Tragedy, comedy, and agon

An essai on chapter 4 of Aristotle’s Poetics.


It is clear that the general origin of poetry was due to two causes, each of them part of human nature. Imitation is natural to man from childhood, one of his advantages over the lower animals being this, that he is the most imitative creature in the world, and learns at first by imitation. And it is also natural for all to delight in imitation.

—Aristotle, Poetics

Along with imitation, Aristotle lists harmony and rhythm as natural human faculties that cause us to enjoy poetry and drama; and we should add (though the philosopher omits it) the faculty of language itself. None of us were born speaking Greek, and a few of us here and there have not got the hang of it even yet. But everybody speaks some language, unless prevented by some bodily defect. And every language I know of has some form of poetry.

In the fourth chapter of the Poetics, however, language is a side issue. Aristotle here is concerned with two things: first, to show how and why poetry develops as an imitative art, and second, to briefly trace the history of Greek poetry down to the full development of tragedy.

He begins with some general remarks, which are perfectly sound as far as they go. We do indeed delight in seeing things imitated, whether in pictures, sounds, or language, or by mimicking actions. Some of us can even bear to watch mimes, who are no good for anything else. I am not quite being flippant. Whether we enjoy it or not, it is true that mime has only one element to offer: it represents different kinds of actions solely by the movement of the performer’s body, without speech, props, or any other dramatic aid. This is a very low form of art, in the sense that moss is a low form of plant life: it does not have enough variety of parts to support any complex structure. Mosses do not grow into trees, and mimes do not perform Shakespearean plays. Those things require more specialized and organized forms of life and art.

All forms of art, perhaps, have their origin in imitation; but some stray further from their origins than others. Painting and sculpture are usually most successful when they remain in the realm of imitation; that is, when they present us with concrete and recognizable subjects. Representational art has a ready-made emotional value that abstract art can only clumsily approach. We have built-in responses to a picture of a mother with her little children, but not to a triangle with her little vertices. If the mother is weeping over her dead child, our reaction may be immediate and overwhelming; nothing about a triangle could convey anything like it. Those abstract artists who dispense even with geometry, and try to accomplish everything with shapeless masses of colour, place themselves under an even bigger handicap. They are not to be despised for it; but some of them, perhaps, should have been warned against it.

On the other hand, music works better on the emotions, and is capable of greater elaboration and more ingenious composition, when it it is least representational. You can, as a mere tour de force, make a musical piece out of (let us say) bits of recorded birdsong and the sound of waves on a beach, artfully arranged; but abstract melody and rhythm have a direct appeal all their own. Most musical instruments sound, in one way or another, like noises made by the human voice; but they are not expressive in the way that a voice is, because any given instrument has only a limited range of tone and timbre. The mere sound of a violin being tuned, or a saxophone being blown by a child who cannot play it, conveys no information to the human ear. But add the abstract inventions of melody and rhythm, and a violin or a saxophone can move us to laughter, joy, or tears, for reasons that no words can easily explain.

Dance is an art that falls somewhere between painting and music in this respect. What the mime at the seaside does is purely representative dance, and he can do very little with it. He may make us laugh and applaud for a moment, that is all. We are entertained not because he is doing anything very interesting, but because he manages to do anything at all with such a limited form. This is also why a man on a bicycle is a bore, but a man on a unicycle is a circus act; and why someone carrying a bag of oranges is unremarkable, but someone juggling the same oranges can keep us amused for some time.

One defect of the mime is that he operates in strict silence. Dance moves us more easily in conjunction with music; indeed, some dance music can move us almost as well without the dancers’ help. In modern times, the most thorough union of dance with music has been the ballet; and while ballet dancers sometimes act out particular concrete actions as a mime does, most of the time they are making abstract movements that express the other abstraction of the music. The result can be as moving as a child’s tears and yet as arbitrary as a kaleidoscope.

When you add language to any combination of these arts — or rather, when you add poesis — you will find that you have crossed over from art into magic. The mime can show himself walking against the wind; the actor can explain why it is so important for him to go out in that weather. The gypsy violinist can play a heartbreaking tune; the singer can say why her heart is broken. A painting can show Hector squaring off against Achilles; a play can show their duel in living motion, right down to the tragic finish, and tell you why there was nothing each man could do but fight. Instead of showing an exercise in technique, suddenly you are showing a world.

The drama, as Aristotle rightly observes, is the fusion of all these arts: the actor, the dancer, the musician, and the stage-painter all have their places in it, under the supreme direction of the poet. But the fusion was achieved only slowly and gradually, and of course each art has its own independent existence still. Music and dance have always been linked, but it was the combination of music and poetry that began the process that led to the development of drama. The ἀοιδοί, like their cousins in other cultures, bards and skálds and so forth, combined instrumental music with poetic recitation. At particular moments, and to express particular moods, they would break into song, but ordinarily the strummed harp or lyre and the spoken word would each go their own way, playing to the same rhythm, evoking the same mood, but not united in the full and obvious way. This is what the show was like in Homer’s time; it was still being done a hundred generations later, when Arlo Guthrie played his guitar and told us about ‘Alice’s Restaurant Massacree’.

From this beginning, drama developed by annexing other art forms consecutively, so as to give greater immediacy and verisimilitude to the performance. But along the way, as Aristotle reminds us, there was a division of the ways. The single face of the ἀοιδος was bifurcated into the laughing mask of the comedian and the crying mask of the tragedian. And this, I am afraid, is one point that Aristotle has exactly wrong.

Here is his version of the story:

Poetry, however, soon broke up into two kinds according to the differences of character in the individual poets; for the graver among them would represent noble actions, and those of noble personages; and the meaner sort the actions of the ignoble. The latter class produced invectives at first, just as others did hymns and panegyrics.

But in the same passage, he admits that the earliest known ‘mean’ and ‘ignoble’ poem in Greek literature was the Margites, which was attributed to Homer himself. Aristotle did not for a moment believe (whatever modern critics may say) that there was one Homer who wrote the Iliad and the Odyssey, and another who wrote light comic verse. There is another tradition, more commonly accepted, which says that the Margites was written by Pigres of Halicarnassus; but that same Pigres performed a feat of astounding virtuosity in ‘serious’ verse, by interpolating a pentameter line of his own composing after every hexameter in the Iliad — in effect, writing a parallel epic (though in a different metre) as a commentary in between the lines of the original.

Again, Aeschylus is remembered chiefly for his tragedies, but (following the rules of the ancient Greek festivals) he presented his plays in groups of four: three tragedies followed by a satyr play, which was basically a farcical treatment of a serious subject. In relatively modern times, Shakespeare wrote both tragedies and comedies, in almost equal numbers and with equal skill; and I have never heard anyone say (for instance) that Shakespeare’s tragedies were own, but his comedies were written by Sir Francis Bacon.

It is clear, then, that it is not any difference between poets that separates comedy from tragedy; it is a difference of mood and technique, but both moods and both techniques can be done by the same writer at different times. Indeed Shakespeare, and the moderns generally, accomplished something that the Greeks would have considered impossible: they inserted comic relief into the action of a tragic play, without in any way impairing the seriousness or the emotional power of the tragedy. In fact, comic relief can often heighten the effect of tragedy, just as the dark parts of a chiaroscuro painting make the light parts seem more brilliant.

In fact, there are only two kinds of writers who write comedy or tragedy exclusively. One is the comic genius, like P. G. Wodehouse, who is so famous for his comedies that he dare not disappoint his audience by venturing a work that does not make them laugh. The other is the writer of limited gifts, who does only ‘serious’ work because he is not skilful enough for comedy. The latter, it is true, may console himself by sneering at comedy as something not worthy of adult attention. But he is really in the position of a colour-blind man sneering at the rainbow. He mocks what he cannot appreciate or make for himself, because he hates to see anyone else get a pleasure that is denied him. Aristotle was a man of many gifts, but I have never heard that he made a successful joke. We must not take him at his word that comedy is an inferior form of art. He was not fitted by experience to have an informed opinion.

For in fact comedy is the hardest art of all. Anyone who can write at all can write tragedy; tragic situations will suggest themselves to the meanest imagination. We can all conceive of suffering, even if we are the odd sort of people who never suffer much themselves. But there is no such thing, properly speaking, as a comic situation. A skilled humorist can find comedy in any situation; someone without that skill could not find it in a barrel of circus-trained monkeys. George Lucas was a good enough director for action-adventure and melodrama, and he could even include bits of comic relief (with a lot of help from his actors); but when he tried to create a purely comical character, he produced the ghastliest failure of his career — Jar Jar Binks.

In fact, there is something heroic about the best kind of comedy, something that matches the traditional happy ending with particular felicity. In both comedy and tragedy, the Greeks anticipated something of the ethics of Christianity. Tragedy takes a great and noble hero and casts him down in a flood of tears; he is weighed in the balance and found wanting. Comedy takes a small and humble person and raises him up on a current of laughter; he is tested in his heart and found deserving. You can write comedy about a king, but he must be an unhappy king. You can write tragedy about a beggar, but he must be a self-satisfied beggar. All this is exactly in the mood of the Magnificat, the oldest and best of all Christian hymns. The Fate that governs the characters of drama happens, in this respect, to exactly resemble the Christian God: Deposuit potentes de sede: et exaltavit humiles. God and Drama both cast down the mighty from their thrones, and exalt the humble.

Comedy has always been more popular with the general public, and tragedy with the elites, partly because of the usual nature of their leading characters, but chiefly because human nature craves the refreshment of change. The hero of comedy is democratic: he is an ordinary man; even if he is a king, he is only comic because beneath his robes of state, he is merely human. The hero of tragedy is aristocratic: he is a man of destiny; he alone rises up high enough to challenge the gods, and attract their wrath to throw him down again. But also, it is only very happy and contented people who can bear to contemplate suffering as a form of entertainment. Ordinary people have enough suffering of their own to do, and want to think about something else in their leisure time. This is why it is so hard to write a bestseller; and why bestsellers almost never please the critics. The critics are aristocrats, and despise the public for its ‘escapism’; but then, the public has things it needs to escape from.

G. K. Chesterton observed that in Victorian London (his own native habitat), the very best humour was the spontaneous invective of navvies and costermongers — the lowest of the working class, the people who were lucky to get food and shelter on the same day. It was as legendary as the verbal paucity and parsimony of the aristocracy; and as true. The London ‘mob’ could destroy an earl with funny insults and shouted lampoons, and suffer nothing for it, for their own kind would laugh with them; the earl (being too well-bred to offer insults and too dull to make jokes) could retaliate only by force, which would land even an earl in prison.

A bit later, in the 1920s, a tide of unprecedented affluence made even middle-class people blasé enough for tragedy; the result was a flood of unhappy stories and deliberately ugly pictures, what Orwell called ‘cheap return tickets to the end of the night’. In the 1930s, the Great Depression threw many of the same people into a nightmare of privation and despair; therefore that was the great age of screwball comedy, and above all of the animated cartoon. Prosperity dined on Proust; adversity lived on Donald Duck.

In such times and circumstances, there is something heroic in the capacity for humour. At the very worst moments of life, the most heroic thing a man can do is simply to endure — to survive, to carry on, to keep being human, until hope and opportunity return. Then the ability to laugh at one’s own plight, or still better at the ones who have caused it, is as clear a sign of humanity as the power of standing on two legs. The joker and the biped both resist the gravity that is pulling them down; they both rise above the level of their fate by the power of levity.

So instead of a single Poesis annexing the other arts on her way to the empire of Drama, we have the conjoined twins, the laughing mask and the crying mask, moving together towards their respective ends in Comedy and Tragedy. It is a pity that Aristotle cannot trace the history of comedy for us; even in his own time, the relevant documents did not exist, because comedy (being the art beloved of the masses) was looked down on by the aristocrats who wrote the books. But he does describe the evolution of tragedy in some detail.

The earliest stage (which he takes somewhat for granted) combines the recitals of the poets, already accompanied by music, with the choral dances of Dionysian religious rites. Tradition has it that Thespis of Icaria, about two hundred years before Aristotle’s time, was the first poet to play a character instead of merely reciting his poems in propria persona. By this invention, he turned the dithyramb into the earliest form of drama — a series of soliloquies alternating with the songs of the chorus. Aeschylus added a second actor, so that the story could be developed on the stage by means of dialogue. Sophocles added a third actor and scenery; later playwrights enlarged the scope of the drama by writing plays with more than one act.

Along the way, the Greek dramatists stopped writing in a trochaic metre, which was well suited for the song and dance of the chorus, and began to write in iambic lines, which sounded more like natural speech. The same development occurred in modern drama, when the Elizabethan playwrights settled on iambic pentameter, and the French on alexandrines, keeping the rhythm of poetry without the sing-song quality. The next logical step was to write in natural speech — that is, in prose — but, oddly enough, that step was not definitely taken until Ibsen rejected verse in the nineteenth century. Today, no one would think of writing a play for the stage or screen in anything but prose. The last successful dramatist to use verse, perhaps, was Dr. Seuss. Aristotle, who never imagined such a thing as a play written in prose, would be astonished.

The climax of this trend in ancient times was the classic tragedy, with three actors and a chorus, songs, dances, masks, scenery, props, and some limited (but very ingenious) ‘special effects’. In modern times, the climax seemed to have been reached with grand opera; and since the attitudes of ‘high’ art and its critics were frozen about 1900, there are critics who believe it still. In fact, opera began to die as an art form when movies came into their own, and even its bastard child, the Broadway musical, has become merely a tail wagged by the dog of Hollywood. The most characteristic stage production of recent years was The Producers, a musical based on a movie; and the moment it proved a success, it was turned into a movie again. The reason is plain: a film, having been performed once, can be viewed many times and in many places. Sophocles’ audience had to travel to the theatre in Athens to see his latest play, which was performed one time only, at a set public festival. Spielberg’s audience can watch his latest film whenever they wish, wherever they happen to be. So it becomes feasible to lavish all kinds of technological gimcrackery on the production (and post-production, a new art form in itself), to create a spectacle that no stage could contain and no stage-producer could afford. Nothing in opera can compete with that, except the arias; and so opera, instead of being the glory and summit of the arts, has retreated once more to become merely a branch of what is called ‘classical music’.

It is the dream, and sometimes the ambition, of nearly every modern writer of poesis to have his work made into movies; but most of us will never see it done. And those who do, alas, will also see their work transmogrified almost beyond recognition; for in Hollywood, the Director is the Author, and the writer is merely ‘a schmuck with an Underwood’. So for the vast majority of works that are created, the arts have gone their separate ways again; we have to fall back on the unaided resources of language. In fact, we are moving in the opposite direction to what Aristotle described as happening in ancient Greece:

As soon, however, as Tragedy and Comedy appeared in the field, those naturally drawn to one line of poetry became writers of comedies instead of iambs, and those naturally drawn to the other, writers of tragedies instead of epics, because these new modes of art were grander and of more esteem than the old.

Movies, and even television series, are certainly grander than books, but they are not of more esteem; indeed, they are the very best proofs of the old adage that ‘too many cooks spoil the broth’. If you are seized with a strong imaginative vision, if you want to tell a story about it, impose your own form on the vision and your vision on the world, you have to do it alone. Hollywood (or even London) will not help you. They will pick your brains for their own purposes, but not for yours; and sometimes they will pick your pockets as well. So as a practical matter, whatever Aristotle may say about the theory, we have to consider the written word in its pure form. We have returned to the age of the epic — this time in prose.

If the prose epic, or as it is often miscalled, the novel, has a shortcoming, it is that it has too many advantages. True, it has no music, no visuals, no actors; but also no associate producers and hairdressers’ boyfriends contributing their own bad ideas to ruin the script. In its own realm, it is supreme and its power is unlimited. If you want to write a world-wrecking science fiction epic, you can wreck worlds; if you want your fairy-tale princess to ride in a palanquin at the head of ten thousand white elephants, you have only to write the words. But unlimited power is not Art, because the essence of every art is limitation. If I choose to write a sonnet, I have to write in rhyme, because a poem that does not rhyme is not a sonnet; and then I cannot rhyme moon with fork, or I break the rules and forfeit the game. Poesis in prose has no formal limits of that kind.

This absence of limits, this lack of a prescribed form, is an open invitation to formlessness. I have heard it said by teachers of ‘creative writing’ (that most dismal of vices) that pretty nearly all their students have a good ear for prose, and some have natural gifts for dialogue, description, and characterization; but not one in a hundred has a nose for story. They do not know how to string a story together; they have no idea of its anatomy. Anybody can write a sonnet, because the form of a sonnet gives the recipe for writing it. It may be a bad sonnet, it may even be meaningless; but if it contains fourteen lines in iambic pentameter with a certain rhyme scheme, it is a sonnet just the same. But not anybody can write a story, because there is no recipe for a story. The lack of a set form encourages people to write formlessly; but a story is not a formless thing. I shall return to this point later; but for now I want to touch on one matter in particular.

One of the troubles that can sink a story is that the writer does not clearly know who the main character is. Sometimes, indeed, you have to tell the story (in rough draft) to find that out; and then, if you are wise, you will go back and revise with the knowledge in mind. Since a story can have as many characters as you want to put in, the problem can quickly become insoluble. Fortunately, it is not always necessary to solve it; and the larger the scope of the story, the better it can stand without a solution. Who, for instance, is the main character of War and Peace? Yet even that masterpiece has come in for criticism because of its lack of narrative focus. It has unity of theme and subject-matter, but no unity of action. There are so many important characters that none of them ever acquires preeminence over the others or over the whole plot.

This problem can hardly arise in a Greek drama, with its strict limit on the number of actors. It is true that the limit is three actors, not three characters; the actors can each play several roles, changing masks and costumes for each part. But only three characters can be on the stage at one time, and the maximum length of the performance — really, the number of hours the audience will sit still — naturally limits the number of changes each actor can do. In any case, it is understood that if there are three actors, they will spend most of their time performing the three leading roles. The Greeks had technical names for the three actors in a tragedy. The first actor, the one Thespis pulled out of the chorus to deliver the first spoken lines, was called the πρωταγωνιστής — the protagonist. The second and third actors were, accordingly, the deuteragonist and tritagonist. (Minor parts, like messengers and soldiers, were not counted among the actors. A member of the chorus could pop onstage to say something like, ‘Sire, a message! King Oedipus is dead!’)

Since actors in ancient Greece were as vain as actors today, and as tenacious in their battles over top billing, it was a matter of great importance to assign the leading roles in the correct order. A protagonist outranked a deuteragonist as surely as a colonel outranks a major. I suppose that the problem became really acute when older plays, with fewer actors, were staged. If a troupe performed a revival of Aeschylus, whose plays have only two actors, the poor tritagonist would be relegated to the chorus.

We who practise our poesis in book form do not have to worry about the egos of actors, but it can be a useful exercise to act as if we did. It may help us bring our stories into sharper focus. A question that we need to answer (so that our readers can answer it also) is, ‘Whose story is this really?’ If the answer is not obvious, I can suggest the following exercise:

Imagine that you are adapting your story as a Greek tragedy (whether the story is tragic or not). Make a list of your characters as if they were the dramatis personae of a play. Now take every character who plays only a bit part — the sort that could be done by a member of the chorus — and strike out his name. This will probably shorten your list. Then strike out the least important remaining character, the one whose lines could be cut out with the least damage to the story. Keep doing this until you have only three characters left; these, then, will be your Protagonist, Deuteragonist, and Tritagonist.

It is not necessary, by the way, for any of these characters to be the villain. Your villain may be the strong, silent type; your villain may be as impersonal as Fate, in which case he can be represented by the chorus. If you are making a Greek tragedy of The Three Little Pigs (a ridiculous idea), then your three actors have got to play the pigs: Practical, Fiddler, and Fifer, in that order. The chorus can be the wolf; choruses are good at huffing and puffing.

Now suppose that your play is being cut down to suit Aeschylus, and get rid of the third actor. Which character do you cut? Then repeat once more, to suit old Thespis, and get rid of the second actor. If your story had to be put on stage in such a way that only one actor had a speaking part, and everybody else had to be represented by the chorus (or by a narrator offstage), which character would speak? Which character’s actions do not explain themselves, so that he needs the power of speech to make sense of himself?

Sometimes this last step is impossible, because the story is essentially dependent on a double-act; but it is often easy to tell which of those last two characters takes top billing. The Sherlock Holmes stories clearly have Holmes as the protagonist, even though it would be difficult to cut the actors below the customary three. Holmes needs Watson as a foil, and the criminal has to have his say. Of course, there would not be much of a mystery if there were only one speaking part besides Holmes and Watson; but then, the purpose of this exercise is to simplify the story until it breaks. What is artistically necessary is often more than what is structurally necessary. A human body can be identified just by its bones, but to be alive, it needs flesh as well.

The Hobbit could be reduced to two actors, playing Bilbo and Gandalf, plus a Greek chorus of assorted Dwarves. Snow White also has a chorus of dwarfs; in this case the deuteragonist is the wicked Queen, without whose envy and vindictiveness there is no plot. Hamlet could be reduced to one actor without losing anything essential; he could argue out his soliloquies with the chorus, who would double by representing the other characters in silent pantomime. (Polonius’s advice could be represented by the chorus chanting, ‘Blah, blah, blah.’)

There is a kind of superstition, founded on folk etymology, that a protagonist is the opposite of an antagonist. Actually, the two words are not related in that way. Antagonist begins with the prefix anti-, but protagonist does not begin with the prefix pro-. (It really means proto-agonist — the first competitor.) The opposite of an antagonist is — an opposed antagonist. The two boxers in a ring are both antagonists. Sauron is ‘The Enemy’, the antagonist of all Middle-earth; but Gandalf says, ‘I was the Enemy of Sauron.’ It takes only one foe to make a war, but it takes two to have a battle; antagonism, in the dramatic sense, is always mutual.

The protagonist of a story is always someone who wants something. (The exercise above works for that very reason: the protagonist is the person best qualified to say what he wants, so if only one character can speak, he had better be that one.) If the protagonist does not want something, you have no motive; if he is not opposed in getting it, you have no ἀγών — no contest, no competition — therefore, no story. So by definition, the protagonist is always one of the antagonists in his own struggle; what we usually call the antagonist really means the other antagonist — the one who takes the opposite side. It is usual to divide all story conflicts into three classes: Man vs. Man, Man vs. Nature, Man vs. Himself. Clearly, only in the first case will the antagonist be a separate character; and often, the antagonism will be such as to require no explanation. Sauron never speaks in The Lord of the Rings; but Sauron only wants to rule the world. It takes no dialogue to explain such a simple desire. If any explanation were needed, we would soon find it by watching Frodo as the Ring slowly corrupts his mind with the same desire.

For this reason, the ‘antagonist’ of a story (that is, the one opposed to the protagonist) need not be one of the leading characters. And it follows from this that the frequent complaint of critics, that the villain of a story is ‘cardboard’ and his motivations lack nuance, is generally empty. If there is a villain, he may be acting from the simplest motives, or merely from his nature. St. George’s antagonist was the Dragon, but the Dragon was not antagonistic because it had any special grudge against St. George; it was antagonistic because that is what dragons do. In the whole Western tradition, they are the natural enemies of humanity in general, as implacable as a force of nature, and as unnecessary to explain.

Sometimes, however, the antagonist will also be the deuteragonist or tritagonist, or at any rate a major speaking character; then we had better put some effort into delineating his psychology and describing his motives. If the central story of the Iliad were a Greek tragedy (as it well might be), then Achilles would be the protagonist, Hector the deuteragonist and Achilles’ antagonist. And Homer, though always a Greek, did some of his best work in bringing Hector to life; so that our sympathies are almost evenly divided, and though we know that these two are bound to fight to the death, we are sorry to see either one of them die.

Modern critics like antagonists to be Hectors instead of Dragons. Partly this is because modern critics like to think they are above ‘naive’, ‘simplistic’ ideas like Good and Evil, and look down on anyone childish enough to take sides in a quarrel. But chiefly it is because they like psychological stories, and think all stories should be about characters who are analysed instead of characters who act. Aristotle said quite clearly that they are wrong; but for him (and us) that is a matter for another chapter.


  1. When my brother took classes on classic Greek works, he mentioned that it appeared that Homer actually made Hector a far more honorable and sympathetic character than Achilles–even though Achilles was the protagonist, you were supposed to sympathize with the tragic Hector.

    Great essai, as usual!

Speak Your Mind