s recital of Homer, there is a narrator (the Evangelist) to tell the story, and speeches sung in recitative by the different characters. There is also a Chorus, which plays two roles. It is both the jeering voice of the crowd hostile to Jesus and, in the Chorales, the voice of the Congregation reacting to the events with sorrow and repentance for what humanity did to the Son of God. This dual role expresses rather well the idea I think is fundamental to mimesis, that the audience – in this case, the Congregation – is actually present, in a certain sense, at the events depicted. They do not merely hear about them. In a Greek tragedy the Chorus has a similar dual role, both participating in the drama and voicing the audience’s reaction. The Athenians did not merely hear about Antigone’s conflict with Creon. In a certain sense, they witnessed it.

We may find it easier to speak of seeing Jesus in a picture than of hearing him in Bach’s music. Plato relies on the analogy with painting to make his point vivid. But no help is needed when we move to opera, which began as Monteverdi’s and others’ attempt to re-create the multi-media experience of Greek tragedy, where speech (for the iambic verse) alternated with flute-accompanied recitative or lyric choruses sung and danced. We do not merely hear the characters of an opera, as in the St Matthew Passion. We also see them – moving, dancing, fighting, dying; not motionless as in painting and sculpture. The absence of a narrator is another contrast with the St Matthew Passion. Tchaikovsky’s Eugene Onegin cuts out the narrator whose ironic commentary is crucial to Pushkin’s poem, and shows us Tatiana herself in the intimacy of her bedroom, writing the fateful love-letter. Afterwards we see and hear Onegin crushing her hopes. At the end we see and hear Onegin declare his love – too late. It would be ridiculous to refuse to describe the opera-goer’s experience in these terms; absurd to insist that all we see and hear is singers playing their parts. As Stanley Cavell said in reply to a parallel suggestion about film: ‘You might as well tell me that I do not see myself in the mirror but merely see a mirror image of myself.’

It is this sense of being present at the events enacted on stage, not merely at the theatrical event of enacting them, that Plato aims to capture when he introduces the concept of mimesis. Mimesis is the production of visual and auditory likenesses which give us that sense of actual presence.

For the second stage of the discussion of the Guards’ musical education in Republic II-III, Socrates turns from the content of poetry to the manner of its performance. He introduces a distinction, which at first Adeimantus is slow to grasp, between mimetic and non-mimetic storytelling. I take Adeimantus’ initial slowness as Plato’s signal to his readers that the distinction will be new to them. ‘Mimesis’ is of course an ordinary Greek word, meaning ‘imitation’, but the distinction between mimetic and non-mimetic storytelling cuts across the more familiar classification by poetic genres. The distinction is probably Plato’s innovation.

Non-mimetic storytelling is third-person narrative, as in ‘Once in Royal David’s City’ and ancient dithyrambic choral singing. The Iliad starts out that way, but at line 17, Chryses, the Trojan priest of Apollo, speaks to Agamemnon, Menelaus and the Greeks, imploring them to release his daughter. His words are in direct speech: ‘you’ and ‘I’ replace the pronouns ‘they’ and ‘he’ of the preceding narrative. Here is how Socrates describes the difference:

You know then that up to these verses, ‘and he made prayer to all the Achaeans,/But especially to the two sons of Atreus, the marshallers of the host,’ the poet himself is the speaker. He does not try to divert our mind into thinking that someone else is speaking. But the following verses he delivers as if he were himself Chryses. He tries his best to make it seem that the person speaking is not Homer but the priest, an old man.

Much more is packed into the concept of mimesis here than results from the change of pronouns.

When I read the Iliad to my children at home and came to the words of prayer at line 17 –

Sons of Atreus and the rest of you strong-
                                   greaved Achaeans,
May the gods who dwell on Olympus grant
That you sack the city of Priam and return safe
                                                to your homes;
But release my dear daughter to me, and
                                accept the ransom,
Out of awe for Zeus’ son Apollo, who strikes
                                                      from afar.

– I did not put on a quavering voice to make it seem that an old man was speaking. Evidently, Socrates has in view a performance of some kind, not just reading aloud to an audience. A performance that involves impersonating an old man or some other character-type.

The performer Socrates talks about is Homer, the poet himself. But he is long dead. What Socrates and Adeimantus are actually familiar with is rhapsodes reciting at the festival of the Panathenaea from the official Athenian text of Homer, fixed a hundred years earlier by order of the tyrant Peisistratus. The rhapsode Ion is about to do just this in the dialogue Plato named after him. But the message of the Ion is that the rhapsode is a mere mouthpiece for the poet. The poet’s voice speaks through his, as the Muse speaks through the poet. There is a chain of inspiration, which Socrates compares to a chain of iron rings suspended one after another from a magnet, through which the divine power pulls the audience’s emotions this way and that. So when Ion recites, it is the divinely inspired Homer we hear.

This makes the rhapsode rather like an actor, voicing someone else’s words. In the Ion he is pictured in terms that bring to mind a modern pop-singer: up on a dais in extravagant clothes before a festival audience of over twenty thousand people, he chants the verse, melodiously and dramatically, with tears in his eyes during the sad bits. The innovation in the Republic is Socrates’ stress on the way ‘Homer’ modulates his voice or diction (λέξις) so that it becomes like that of an old man praying. The poet-performer ‘hides himself’ and does everything he can to make it seem that Chryses is present to our ears.

From this introductory example Socrates proceeds to a generalization that covers visual as well as auditory likenesses. It is mimesis, he says, if the poet likens himself to someone else either in voice or in σχημα. Σχημα can refer to gesture, posture or movements, including the movements of dance. This extends the concept of mimesis to the silent miming (as we still call it) of Jean-Louis Barrault in Les Enfants du paradis, or the dance and music of modern ballet. For a case fulfilling both clauses of the disjunctive generalization, imagine a performance where not only the rhapsode’s voice, but also his gestures, posture, perhaps even some movements, are like those of an old man’s supplication. He goes down on his knees (rather stiffly) and stretches out his hands. Chryses seems to be present to our eyes as well as our ears.

The generalisation still does not provide a definition of mimesis, only a sufficient condition. Socrates will not offer a general, explanatory account of mimesis until Book X. We have to catch on piecemeal as he adds in new types of example. Next come tragedy and comedy, which are entirely mimetic, without any narrative in the poet’s voice. Yet Socrates continues to speak of the poet as the imitator. Just as Homer speaks through Ion, so in drama it is the poet who tells the story – through his characters’ speeches. It is as if the actors, like the rhapsode, are mere conduits for the poet’s own voice. Euripides speaks the words of Medea, his voice modulating like a ventriloquist’s into that of the (male) actor playing the part.

This way of thinking about actors as extensions of the poet is taken further when Socrates goes on to say the Guards should not imitate neighing horses, lowing bulls, the noise of rivers, the roar of the sea, thunder, hail, axles and pulleys, trumpets, flutes, Pan-pipes and every other instrument, or the cries of dogs, sheep and birds. Is he talking about some crazy pantomime, in which people mimic everything under the sun, including axles and pulleys? Or about the dramatist’s use of sound-effects? I suggest the latter. In Aristophanes’ Frogs the Chorus croak ‘Brekekekex, koax, koax’ – after all, they are a chorus of frogs. If the imitator is taken to be the poet rather than the actors, then it is Aristophanes himself who makes these noises, while his voice modulates into the trumpets and flutes of the accompanying music, or rumblings from the thunder-machine off-stage.

If you find it grotesque, this picture of the poet sprouting extensions of himself and his voice all over the theatre, Plato will be well pleased. His point is to forbid the Guards to engage in dramaturgy. They must practise one craft only, that of defending the freedom of the city. They are not even to do what cultivated Athenians often did, combine their main pursuit with the writing of tragedies. (In real-life Athens, Sophocles did it the other way round: he served twice as general.) The ideal city is founded on the principle that each man devote himself to a single craft.

In itself, this is not an argument for a ban on purely mimetic storytelling. There are lots of things the Guards must not do which, nevertheless, someone in the ideal city has to do: pottery and painting, for example. But the ‘one man – one job’ principle can be reapplied to block the suggestion that, provided he made tragedy or comedy (not both) his specialty, a professional dramatist could be admitted into the city. The ideal city is like a symphony orchestra, in which each member plays just one instrument, so that together they create a beautiful whole called ‘Kallipolis’. The dramatist is a walking-talking-singing-trumpeting-thundering subversion of the ‘one man – one job’ principle responsible for this happy result. Not only must no Guard write plays, but if a professional dramatist turns up at the city gate and asks to present his works, he will be treated as if he were a one-man band at the street corner asking to join the Berlin Philharmonic. It is not even lawful for such a multiplex personality to grow up within the ideal city, let alone for one to be let in.

You may object that a professional dramatist does not really exhibit the multiple-personality disorder Socrates ascribes to him. He only seems to do so. Plato knows this very well; in Book X he will insist on it. But he also knows that ‘imitations, if continued from youth far into life, settle down into habits and (second) nature in one’s body, voice, and thought.’ In John Banville’s novel The Untouchable, a young recruit to MI5, out on his first assignment and moving in to detain the spy for questioning, ‘narrows his eyes as the thrillers had taught him to do’; by the time he retires, that eye-movement will be second nature to him (thereby proving the realism of the next generation of thrillers). Imitation may have consequences. It is not a thing to take up lightly, still less to make a profession of. Some film stars have been said to lack a stable self of their own, to live only in the public appearance of a bundle of different roles. Given Plato’s conceit of the actors as so many extensions of the poet, for him it is the dramatist who is like that. Not a person who will contribute to the austerely civilized life of Kallipolis.

At this stage, then, Plato’s objection is to the dramatist rather than the drama. His ban on dramaturgy (amateur or professional) is not primarily due to concern about what will happen to the souls of Guards who recite speeches from Euripides or act in his plays, nor to worries about Euripides’ effect on the souls of his audience; this will be discussed in Book X. In Book III the decision is political. Euripides is an undesirable character to have around; so are politicians and military men who write plays in their spare time. Plato here is like someone who would ban rock music not because of its heavy beat and racy words, but because of the singers’ lifestyle. And beware of politicians (like Tony Blair and Bill Clinton) who play musical instruments.

Contemporary readers would be sensitive to the political aspects of Plato’s decision. Athenian tragedy and comedy were intensely democratic institutions, not only in the way they were organized, but also in their physical presence. During the Great Dionysia, 1200 citizens – 700 men plus 500 adolescents – took part in the choral singing and dancing of the various competitions (tragedy, comedy, dithyramb). Under Pericles’ cheap ticket scheme, even the poorest of the rest could join the audience, which was further swollen by visitors from the Empire and abroad, reaching a total of between ten and 14 thousand people. In oligarchic Sparta there were choral festivals, but no theatre. The link between theatre and democracy is not explicit in Book III of the Republic, but elsewhere the connection is loud and clear.

Book VI includes a discussion of what is likely to happen if, in a non-ideal state like Athens, a truly philosophic nature is born, capable of becoming one of the philosopher-rulers of the ideal city. Would the young man escape the corrupting influence of the culture under which he grows up? The chances are small, says Socrates. Think of the impression made on a really talented soul by the applause and booing of mass gatherings in the Assembly, the courts (an Athenian jury was not 12 good men and true, but several hundred and one), theatres and military camps. Is not the young man likely to end up accepting the values of the masses and becoming a character of the same sort as the people he is surrounded by? A democratic culture does not nurture reflective, philosophical understanding. Mass gatherings set the standards of goodness, justice and beauty, in painting, in music (where ‘music’ includes poetry and drama) and in politics. Plato knows all about democratic control of the general quality of the culture; in the Laws he will call it ‘theatrocracy’. His vitriolic denunciation of the mass media of his age argues for rejecting democratic control in favor of his own, authoritarian alternative.

Even stronger is the claim at the end of Republic VIII that tragedy both encourages and is encouraged by the two lowest types of constitution, democracy and tyranny. Note the interactive model of cultural change. As in a bad marriage, playwright and polity bring out the worst in each other. Each indulges the other’s ways.

So what occasions for the performance of poetry will remain in the ideal city, after the dramatists have been turned away at the gate? The Guards’ musical education will include dance, which usually implies singing too. They will eat, as if they were permanently on campaign, in common messes (ξυσσίτια); this Spartan practice implies sympotic drinking after the meal and much singing of lyric poetry. Despite a stringent ban on innovation in musical technique, new songs are allowed – provided they are in the same old style. Delphi will be invited to prescribe rules for religious ceremonies (founding temples, sacrifices, burials etc), all of which would in the Greek world involve singing hymns and other poetry. Hymns are an important element also in the ideal city’s annual breeding festivals. ‘Our poets’ will compose verse and music appropriate to the forthcoming unions. Again, at sacrifices and ‘all other such occasions’ there will be hymns (i.e. songs of praise) to honour men and women who have distinguished themselves in battle. Like Heroes of the Soviet Union, the good will be constantly extolled in public – to reward them and hold up models for everyone else.

This list is enough to show that poetry, of the approved sort, will be a pervasive presence in the life of the warrior class. Republic X sums it up as ‘nothing but hymns to the gods and encomia for the good’, yet the occasions for these will be plentiful enough to keep the poets of the ideal city busy. But I have had to compile the list from scattered remarks. No detail is given about how the various ceremonies will proceed. Worse, phrases like ‘hymns to the gods’ may suggest the wrong sort of detail to a modern reader. The Greek ύμνος covers a variety of forms more interesting than the hymns we are used to. The Homeric Hymn to Hermes, for example, is an engaging narrative, nearly six hundred lines long, with lots of mimesis, about the birth and impudent tricks of the robber god. Equally, any Greek reader would expect ‘encomia for the good’ to include tales of their noble deeds. Adventure stories will often be the order of the day.

One occasion for poetry does receive fuller treatment – the symposium. Book III’s discussion of poetry reaches its climax with a set of norms for symposia. This has not been noticed, partly because Plato expects readers to recognise the familiar setting without being told. Another reason is that in the past scholars have preferred not to wonder why the discussion of poetry ends by imposing austere limits to homoerotic sex.

Drama is not all the Guards are deprived of. Their epic recitals will be very unlike those the ancients were used to. No rhapsodic display, and much less speechifying than in the Iliad and Odyssey. The story will be mostly plain narrative, interrupted by the occasional stretch of mimesis. The mimesis will be largely restricted to auditory and visual likenesses of a good person behaving steadfastly and sensibly. The impressiveness of this steadfast, sensible behaviour will be reinforced by the speaker’s even delivery (λέξις). There will be little variation in his voice, and the accompanying music will stick to a single mode and a single rhythm. Even good people are struck down by disease, fall in love or get drunk, but mimesis of such events is to be very sparing. The other side of the coin is that a villain may do the odd good deed: mimesis of that is admissible, but it is not likely to happen often. The final exception is that poets may imitate bad characters in jest, to scoff at them.

Thus far, Chryses’ prayer would survive, but not Agamemnon’s angry, unrelenting response at line 26. Already it seems that the Iliad will have to stop as soon as it has started, but Plato delays until Book X the shocking news that Homer will be banished as well as the dramatists.

But remember that Book II implies that a purged tragedy will still be allowed. Tragedy and comedy are not explicitly banned until Book III. Plato deals out the pain in measured doses, allowing his readers to get used to one shock as preparation for the next. No objections have been raised to mimesis or to poetry in themselves. There will in fact be lots of poetry in the ideal city, some of it mimetic. The shock is, how little is to be mimetic; and how thoroughly edifying it all has to be.

The third stage of the discussion confirms that Plato has no objection to mimesis as such. Here Plato deals with the non-vocal side of music: the modes, instruments and rhythms which make the music in our narrower sense of the word. Socrates’ norms in this department are as austere as the norms governing content and performance. Some Bach might scrape by; certainly not Beethoven, Mahler or Stravinsky. This is where Plato gives examples of the kinds of mimesis to be permitted. The examples remove all doubt about the answer to the question: ‘What does Plato think is so bad about mimesis?’ Nothing – provided it is mimesis of a good and temperate character, the character (we later discover) of which gracefulness in architecture and bodily movement is also a likeness (μίμημα). On the contrary, mimesis has a formative educational role to play in the culture. What you imitate regularly is what you become, so from childhood the Guards must imitate appropriate models of courage, temperance and other virtues. These things must become second nature to them. Just as graceful architecture and bodily movement have a gradual, unnoticed influence on the souls of those who grow up in their presence, so, too, do the mimetic likenesses of the poetry Plato allows for the Guards. The passage I shall quote is designed to illustrate the permitted modes of music, but appropriate words are taken for granted. In the songs permitted at social and sacred gatherings, both music and verse will imitate the way persons of good character deal with the ups and downs of fortune; later we will meet the contrasting case of bad mimesis, the way a tragic hero reacts to misfortune.

The musical modes (άρμovίαι) under discussion are the ancient alternative to our musical scales. A mode is an attunement, a way of tuning the instrument to certain intervals, which lends a particular character to the tunes that can be played with it. When Socrates bans all but two modes, the Dorian and Phrygian, it is like saying: ‘Scrap all the minor keys, but leave just two of the major keys.’ Here are Socrates’ examples of good mimesis:

Leave me that mode which would fittingly imitate the tones and cadences of a brave man engaged unsuccessfully in warfare or any other enforced endeavor, who meets wounds, death or some other disaster but confronts it steadfastly with endurance, warding off the blows of fortune. And leave me another mode for the same man engaged in unforced, voluntary activities of peace: he may be persuading someone of something or entreating them, either praying to a god or teaching and admonishing a human being. Or, contrariwise, he may himself be attending to another’s entreaty, teaching, or attempting to change his opinion. In either case he does what he is minded to do without arrogance, acting throughout and accepting the outcome with temperance and moderation. Just these two modes, the one enforced, the other voluntary, which will best imitate the tones of brave men in bad fortune and of temperate men in good – leave me these.

If it was always these two types of song that we heard when we turned on the radio or went out to a social gathering, our culture would be very different. But not necessarily boring. Nothing stops a poet weaving the permitted types of mimetic display into a gripping third-person narrative, short or long; nothing stops a story including the imitation of more than one good character. A narrative of comradeship and dignified courage before death in a concentration camp could well satisfy Socrates’ norms for what he calls ‘enforced endeavour’. We might even be sympathetic to the idea that it would be indecent to give the Nazis any significant speaking parts.

The second type of permitted mimesis is for ‘voluntary’ activities. In Oliver Sacks’s Awakenings a doctor persuades the hospital authorities to let him try a new treatment on patients sunk in a permanent catatonic trance. They are unable to react to people or the world around. This treatment brings the patients to life again, but only for a while. The doctor accepts the outcome with temperance and moderation. He did what he could; medical science made a modest advance. It is an engaging, sympathetic story. But if you want more action, Plato has nothing against adventure stories. Heroism in military and civil life is exactly what this education aims to promote.

So do not think of the artistic culture of Plato’s city as boring. Austere, yes; an eventoned, calm expressiveness prevails. Plato’s word for it is ‘simplicity’ (άπλóτης). Growing up in such a culture would be like growing up in the presence of sober people all of brave and temperate character.

But the ideal city already ensures, so far as is humanly possible, that the young grow up in the presence of sober people of good and temperate character. Why worry about likenesses, the cultural icons, if kids are already surrounded by the real thing in flesh and blood? Plato’s answer is that, even in the ideal city, where the family and private property have been abolished, the people you know are only one part of the culture. When the influence of human role models is at odds with the cultural icons, there is a risk of change. It is not just that multiplicity and variety are bad in themselves. That is indeed at the heart of Plato’s objection to Homeric epic and Athenian drama, which revel in variety and the clash of different characters. But the main point is that change from the ideal is change for the worse. To avoid change as long as possible, the entire culture must be in harmony both with the people you meet in life and with those you know from poetry. That is why the discussion of musical poetry turns next to gracefulness in architecture, clothing, and everything that craftsmen make. A graceful material environment will ensure that the young are always and everywhere in the presence of likenesses of the same good and temperate character as the people whose lives and stories they know. The entire culture unites in harmonious expression of the best that human beings can be.

A musical education which forms a sensibility able to recognize gracefulness, and respond to it as an image of good and temperate character, also lets you recognise, and respond to, other images of good character – images of courage, liberality, high-mindedness. A Guard so educated, and old enough to understand some of the reasons why these are images of goodness, is ready to fall in love. Thanks to his education, the younger male comrade he favors will be one with beauty of character to match the beauty of his physical appearance. Love (ἐρως) of such a person is the goal and consummation of musical education. Socrates’ last word on poetry in Book III is a summons to erotic desire: ‘Music should end in the love of the beautiful.’

Socrates has now moved from the material environment to the social setting for musical poetry. The symposium is not the only social gathering where musical poetry is performed, but it is the one most relevant to love. Among the musical modes banned earlier, at stage three of the reform, were certain soft ‘sympotic’ modes, which encourage drunkenness; in the ideal city, as in Sparta, drunkenness is forbidden. But the rule presupposes they will drink wine. No Greek ever equated sobriety with abstinence. After the meal in their Spartan-style common messes, the Guards will drink in convivial moderation. (We have actual figures for Spartan wine consumption: Sparta was famous for its sobriety, yet their daily ration was well over our driving limit.) And the symposium is the main social occasion for dalliance: the couch is wide enough for two. In the ideal city, a lover may ‘kiss and be with’ his beloved, and ‘touch him as if he were a son, for honorable ends, if he persuade him’ – but nothing further, on pain of being stigmatised as ‘unmusical and unable to enjoy beauty properly’. The combination of wine, music and homoerotic love at the symposium was widely used in the Greek world (not only in Sparta) to forge bonds of loyalty and comradeship among those who fight for the city. Plato is adapting this institution to the austerely controlled ethic of Kallipolis.

Later, when readers have recovered from the shock of being told in Book V that in this city women, too, are to be warriors and rulers, equally with men, they learn that those who distinguish themselves on campaign (which would include symposia in camp, on beds of leaves) will exchange kisses with everyone else. Indeed, they will have an unrefusable right to kiss anyone they desire, male or female, and will be given more frequent opportunities to take part in the breeding festivals. The better you are, the more you can breed. Heterosexual desire, like homosexual, is harnessed to the ends of the city.

Looking back over the long discussion of musical poetry in Republic II-III, we should be struck by how widely it ranged. Starting with religion, ending with sex, taking in architecture and embroidery by the way, Socrates has broached all the issues that affect the ethos of society. All were woven around the central thread of musical poetry, because this for Plato is the main vehicle of cultural transmission, the main determinant of the good or bad character of the city.

In recent years, we have seen the ethos of British society go through a quite dramatic change as a result of the Thatcher years. The change was not planned in every detail from above. But there was a deliberate, concerted effort by the Conservative Government to purge the prevailing values and substitute the values of ‘enterprise’ and the spirit of the free market. In the political arena, whether national or local (including universities), it became increasingly difficult to appeal to the idea that the better-off should contribute to the welfare of the disadvantaged, for the overall good of the community. This attack on the values of community was pursued in every area of life, even in areas (like universities) where ‘the market’ is at best a metaphor. Metaphors and images, as Plato knew better than anyone, are potent weapons, especially in the wrong hands. If there are lessons for today in Plato’s discussion of musical poetry in Books II-III, the unit of comparison I would propose is not the details of censorship in the carefully guarded, closed world of the ideal city, but Plato’s concern for what he calls the ethos of society. Plato, like Mrs Thatcher, saw this as a prime political responsibility. Democrats can only undo the damage done to our society by the excesses of market ideology if we find democratic alternatives for fostering a better ethos in society at large.

Most of us do not share Plato’s confidence that objectively correct answers to these questions exist, and that, given the right education, men and women of talent can come to know what the answers are. Even if we did have that confidence, we would not think it right to impose our answers on everybody else. Democracy, both ancient and modern, puts a high value on individual choice and autonomy. That complicates the task. But it hardly relieves us of responsibility for thinking about what we can do to improve the world in which our children grow up.

It is not until Republic X that Socrates braces himself to denounce Homer openly as ‘the first teacher and instigator of all these beauties of tragedy’. Even the revered Homer, whom Socrates has loved since boyhood, must fall to Book III’s ban on the mimetic genres of musical poetry: tragedy and comedy. Homer is expelled because he is the master of tragic mimesis. But the main task of Book X is to explain why, in existing cities like Athens, it is dangerous, even for the most morally secure individual, to go to the theatre, or to Ion’s performance of Homer at the Panathenaea. The passive mimesis you undergo when you join that audience is a threat to the constitution of your soul.

The problem with uncontrolled mimesis, as Plato sees it, is not just the character of the likenesses it brings into our presence. It is how those likenesses gradually insinuate themselves into the soul through eyes and ears, without our being aware of it. Unlike narrative stories, which tell us about something, the seeming presence to our senses of the imitated characters can bypass the rational mind’s normal processes of judgment. To account for this phenomenon, we should return to the painted couch.

When we look at a painting, or (to take a second example) when we look at an oar half submerged in water, we know perfectly well that the painting is flat with no depth to it, that the oar is straight. But knowing this does not stop the oar looking bent or the painting seeming to have depth. How is the persistence of the false appearance to be explained? Only, so Socrates argues, by supposing there is some part of us, some level of the soul, which believes, or is tempted by the thought, that the oar actually is bent, that the painting does have depth. We are not inclined to believe it, but something in us is – just as something in the most sceptical person may shiver at a ghost story. At some level, we entertain beliefs, thoughts and fantasies that run counter to our better judgment.

Similarly, when we sit in the theatre and witness Oedipus discovering who he is, we know we are not hearing Oedipus’ own voice. Not because Oedipus is a fiction (for the ancient audience Oedipus is no more a fiction than Agamemnon or other heroes), but because Oedipus is not really there, only a likeness of him, just as there is no couch there in the picture, only the likeness of one. But knowing this does not stop us being affected by the appearances before us. Oedipus still seems to be on the verge of his terrible discovery. Even though we know they are only images, the false appearances persist, and stir our feelings. It is as if eyes and ears offer painter and poet entry to a relatively independent cognitive apparatus, associated with the senses, through which mimetic images can bypass our knowledge and infiltrate the soul.

In modern discussions of the influence of the media, it is often said that a normal, healthy individual is not unduly influenced by images they know are unreal. For Plato, the audience’s knowledge is the source of his deepest anxiety about mimesis. Normal, healthy individuals are undoubtedly influenced, all the time and in ways they are mostly unaware of, by images that pervade the culture. So knowing the image is only an image is no protection. Schools used to give lessons to make the young more aware of the wily tricks of the advertising industry. The advertisers had no need to protest. They knew that Plato has the better of the argument. A sexy jeans ad invites the viewer to notice its brazen appeal – and then go shopping.

Similarly in the theatre:

Even the best of us, you know, when we listen to Homer or some other tragedian imitating one of the heroes in a state of grief, delivering a long speech of lamentation, or chanting and beating his breast with the chorus, we enjoy it and give ourselves up to it. We follow it all with genuine sympathy for the hero. Then we praise as an excellent poet the one who most strongly affects us this way ... And yet when the sorrow is our own, you notice that we plume ourselves on the opposite response, if we manage to stay calm and endure. The idea is that this is the conduct of a man, whereas the sort of behaviour we praised in the theatre is womanish.

In the theatre we take pleasure in emotions we would try to restrain in real life: grief, joy, pity, fear, erotic excitement, anger, scorn. (The point does not depend on agreeing with Plato’s ideas about restraint: anyone will accept that there are times when emotion should be restrained.) Worse, we deliberately allow ourselves to indulge these feelings. As Socrates puts it, in the theatre our better judgment relaxes the guard it would maintain in real life. There are two rather different ways in which our guard is relaxed.

One is what we now call suspension of disbelief. We do not keep reminding ourselves of what we know perfectly well, that the events on stage are not really happening there now. They may have happened in the past. (For the Greek audience, a tragedy’s plot is not fiction; it is more like Shakespeare’s history plays or medieval mystery plays.) But the events are not actually unfolding before our eyes and ears. We would be upset if we turned on the television one evening, watched what we took to be the end of a rather violent film, and then the announcer came on to say: ‘That’s the end of the News.’ The jolt would prove how completely we had suspended normal judgment about what was apparently taking place. Conversely, I recall a news commentator during the Los Angeles riots exclaiming in disbelief: ‘This is not a film; this is for real.’

But Plato worries more about our suspending moral judgment about what is apparently taking place. When we sympathie with a grieving hero, we not only allow ourselves to share feelings we might wish to restrain in real life. We also allow ourselves, as part of that emotional bonding, to share a while, at some level of our soul, the hero’s belief that a great misfortune has happened. And here the mistake is not that no such event took place, it is only a play. The mistake in Plato’s eyes is allowing yourself to believe, even vicariously and for a short while, that an event like the death of your child would be a terrible loss, a great misfortune, if it really happened. The law in the ideal city is stern:

The law declares, does it not, that it is best to keep as calm as possible in calamity and not get upset, (1) because we cannot tell what is really good and bad in such things, (2) because it will do us no good in the future to take them hard, (3) because nothing in human affairs is worthy of deep concern, (4) because grief will block us from taking the necessary measures to cope with the situation.

The whole culture is set up to reinforce this law – remember the songs about calm endurance in adversity. The mimetic genres of poetry – epic, tragedy and comedy – encourage people to suspend the moral principles they try to live by, so as to enter the viewpoint of emotions which their better judgment, if it were active, would not approve. This is how the analogy with visual perspective carries over to the theatre. When we share an emotion with a character on stage, we enter (despite our better judgment) the moral outlook from which the emotion springs. The images created by theatrical mimesis are so sensuously present to eyes and ears that they lock the audience into a distorted moral perspective. Epic and drama encourage us to feel, and to some extent believe, against our better judgment, that the ups and downs of fortune are much, much more significant than they really are.

This is not Mrs Whitehouse’s argument that showing a violent film on Tuesday brings about a rape on Wednesday. It is a more interesting claim about the longer term influence of mimesis. By encouraging us to enter into the perspective of strong emotions, epic and drama will gradually erode the ideals we grew up with, even if they go on being what our better judgment tries to live up to. This argument does not depend on the stern, other-worldly morality on which Plato’s ideal city is founded. Let the prevailing morality be more relaxed and humanistic: it will still include ideals we think we should live up to, and Plato will still caution us about mimesis. It is dangerous to enter feelingly and uncritically into viewpoints that our better judgment, if it were active, would not approve. That is why he would banish Homer, tragedy, comedy and their modern equivalents.

Some writers have naively supposed they could defend Homer and imaginative literature generally against Plato’s critique by claiming that literature enlarges the sensibility and makes us more feeling people, because it fosters empathetic understanding of all sorts of different characters, both good and bad. As if Plato did not know that. ‘Yes,’ he would reply, ‘that is what we need to prevent.’ Opposite conclusions are drawn from the same premise. What you cannot do, it seems to me, is accept that mimesis has the effects on which Plato and these critics are agreed, and then argue that anything and everything should be allowed. If we agree with Plato about the power of mimesis (ancient or modern, epic and drama, or advertising, film and TV), but reject his authoritarian solution, then democratic politics has to take responsibility for the general ethos of society. Plato’s problem is still with us. It needs a modern solution.