Poetics

Overview

The Poetics is about sixty pages and reads like what it probably is: notes for a course, compressed, occasionally repeating itself, breaking off where a second book on comedy should begin. Do not expect an essay. Expect a working analysis, in the manner of Aristotle’s biology, of an object he found interesting enough to take apart.

The object is poetry, by which he means made things in verse: tragedy, epic, comedy. His starting claim is that all of them are species of one activity, mimesis, usually translated imitation and sometimes representation. They differ in the medium they imitate with, the objects they imitate, and the manner. That sounds dry and it is doing real work: it lets him compare tragedy and epic as two solutions to the same problem, and it lets him ask what tragedy is for rather than whether it is nice.

Most of the book is about tragedy. He gives a definition, breaks tragedy into six parts, ranks them, and then spends the middle chapters on the part he ranks first. The last third turns to language and to the comparison with epic, where he argues that tragedy is the better form because it does the same work in less space.

The argument

Chapters 1 to 5 set up mimesis and sketch how poetry developed, with a historical account that is partly guesswork and worth reading lightly.

Chapter 6 is the hinge. Tragedy is the imitation of an action that is serious, complete, and of a certain magnitude, in language made pleasurable, performed rather than narrated, and through pity and fear accomplishing the catharsis of those emotions. Then the six parts, in his order of importance: plot, character, thought, diction, song, spectacle. Spectacle comes last, which is a deliberate insult to the stage machinery of his day.

Plot first, and this is the claim students argue with hardest. Tragedy imitates an action, not a person. Character exists to make the action intelligible. You could have a tragedy with no character work at all, he says, but not one with no plot. Chapters 7 and 8 define what makes a plot whole: a beginning, a middle and an end, of a magnitude the memory can hold, with the events so bound together that moving or removing one damages the rest.

Chapter 9 contains the best-known sentence in the book. Poetry is more philosophical than history, because history tells you what Alcibiades did and poetry tells you what a man of a certain sort would probably or necessarily do. The poet’s job is not accuracy but probability. This is his answer to Plato: poetry is not a lie about particulars, it is a truth about kinds.

Chapters 10 to 14 give the machinery. The best plots are complex, meaning they turn on peripeteia, a reversal of the situation into its opposite, and anagnorisis, a change from ignorance to knowledge. Oedipus is the model because both happen in the same stroke: the messenger comes to reassure him and destroys him with the same news. Then hamartia. The tragic figure must be better than average but not perfect, and must fall through an error rather than through vice, because we pity the undeserving and fear for the like ourselves.

Chapters 16 to 18 handle recognition scenes and plot construction, chapter 19 onward diction, metaphor, and the parts of speech. Chapters 23 to 26 compare epic and tragedy and answer objections to poetry, ending with the claim that tragedy is superior because it achieves its effect in a shorter compass.

How to read it

Read Heath’s introduction first. The technical terms are the whole difficulty, and forty pages of orientation will save you an hour of confusion.

Read Sophocles’ Oedipus the King before or alongside, if you have not. Aristotle refers to it constantly and the argument is much harder in the abstract. Half the book is a close reading of a play he expects you to know.

On a first pass, read chapters 6 to 14 closely and let the rest go quickly. That is the section every later critic argues with. The material on diction in chapters 19 to 22 is technical Greek grammar and can be skimmed unless you read Greek, though chapter 22 on metaphor is worth the time.

Two things to watch for. First, catharsis appears once in the definition and is never explained, which means every account you have heard of it, purgation, purification, clarification, is a reconstruction rather than a summary. Notice how little Aristotle actually commits to. Second, hamartia is not a moral flaw. The common translation “tragic flaw” imports a character defect the Greek does not require; the word covers a mistake, a misjudgement, missing the mark. Oedipus’ error is that he does not know who his parents are.

The common misreading is that Aristotle is prescribing rules for writing plays. He is describing what the best existing tragedies have in common. The “three unities” that French neoclassical critics enforced for two centuries are not in this book: Aristotle mentions unity of action, remarks in passing on the length of a day, and says nothing about place at all.

Questions it raises

• If plot matters more than character, why do most readers remember Oedipus and Antigone as people rather than as sequences of events?
• Catharsis is left undefined. Does tragedy purge pity and fear, train them, or merely give them somewhere to go?
• Aristotle says poetry is more philosophical than history because it deals in the universal. Is that true of the tragedies you have actually read, or do they work precisely because their particulars are so specific?
• If the tragic figure must be neither wicked nor faultless, what happens to a play whose protagonist is genuinely monstrous, or genuinely innocent?
• Aristotle answers Plato’s charge that poetry corrupts by arguing it does something useful to the emotions. Has he met the charge, or changed the subject?
• He ranks spectacle last and says a tragedy should work when read as well as performed. Is that defensible for a form written to be staged?