On Translating Aristotle’s Poetics
The obvious question is – why again? Even a select list of English translations in this century makes quite a litany: Butcher, Bywater, Hamilton Fyfe, Lane Cooper, Allan Gilbert, Preston Epps, Seymour Pitcher, L. J. Potts, George Grube, Gerald Else. I admire three or four of these, and decry none of them. While the study of English literature has – in part at least – taken the place of Greek and Latin as a central humanist discipline and literary criticism has tried to assume the role almost of an autonomous discipline, Aristotle’s Poetics has continued to be a document of great historical and critical importance. Because almost nobody in the field of English studies reads Greek any more – if indeed anybody ever could read fluently and without dismay the Greek of the Poetics – translations have accumulated, all highly accomplished.1 But many of them are of a marmoreal smoothness; almost, the more eloquent and stylish the translation, the farther it is from inducing the direct tactile qualities of the Greek original. For many students of English literature, even some pretty mature ones, the Poetics is either a doctrinaire statement that can be readily mastered from a translation, or a very limited account of poetry, interesting enough as the oldest surviving treatise on poetry but distant, foreign, and not very much to the point. Certainly the continuous reprinting of Butcher’s translation in collections of critical texts has not encouraged the currency in English studies of certain important developments in Aristotelian scholarship in the past forty years.2
As I have worked repeatedly through the Poetics, trying to unfold the original to students of English who have even less Greek than Shakespeare had, I have gained an increasingly vivid sense of the activity of Aristotle’s mind in this broken and intermittent little document; and have wondered whether a translation could conceivably be prepared that would bring a reader to ‘the revelation ... of the driving energy of Aristotle’s thought’.3 ‘An editor in these days’, Ingram Bywater wrote sixty years ago, ‘can hardly hope to do much to advance the interpretation of a book which has been so carefully studied and re-studied by a long succession of editors and translators, many of them among the more illustrious names in the history of classical scholarship.’4 To think of doing anything about the interpretation of the Poetics would make the heart even of a classical scholar quail.5 Of interpretation there is great store, not least in the work of those Chicago scholars whose enemies have called them neo-Aristotelians – Crane, Olson, McKeon, Maclean, Weinberg, to name but a few. These know their Greek as well as their English literature; and there is no sign that as critical theory has effloresced classical scholars have failed to apprise themselves of what might conceivably be profitable in the criticism of English letters to enrich and refine the commentaries they write for classical scholars. And still I feel there is something that needs to be done that has not yet been done for students of English literature; and it would probably take more than a plain translation. My purpose is simply to recover for Aristotle’s Poetics what Werner Jaeger said was Plato’s aim in writing his dialogues: ‘to show the philosopher in the dramatic instant of seeking and finding, and to make the doubt and conflict visible’.6
Aristotle’s works, as we know from the three lists that have come down from antiquity, fall into three groups, only one of which survives. His early reputation as a writer rested on a number of dialogues in the Platonic manner, many if not all written before he founded the Lyceum; all are now lost, and what little we know about them is from a few fragments and a few comments by other writers. He also compiled very extensive memoranda and compendious collections of material put together (sometimes with the help of others, he being perhaps the first to make systematic use of research assistants) for purposes of study and as a basis for future scientific works. Beyond fragments only one of these survives – the Athenaiōn Politeia, notes on the constitutions of 158 states, mostly Greek, prepared for publication and stylishly written, a manuscript which was recovered almost intact from the Egyptian desert as recently as 1889. Thirdly, he wrote philosophical and scientific works, still extant, about thirty in number, to which are attached two doubtful works and some seventeen spurious works. None of the works in this group was prepared for publication, and as a group they show varying degrees of finish; the Nicomachean Ethics is one of the most finished, the Poetics one of the least.
There are few indications even of the relative order of the works, though some of them have evidently been worked at over a period of time – the Politics, for example, and the Poetics. Most of them are too elaborate and detailed to be regarded as mere lecture notes. Cicero speaks of Aristotle’s works in two classes: ‘esoteric’ works and ‘commentaries’. The esoteric works were presumably the ‘published’ dialogues whose style Cicero praised and which he sought to imitate in his own dialogues. The word ‘commentaries’ (hypomnēmata) is not a very specific term: it could mean anything from rough notes to ‘such sophisticated works as Caesar’s records of his campaigns’,7 and so could cover both the encyclopaedic collections and the treatises that now survive. What we now have would be called ‘esoteric’ in Cicero’s terms; meaning, not that they were secret or available only to initiates, but simply that they were for use ‘inside’, in the school. The early commentators – but not Aristotle – referred to these as ‘acroamatic’ – ‘works for listening to’. Though nothing is now known about the way these were actually used in the Lyceum, it is generally agreed by scholars that they were used in oral instruction and were not intended to be widely circulated outside the school. The Poetics is one of these – and a very small one. It takes up only fifteen pages (thirty columns) in Bekker’s Berlin Academy edition (1830–1) compared with the ninety-eight pages of the Nicomachean Ethics and 114 pages of the Metaphysics. The Poetics runs to about 10,000 words – that is, about one-hundredth of Aristotle’s extant writings.
A translator has to make up his mind about the primary document he is working with. Aristotle was, as we know, the inventor of what we now call a library, but we have no way of telling whether his successors regarded his working-manuscripts as sacrosanct in the way we now regard even the scribblings of some very minor writers: they were not, after all, drafts for finished written work. Plato’s disciples prepared a sort of Academy edition of his works; Aristotle’s successors seem not to have done so, and indeed we are lucky to have even the text of the Poetics as we have it. If the original Poetics, as a group of materials to be used in oral instruction, was the property of the school (as there is no reason to doubt) and remained in use after Aristotle’s death, the manuscript could well – for successive uses – have been revised, cut down, altered and added to. It is impossible to deny on theoretical grounds that whatever Aristotle had originally set down could have been altered and revised entirely out of existence, leaving behind a manuscript ostensibly Aristotelian (and certainly Peripatetic) that contains nothing of Aristotle’s beyond transmitted echoes. My own view, however, is that what has been passed down to us is genuinely Aristotle’s; that a primary text – or part of it – is preserved; that the text as we have it includes revisions, additions, and afterthoughts by Aristotle, and that at least some of these can be detected with varying degrees of confidence; that a number of spurious glosses have wandered into the text (perhaps from later marginal and interlinear notes) and that these can also be identified with some certainty, without working to the high-minded principle that everything inconsistent, paradoxical, unexpected or difficult is not Aristotle’s. I believe further that the substantial nucleus around which these accretal activities have occurred is distinct, coherent and shapely enough to give impressive evidence, at first hand, of Aristotle’s intelligence and imagination at work.
This is an expression of faith, but not on that account a shot in the dark; for it arises from many detailed considerations, not least the minutiae of the text itself. But when we cry, ‘Back to the Greek text’, the question arises ‘What– or which – Greek text?’ If the style were less terse and abrupt, if the state of the text were less problematical than it is, and the line of transmission of the text more direct than we know it can have been, there would be fewer difficulties in translating, and fewer chances of being deflected into anachronistic misreadings.
Although the Poetics first came to the Western world in the Latin translation of Giorgio Valla in 1498 (from a good manuscript), the only readily available Greek text was the Aldine edition of 1508 (in Rhetores Graeci, for it was not included in the great Aldine edition of Aristotle of 1495-8), which, though poorly edited from an inferior manuscript, reigned for more than 300 years. In the Renaissance, when the authority of Aristotle’s philosophy was already in decline, interest in the Poetics was widespread; through the commentaries of Robortello and Castelvetro it assumed a menacing and authoritarian aspect and gathered to itself some non-Aristotelian doctrine. By the end of the seventeenth century the wave of doctrinaire-expository enthusiasm had subsided, leaving the Greek text in an unpurified form, even though some of Castelvetro’s emendations are still worthy of consideration. The Greek text reaches us along a very shadowy route. With Aristotle’s other manuscripts bequeathed to his friend and successor Theophrastus it came eventually to Rome in 84 BC after the sack of Athens and must have been included in the edition (long ago lost) made a few years later by Andronicus – the basis for our present Aristotelian corpus. But the Poetics, unlike the other works, received no commentary and so was not submitted to early detailed textual examination, and for a time seems to have disappeared. No passage from it is certainly quoted before the fourth century AD; the earliest manuscript with which we have any direct connection was of about the ninth century, and the link is very tenuous; the earliest authoritative Greek manuscript is dated on palaeographical evidence as having been written at the end of the tenth century. The history of the modern text of the Poetics begins in 1867, when Johann Vahlen established that MS Parisinus 1741 (MS A) –already known to Victorius, Tyrwhitt and Bekker – was the best and oldest surviving manuscript. Three important discoveries followed: the identification of MS B (in MS Riccardianus 46) as independent of MS A but deriving directly from a common source from which MS A derived at second remove; the discovery of the Arabic version of a Syriac version older than the common source of MSS A and B; and the discovery of a thirteenth-century Latin version of a manuscript closely related to MS A. Butcher’s text of 1894 was the first attempt to combine MS A with other texts then available. In the light of successive discoveries, other editions have followed – Bywater (1909), Gudeman (1934), Rostagni (1937, 1945), Daniel de Montmollin (1951) – all of which are superseded by Rudolph Kassel’s edition of 1965.8
This can be said with confidence: the best Greek text a translator can now work from is a great deal better than any we have had before, not only for the reliability of the central text but for the variety of carefully examined considerations it brings to bear upon the many cruces. Nevertheless it is a long way away – in time and space – from whatever it was that Aristotle wrote down and bequeathed to Theophrastus. That does not necessarily mean, however, that what we have is a wildly distorted or truncated relic of the original. The second ‘book’ of the Poetics – the whole section on ‘iambic’ and comedy that balanced the long account of tragedy and epic – is lost, and must have been lost before the manuscript lambda from which MSS A and B derive on one side, and on the other the Syriac version; for there is no trace anywhere of the section on comedy beyond the few, partly conjectural, words in MS A which may have introduced it. If the heirs of Neleus had a cellar anything like mine it would have taken less than a century for a manuscript to suffer irremediable damage, and it is known that Aristotle’s manuscripts did not survive their incarceration without physical damage. Meanwhile it is clear that a translator cannot, without serious danger of systematic distortion, ignore the textual evidence that has been examined, refined and accumulated by a succession of Greek scholars of great distinction; and in the end, for better, for worse, he will have to make a number of textual decisions on his own account.
More than forty years ago, in 1923, Werner Jaeger established effectively for the first time the principle that Aristotle’s canon represents a development, and that in order to understand and interpret the individual writings it is essential to imagine as vividly as possible the man and the mind that made these writings and in what order.
Aristotle ... was the inventor of the notion of intellectual development in time, and regards even his own achievement as the result of an evolution dependent solely on its own law. ... It is one of those almost incomprehensible paradoxes in which the history of human knowledge abounds, that the principle of organic development has never yet been applied to its originator, if we exclude a few efforts which ... have been ... without influence. ... The main reason why no attempt has yet been made to describe Aristotle’s development is, briefly, the scholastic notion of his philosophy as a static system of conceptions. His interpreters were past masters of his dialectical apparatus, but they had no personal experience of the forces that prompted his method of inquiry, or of his characteristic interplay of keen and abstract apodictic with a vivid and organic sense of form. ... Everybody knew indeed, that he was a power to be reckoned with, and one of the foundations of the modern world, but he remained a tradition, for the reason, if for no other, that even after the days of humanism and the reformation men still had far too much need of his content.9
I am concerned here, not with the development of Aristotle’s work altogether, but with the Poetics (to which Jaeger makes only two references, one of them concealed10) and with the Poetics as Aristotle’s; or, to put it in Werner Jaeger’s words, I wish to disclose ‘his characteristic interplay of keen and abstract apodictic with a vivid and organic sense of form’. Recognising that the Poetics is by Aristotle, we may be expected to adopt an attentive attitude, and even to expend a little intellectual effort; but the labour may go to gathering ‘content’, and our interpretation could become – like much mediaeval and Renaissance commentary – minute, immensely learned, and totally devoid of any sense of the whole conception or of the energy that imparts wholeness. I feel Aristotle’s presence in the Poetics, and find myself saying, ‘We have a given text, made by Aristotle; it has a form which implies not only why it exists, but what it is, and what energy is disposed in its realisation, and what patterns of resistance have been interposed to lead that energy into self-expository form.’ But the text is in Greek, which few read; if there is to be a translation, I should want it – whatever else it did – to bring the reader to a vivid sense of the energy and shape of Aristotle’s thinking, and so to bring him into the presence of Aristotle thinking – Aristotle making this thing, Aristotle inventing for his purpose a method that allows him to do what he sees he must do. This after all is a very Aristotelian way of coming at things; to accept the poiēma as given and made; to consider its physis (nature); to infer the dynamis (power) that realises itself in the given poiēma, and to work out from this why it has assumed the form it has – which is to say, simply, what it is.
For I hold the view that a piece of vigorous thinking is an activity of imagination, with its own peculiar spring and set, an action of discovery; and that its form, though overtly discursive, is yet imaginative. If so, the outcome could be expected to be not a group of ‘conclusions’ or doctrinal precepts, but rather the record of a feat of inventive thinking and the starting-point for fertile, elucidatory, finely controlled and energetic reflection in response to it.
I should like a translation of the Poetics to disclose the drama of the discourse – the gesturing forth of the argument (for, as Aristotle notes in passing, drama means doing, acting) – so that the reader may be able to ‘experience’ or enter into that drama. If we were not dealing with Aristotle, that might not be either necessary or even much to the point.
But
Aristotle was the first thinker to set up along with his philosophy a conception of his own position in history; he thereby created a new kind of philosophical consciousness, more responsible and inwardly complex. ... Everywhere in his exposition he makes his own ideas appear as the direct consequences of his criticism of his predecessors, especially Plato and his school. It was, therefore, both philosophical and Aristotelian when men followed him in this, and sought to understand him by means of the presuppositions out of which he had constructed his own theories.11
The drama of his thinking in the Poetics flows out of the Platonic background, and is yet the unfolding, in an invented mode, of an energetic process of discovering and seeing quite his own;12 a self-clarification in the presence of what he is examining – in this case certain kinds of poetry.
As for Aristotle himself, his credentials as a person to speak authoritatively about poetry are rather strange. It is known that he compiled a list of all the dramatic performances given at Athens; he wrote dialogues On Music and On Poets; in addition to the surviving acroamatic Rhetoric, he wrote a dialogue in three books On Rhetoric, a summary of rhetorical theories in two books, and a summary of Theodectes’ Handbook of Rhetoric; he annotated or corrected a copy of the Iliad for his pupil Alexander (which Alexander treasured), and wrote out six books of Homeric Problems (some traces of which seem to survive in Chapter 25 of the Poetics). On the other hand, although Aristotle is known to have gained a reputation for his dialogues and wrote some verses, it is clear that he is not much interested in what we think of as ‘poetry’; he does not respond to the touch and tune of poetry as Plato did; neither in the Poetics nor elsewhere is there any notice of lyrical poetry, nor of the choric writing that we consider the glory of Greek tragedy; and his theory of metaphor, as far as it goes, is informed more by logical considerations than by a sensitive understanding of the transfigurations language can undergo in poetry. Yet his admiration for Homer is unbounded and declares itself repeatedly in the Poetics and elsewhere. And, if we have any tendency to suppose condescendingly that his theory of tragedy is limited by the small number of examples he happens to have had at hand to study, we do well to recall that, out of more than 300 plays by Aeschylus, Sophocles and Euripides he could have known, only thirty-two or thirty-three have come down to us; that he could have known over a thousand plays, and that his well-known compendious habits of inquiry tempt us to suppose that he may well have done.
When we try to place the Poetics in the context of his other work or to trace the development of the work itself, the evidence is far from conclusive. In Aristotle’s other works there are a few references to what must be the Poetics: one in the Politics (the promise of a fuller treatment of katharsis which has not survived), and five or six in the Rhetoric. We know that parts of the Rhetoric go back to the last few years of Aristotle’s stay at the Academy, and this would not be an implausible date for the earliest elements of the Poetics. I am willing to hold with Gerald Else that the earliest parts of the Poetics could have been set down at the Academy in the last years of Plato’s life, and that the document may have been worked over in the Assos-Mitylene period of his teaching and perhaps also while he was instructing Alexander, and may have been worked over again in the Lyceum.13
From classical scholarship a translator can take a sound Greek text, and can gain some acquaintance with Aristotle’s works and his ways of thinking so that appropriate connections can be made between the Poetics and other works of Aristotle and of Plato. Something further is needed, and for this there is little precedent – a prose style that will remain in close and continuous contact with the details of the Greek, an English vocabulary, syntax, and rhythm that will catch the immediacy and movements of the Greek.14
English, with its eclectic vocabulary, a strong tradition of Latinism in its philosophical terms, and of Latinistic structures in its formal writing, is not very much like the Greek Plato and Aristotle thought in. The Attic dialect, the language of Athens at the height of her literary magnificence, is the most cultivated and refined form of Greek: this was Plato’s dialect, and Aristotle had it by inheritance, even though – an Ionian by birth – his usage lies in the border country between pure Attic and the less strict, less eloquent koinē that was beginning rapidly to develop in the wake of the Greek empire. Attic Greek differs from English in being highly inflected, in the verb as well as in the noun, and is much more highly inflected and supple than Latin (to which in some other respects it is obviously similar). Greek is extremely rich in participles, which with a fully inflected definite article offer a wide range of substantival adjectives which function like verbal nouns, preserving the active initiative of the verbs that are radical to them. This alone goes far to account for the vivid directness typical of Greek philosophical writing – the general absence of special terms and a happy restraint from abstraction. Furthermore, Greek is capable of providing a wide range of cognate words on a single root: this allows for great variety of self-expository compounds, and also adds to the range of participial nouns which by altering their terminations can refer the root to a person, a thing, a product, a process, an intention even. Poiein, prattein aran, and mimeisthai are crucial instances in the Poetics. From poiein (to do or make) we have poiēma (a thing made – roughly our ‘poem’); poiētes (a maker – roughly our ‘poet’ – but poiētria is not poetry but a poetess); poiēsis (the process or activity of making – only very roughly our ‘poetry’, and unhappily the eighteenth century fumbled the ball in allowing ‘poesy’ to become an elegant variant of ‘poetry’ when we badly needed a word for poiēsis). From the noun poiēsis, the adjective poiētikos is regularly formed (to do with making, capable of making); and, since we have allowed the word ‘poetic’ to become merely the adjective of ‘poet’ and ‘poetry’, I should like to be able to use both the ‘poetic’ (in our sense) and ‘poietic’ (in the Greek sense). Also, a number of compounds can be formed by attaching a noun to –poiia (making) and –poios (maker) to provide ‘myth-making’, ‘song-making’, ‘a tragedy-maker’, ‘an epic-maker’, and the like. Greek is seldom at a loss for alternative words in any verbal situation; yet it may be that the many subtle variants it can devise upon a single root accounts for the semantic clarity that Greek words preserve over a long period of time, so that, even when transliterated into Roman letters and converted into English forms, they preserve – at least to those who know even a little Greek – their pristine clarity.
It is in words of active or indicative termination that English seems to me particularly weak for the business of translating the Poetics – words that by their form clearly imply process or continuous action. English has no word to match the processive implications that abide in the very form of the words mimēsis and poiēsis. Too often we have to fall back on nouns formed from Latin past participles (‘imitation’, ‘conception’, ‘notion’, ‘construction’) or upon collective nouns (‘poetry’, for example, which has to serve far too many uses); and the present participle ‘being’ hovers uneasily between noun and participle (it took a Coleridge to wonder whether ‘thing’ could be the present participle of ‘the’).15 Where Greek is strong, lucid, flexible and precise, and English too often, faute de mieux, driven to Latinism, a translator of the Poetics has to be crafty and unconventional, and write sentences that to an ear attuned to English philosophical writing of the last couple of centuries does not sound like philosophy at all.
Again, Attic Greek uses a variety of enclitics and particles which impart subtle shades of emphasis and relation. These also play an important part in controlling and shaping rhythm. The best Greek prose is wonderfully sinewy and fluent – athletic in its grace and with the superb athlete’s way of disposing energy in repose; by contrast, much English philosophical prose recalls the muscle-bound rigidity of the Hellenistic and Roman boxers. When the Greek is abrupt, without deliberate grace or sustained fluency – as is the case in the Poetics, even when the text is not corrupt – even then the rhythms still trace out the inflections of a speaking voice.
But what tune is it proper to have in the ear while translating the Poetics? Cicero thought well enough of Aristotle’s dialogue-style to fashion his own dialogues on it; but that tells us little enough – Latin not being Greek – except that Aristotle’s dialogues, as might have been expected, used more sustained discursive monologue than Plato does at his best. If, in the hands of a competent writer, prose style is the image of the mind that produces it, Aristotle’s prose cannot be expected to lack force, structural strength, subtlety or complexity. To my ear there are plenty of tokens of all these qualities in the Poetics, even though in his later years any desire he may once have had for literary distinction had been dissolved into a preoccupation with teaching. In the Poetics he is probably writing with only half an ear for the sound of what he is saying; but there are some elaborate sentences there which, by the way they somehow in the end, and contrary to expectation, unravel themselves triumphantly to a close, make me wonder whether we may be dealing with an absent-minded virtuoso. Parts of the Poetics are admittedly broken and terse, and some parts look more like jottings than sustained writing; but the opening chapters – at least fourteen of them – are continuous enough to give an impression of style, even a hint of mannerism, certainly the distinct tune of an identifiable voice. And one thing that emerges from what Aristotle has to say about style is that nothing matters so much as clarity.16
All we know is that the Poetics was acroamatic – something to be listened to. Suppose it is lecture notes on which Aristotle would improvise and expatiate, as many lecturers do: the trouble is that we don’t know how in fact Aristotle did speak from these notes, if they are notes; we have only what is set down. In translating, I have decided therefore to keep very close to the words, to add no grace, to smooth no roughness, thinking rather of Aristotle as a lecturer whose authority rests in the sustained gravity and openness of his speech; a man who chooses deliberate, even angular, plainness in preference to rhetoric, stylishness, or fine and memorable phrasing. The objection to a smooth rendering of the Poetics is that it will probably conceal the difficulties the text presents, and bury the fascinating and exacting cruces that often confront the reader in the Greek.
The requirements I have in mind for a translation of the Poetics are these. The reader must never be allowed to lose touch with the Greek, even if he does not know any Greek. Latinistic words are to be avoided as far as possible. When a suitable English word does not match a central Greek word, the Greek word can be transliterated (for example, ‘mimesis’, ‘opsis’, ‘lexis’, ‘poietic’), not in order to introduce a technical word of invariable meaning (which is the business not of language but of mathematical symbolism), but to remind the reader of the root meaning and implied functions of the word. The writing would have a spoken rhythm to allow for the vigour, informality, brokenness and sudden changes of direction in the Greek; it would be easy in movement syntactically a little ramshackle, perhaps, to catch the sound of a voice that is good to overhear, bespeaking the grave unhurried self-possession of a man who is confident that he can think aloud coherently and inventively.
Even if I could manage all that, it would not in itself be enough for what I have in mind. The counterpart to the gaps in the discourse that Aristotle himself might have filled or elaborated would be some sort of commentary; and the counterpart to knowing the Greek is to jog the reader’s elbow constantly (if need be) to tell him what the Greek is doing, or why at any point the English is markedly different from the Greek. I would show in square brackets in the text whatever the translator has supplied by way of elucidation or implied comment, and would draw the reader’s attention away from the translation as often as and wherever necessary with editorial footnotes leading to a sparse and pointed commentary. This does not make for easy reading; but who ever thought the Poetics was going to be easy reading? The aim is to find Aristotle, not to miss him.
Another editorial or typographical device that seems to me important in a version of the Poetics is to separate out from the main text all identifiably intrusive elements. These are of two kinds, (1) Interpolations into the text by other hands, presumably at some time after the original text was consolidated. These are seldom emendations of the text itself, but are usually marginal or interlinear notes carried into the text by later copyists. The provenance of some spurious interpolations in the Poetics can be traced from manuscript evidence; if there are certainly some of these there may well be others. (2) Aristotle’s own notes and afterthoughts, which in modern book-making would be printed as footnotes and appendixes. To identify these is not easy. A good textual critic, guided by his respect for the integrity of the Greek text, is a curious mixture of daring and conservatism. Any claim to have identified an interpolation or dislocation of the text will be narrowly scrutinised by other scholars equally fastidious, daring, and conservative; few such identifications are accepted without qualification by many scholars. But in some cases there is impressive agreement, and, as long as the motive is not to resolve intractable difficulties in interpretation by tearing up the paper the problem is written on, it is well to give distinctive treatment to Aristotle’s additions. In this matter I am prepared, for pedagogic purposes, not to be excessively conservative.17
As an acroamatic document, the Poetics cannot be envisaged as a draft for a publishable treatise, with corrections, alterations, and additions written in to be accommodated to a final text. Some of Aristotle’s additions look like the sort of additions that in a later draft are ballooned and arrowed into a context without final adjustment of the syntax and adjustment of the argument. Some are noticeable for their expansive and relaxed style; others are evidently later than the original because they suggest a new line of attack or use a revised vocabulary; a few seem to be blocks of material taken out of something written for other purposes but found convenient to extend the argument or to provide broader illustration. In my scheme all spurious intrusions are clearly separated out of the text, but kept in sight; Aristotle’s additions are kept in the text but given distinctive typographical treatment; a few larger additions are printed as appendixes; a few paragraphs are repositioned.18
The point of using these distinctions in presenting the text is not to ‘remove incoherencies and inconsistencies’; rather they give some hope of restoring the document to the status of an organic and living thing – zōion ti (a favourite phrase of Aristotle’s). The purpose is to make clear ‘that provisional form which, being thoroughly characteristic of Aristotle’s philosophy, constitutes the inevitable starting-point for every historical understanding of it’. The Poetics is not chaotic: the schēma is beautifully direct, orderly, and elegant in its logical and thematic development. Yet, for the intelligent and strenuous reader who has no Greek and therefore has no direct access to the textual problems, there seems little point in printing the translation ‘plain’; then the reader would be left to resolve or ignore problems the solution of which would heighten his dramatic sense and energise his understanding. I would therefore insist upon some typographical clarification of the textual problems short of imposing dogmatic finality upon their solution. I would also introduce paragraph-numbering for large-scale reference in place of the rather perverse chapter-numbering that tradition has carried with the manuscript, while still preserving the Bekker lineation for small-scale reference.
A few examples will illustrate the sort of translation I have in mind and the kind of details that I think would be useful in a commentary to go with the translation. The two translations that I find closest to the tone I intend are George Grube’s (for its firm muscularity) and Gerald Else’s literal version in his Argument (for its close contact with the Greek and its grave self-preoccupation). But Grube’s rendering is so polished as to deflect minute inquiry; and I owe too much to Else’s work to venture an open comparison. I have therefore chosen S. H. Butcher’s version, as an example of received standard glyptic, and Lane Cooper’s, for its relaxed and Latinistic verbosity. Let us begin in the natural way at the beginning.
Butcher, 1911:
I propose to treat of Poetry in itself and of its various kinds, noting the essential quality of each; to inquire into the structure of the plot as requisite to a good poem; into the number and nature of the parts of which a poem is composed; and similarly into whatever else falls within the same inquiry. Following, then, the order of nature, let us begin with the principles which come first.
Lane Cooper, 1913:
In this work, we propose to discuss the nature of the poetic art in general, and to treat of its different species in particular, with regard to the essential quality or function of each species – which is equivalent to the proper and characteristic effect of each upon the trained sensibilities of the judicious. Accordingly, we shall examine that organic structure of the whole which is indispensable to the production of an ideally effective poem, together with such other matters as fall within the same inquiry respecting form and function. Turning first to the conception of poetry in general, we may follow the natural order, and begin with what is fundamental, the principle of artistic imitation.
I propose to translate as follows:
The poietic [art] [1] in itself and the various kinds of it, and what [particular] effect each kind has, and how plots are to be put together if the making [2] is to prosper [3]; and how many elements it has and what kind; and likewise everything else that belongs in this area of inquiry – let us discuss all this, beginning in the natural way with first things [4].
The commentary would discuss four points.
[1] The opening words are peri poiētikēs [technēs] – from which the book takes its title. Neither ‘poetry’ nor ‘the art of poetry’ is quite right. The root of poiētikē – poiein (to make, do, fashion, perform) – is a strongly active verb that will dominate the whole discussion in the sense ‘to make’. (Emphatically, it does not mean ‘to create’.) I have written ‘poietic’ art, rather than ‘poetic’ art, partly to emphasise the sense of ‘making’ (and the poet as ‘maker’), partly as a reminder that Aristotle does not recognise a distinction between ‘art’ and ‘craft’.
[2] Poiesis, radically the process of making.
[3] Kalōs hexein – ‘to go well with, to work out luckily’. Else translates ‘to be an artistic success’, but I prefer a more direct and idiomatic word.
[4] The way the discussion later develops in detail shows that this sentence is neither a systematic preliminary outline nor a statement of the programme Aristotle intends to follow. He seems to be sidling comfortably into his discourse. But by taking his starting-point in ‘first things’ he shows that he is thinking of the poietic art as cause, or ‘reason why’.
After the prefatory sentence-paragraph, the plot thickens immediately and the difficulties are formidable.
Now epic-making and the making of tragedy [5] – and comedy too – and the art of making dithyrambs, and most of the art of composing to the flute and lyre – all these turn out to be, by and large, mimeseis [6]. But these arts differ from one another in three respects: for they do their mimesis [7] (a) in different matters (in-what), (b) of different subjects (of-what), and (c) by different methods (how) [8].
[5] In the first sentence poiein or some derivative of it is used three times (even recognising that by Aristotle’s day epipoiia often meant ‘epic’ rather than ‘epic-making’). Aristotle is clearly not talking about epic, tragedy, comedy, etc., as genres or art forms; he is talking about the making of them.
[6] This word, the plural of mimesis, is transliterated to avoid using the word ‘imitations’. Mimēsis is in its form a processive word – a point of great importance for much of what follows. A useful habit is to read ‘mimesis’ as ‘a process – mimesis’. ‘The mimetic process is the activity of poiētikē’ (Else); its dynamis (potentiality) works towards a telos (end) which is in both a substantial and active sense, a poiēma (poem). Aristotle does not define either ‘the poietic art’ or mimesis; he leaves both open for exploration and for progressive self-definition in the body of the discussion.
[7] In this paragraph, as in many other places, Aristotle uses mimeisthai, the verb cognate to mimesis. If the verb is translated ‘to imitate’, the meaning is deflected towards an assumed commonplace definition for ‘imitation’. In order to keep clear that mimesis is an activity or process and not a thing or product, I use the phrase ‘they do their mimesis’ for mimountai; ‘they make their mimesis’ would allow mimesis to be thought of as a product, an ‘imitation’.
[8] This sentence does what is the despair of any English translator, and does it with Greek clarity and forthrightness and in a manner usual with Aristotle. Literally ‘they differ in as much as they do their mimesis in different things, of different things, and differently and not in the same way’. The traditional abstract terms for these three differentiae are ‘medium’, ‘object’ and ‘mode’. I prefer ‘matter’, ‘subject’ and ‘method’ for the following reasons.
Matter (in-what). Even if the word ‘medium’ were not now corrupted below fastidious use, it would not be quite correct here. In current vulgar usage ‘medium’ refers to various means of public presentation – printed matter, public speech, stage, film, radio, television: in short, ‘medium [of communication]’ – whatever the question-begging term ‘communication’ means. Aristotle’s three ‘in-what’ differentiae are rhythm, melody and speech. In our way of thinking, these three are not at the same level: rhythm is radical to both melody and speech. Although Aristotle seems to think of each emerging as dominant in dance, music and (dramatic) poetry, he does not encourage us to suppose that he thinks of any one of them functioning in isolation from at least one other. Aristotle’s ‘in-what’ is the physical stuff in which the action is embodied and assumes form – e.g. for music, patterned sound, and for painting, patterned colour-and-line-in-space. We know too little about Aristotle’s thoughts on the work of art as ‘mediating’ between (say) poet and reader to use the word ‘medium’ confidently. What we do know is that Aristotle has a very strong sense of physical actuality. Since he seems to have been the first to attempt a classification of the arts according to the physical materials they use, the choice of a correct term for ‘in-what’ is important.
Subject (of-what). ‘Object’ is unsatisfactory because (a) it tends to imply that the model imposes a predictable or desirable form upon the work of art, as is sometimes naïvely assumed to be the case for painting: (b) it may be mistaken for ‘aim’ and become so confused with Aristotle’s teleological principle that the starting-point comes to look like the ‘end’. ‘Subject’ presents no difficulty or deflection: we commonly speak of the ‘subject’ of a book, play, picture or poem meaning in the most general way ‘what it is about’ and implicitly what it starts from.
Method (how). The usual word ‘mode’ (as in ‘narrative mode’, ‘dramatic mode’) is unsatisfactory because it indicates a static classification into which individual works may fall. ‘Method’ places the initiative in the maker and helps us to concentrate on the work as in process of making or acting – which is consonant with Aristotle’s emphasis throughout the Poetics. Fortunately this sense of the word ‘method’ is familiar to us from twentieth-century critical analysis of prose fiction, drama and poetry.
Let us go on, straight through the next long paragraph which happens to include two allegedly spurious insertions, one certainly spurious word, and a passage that I treat as a discursive note or afterthought of Aristotle’s.
[Differentiation by Matter]
You know how some people make likenesses of all kinds of things by turning them into colours and shapes – some imaginatively and some [merely] by formula – [9] and how other people do their mimesis with the voice [10]: well, in the same way, the arts we are thinking of all do their mimesis with rhythm, speech and melody [11], but using speech and melody either separately or mixed together. For example, flute-playing, lyre-playing, and any other [instrumental] arts of this sort – like playing the panpipes – use only melody and rhythm [12]; while the other [verbal] art [13] – an art that happens so far to have no name* – uses only prose [speeches] of [unaccompanied] verses, and when verses, either mixed or of only one kind.
*[A discursive note by Aristotle] [Speaking of lack of suitable terms] we haven’t in fact even got a common term to cover the mimes of Sophron and Xenarchus and the Socratic dialogues; and, again, if somebody does his work in trimeters, elegiacs, or some other such verse-form [we have no name for it] – except of course that people get into the habit of attaching the word ‘poet’ to the verse-form, and speak of ‘elegiac poets’ and ‘epic poets’ – not because they are entitled to be called poets for the quality of their mimesis but because as practitioners they are lumped together according to the verse-form they write in. And if a man puts together some medical or scientific work in verse, people usually call him a ‘poet’; and yet Homer and Empedocles have nothing in common except their use of verse, and properly speaking the one should be called a poet, and the other not a poet but a science-writer — and the same would apply even if he used a combination of all the verse-forms (as Chaeremon did in his Centaur [14]) [15].
For these arts, then, let this be our division [according to matter].
Looking back, a few comments are in order.
[9] Aristotle’s word is ‘habit’ or ‘routine’. Coleridge once referred to Southey’s verse as ‘cold-blooded carpentry’, but that is probably stronger than Aristotle intended. The word ‘imaginatively’ is anachronistic, but I cannot think of a better here.
[10] ‘Sound’ will not do. Phōnē is specifically the human voice – ‘the most mimetic of the human faculties’ (Rhetoric 1404a21).
[11] The word is harmonia – the due fitting-together of musical sounds. For Greek music this applied horizontally – melodically. Our use of the word ‘harmony’ implies a vertical or chordal relation.
[12] ‘And the dancer’s art uses rhythms alone, without melody, for it is through their rhythmic figures that dancers represent characters, feelings, and actions.’ Else in his Argument agreed with Vahlen in taking this passage for an afterthought of Aristotle’s: it certainly disrupts the run of the sentence. In his Translation, however, Else omits the passage as spurious – the way it is represented here.
[13] The word ‘epic’ has been introduced here, probably from an explanatory gloss; but it is obviously wrong and is marked as spurious by Kassel. When the phrase on dancing [12] is not allowed to interrupt the sentence, it is clear that Aristotle is making a contrast between ‘bare’ instrumental music (without song) and the ‘bare’ verbal art has no instrumental accompaniment – ‘an art that happens so far to have no name’.
[14] ‘... a mixed epic work (miktēn rhapsōdian) – but he [Chaeremon] is entitled to be called a poet’. Whatever miktēn rhapsōdian means, Chaeremon’s Centaur (which has disappeared except for five iambic lines) was a drama, perhaps a closet drama, possibly a tragedy but more probably a satyr-play. Yet a rhapsody is normally a portion of epic of a length that can be given at one performance. Chaeremon seems to have been a contemporary of Aristotle. Aristotle’s point in any case is not that Chaeremon was not a poet but that he used a mixture of all the metres.
[15] Whether this section is to be regarded as a note or an afterthought or even a ‘later’ addition is probably not worth quarrelling over. To mark it off typographically draws attention to its looser rhythm and more leisurely conduct in contrast to the trenchancy of the argument so far. This difference is felt if the passage is left embedded in the text, but we may get the impression that Aristotle has lost the thread and is drifting away from his announced discussion of differentiation by matter (in-what). It is worth noticing that the ‘art that happens so far to have no name’ is not what we should call ‘lyrical poetry’, but prose by itself and verse without music. And what seems to have led Aristotle to complain about the lack of proper terms was his insistence that the word ‘poet’ should not be used sloppily.
One more passage will give an example of one of Aristotle’s more complicated sentences sustained against fearful odds, and will also show what happens to the central passage about the relation between plot and character in my version. The shape of the Poetics is, in outline at least, straightforward and purposeful. Part I, quite short, deals with the differentiation of mimesis secured by the matter (in-which), the subject (of-what), and the method (in what way). In part II – also quite short – Aristotle discusses the origins and growth of the poietic art. This is not so much drawn deductively from historical evidence (if indeed much was available), but is a theory of how ‘it stands to reason’ the poetic art took its origins and grew towards fulfilment – a very Aristotelian way of working. The poietic art, he says, grew out of two human radicals: a flair for mimesis (which in this context is very much like ‘imitation’ in Plato’s sense), and a feeling for rhythm and melody. His first and basic division for the poietic art is bravely and incontrovertibly moral: two species establish themselves according as the subjects and central figures (? and poets) are spoudaioi (‘serious’, morally superior, praiseworthy) or phauloi (‘mean’, trivial – or, as Else happily suggests, ‘no–account’). Hence on the one side epic and tragedy, and on the other ‘iambic’ (rough lampooning) and comedy. Each species ‘finds itself’ discovers its own nature and form, and progressively – even inevitably – moves towards realising its own peculiar nature. He then turns in Part III (which is all the rest of the manuscript as we have it) to discuss tragedy and epic together, with tragedy in the forefront until the closing chapters, when epic is distinguished from and compared with tragedy and found inferior to it. And all the time he is talking not about things-made so much as about things in the making, coming into being, finding themselves.
At the beginning of Part III Aristotle sets down the famous definition of tragedy after saying ‘let us pick out the emergent definition of its integral nature’ – emergent, that is, according to his theoretical ‘history’ of the way tragedy found itself. Then he discusses the six merē of tragedy, literally ‘parts’ and often translated ‘constituent elements’; but, since Aristotle is thinking of tragedy as a special instance of the poietic art, the merē must be related to the making and coming-into-existence processes of tragedy. Merē are not component parts, and the Poetics is not a do-it-yourself tragedy kit. So I translate merē as ‘aspects’ – various points of vantage from which we can examine the making and functioning of a drama. Aristotle points to six ‘aspects’: opsis (an impossible word – ‘look’? ‘visuals’?, but preferably not ‘spectacle’19), melopoiia (song-making – both words and music), lexis (speech or dialogue, not ‘diction’), characters, ‘thought’, plot.
I take up the text again at the point where he has finished with opsis, melopoiia and lexis; the syntax of the first sentence is left much as it is in the Greek.
Since [tragedy] is a mimesis of an action and [since] it is acted out by certain people acting and these must necessarily have a certain kind of character and cast of mind (for it is in the light of these that we say that their actions are of a certain kind, and according to [their actions] they all succeed or fail); and [since] the plot is the mimesis of the action (for I use ‘plot’ in this sense – the putting-together of the events) and the ‘characters’ are what allow us to ascribe certain qualities to the actors, and the ‘thought’ is the places where [the actors] by speaking prove some point or declare wisdom – because of all this, the [number of] ‘aspects’ to tragedy-[making] as a whole that account for tragedy as a distinct [species] must be exactly six: plot and characters and speech and thought and ‘visuals’ and song-making. ...
But the most important of these is the putting-together (? structuring) of the events. For tragedy is a mimesis not of men [simply] but of an action, that is, of life.20 That’s how it is that they certainly do not act in order to present their characters: they assume their characters for the sake of the actions [they are to do]. And so the [course of] events – the plot – is the end of tragedy, and the end is what matters most of all. Furthermore, you can’t have a tragedy without an action, but you can have it without [clearly defined] characters. ... So it follows that the first principle of tragedy – the soul, in fact – is the plot, and second to that the characters; it is a mimesis of an action (praxis) and therefore particularly [a mimesis] of men-of-action in action.
I am aware of the uncouthness of the style in these passages, but I have retained it for a distinct purposes: to hold the English to what I feel sure the Greek is saying and doing, to the way that argument runs and the emphasis falls. I must now say what guides that purpose and encourages that confidence; but sketchily, because I am not primarily concerned here with points of interpretation but with the attitude of mind that might discover a vivid interpretation if one were wanted.
Far and away the most insistently recurring words and ideas in the Poetics (though not so much near the end as in earlier chapters) are ‘making’ and ‘action/acting’. Nevertheless, the radical error (prōton pseudos) that is most commonly made about the Poetics is to suppose that Aristotle is discussing – as we might – tragedy, epic, comedy, and the rest as genres or as somehow things-in-themselves. This may be a valid-enough way to think of these things, but it happens not to be Aristotle’s way. The second error is to suppose that Aristotle has drawn together all the literary works he can lay hands on, has classified them, and drawn certain general conclusions which he then proceeds (in the standard backward philosophical way) to explicate and ‘prove’. There are a number of reasons for rejecting these two assumptions, inevitable though they may be to some mentalities and even though many translations, especially the earlier ones, imply or endorse them. The dominance of ‘making’, ‘action’ and process in the Greek text makes it plain that – whatever preliminary investigations Aristotle may have made (and we may reasonably guess that they were comprehensive and minute) – he is here not working by deduction but by inference. In short, he is working in the distinctive Aristotelian way. He is seized by the individual, the particular, as substantial. What interests him, as Jaeger puts it, is the fact, ‘not that something is coming to be, but that something is coming to be’:21 something that will be final and normative is making its way into existence; when it has come into existence it will have achieved form, it will have become what it had to be. The form then is the final statement – assertion, if you like – of an activity seeking its own end, its own fulfilment.22
Aristotle sometimes uses the organic example of the seed or the developed organism.23 It has been argued – I think convincingly – that the distinctive nisus of Aristotle’s thinking is most clearly to be seen in his biological investigations: contrary to Plato’s ascription of reality to the Ideas only, Aristotle’s habit is to insist (as he does repeatedly in the Metaphysics) upon individuals – particulars – as substances, the only fully real things. His biological investigations provide commanding instances of a process that he recognises in everything he sees and everything he thinks about, embodied in the notions of potency and act. The power, potency (dynamis) can be latent or active; when it becomes active the dynamis is energeia, actuality, activity from within that drives towards, acts towards attaining its own end (telos): that is, dynamis is self-realising. (Coleridge, though he regarded himself as a Platonist and no Aristotelian, said that ‘Every thing that lives, has its moment of self-exposition.’24) For Aristotle, everything presents itself to him in terms of motion and end; and ‘in every kind of motion his gaze is fastened on the end’.25 Whether it is a snail, an octopus or dogfish, the convolutions of a nautilus shell or the evolution of the government of a city-state, or the activity of man as a moral creature, Aristotle’s fascinated and stern gaze is fixed on the inescapable mystery that this is, that this is the self-exposition of its dynamis, the end of its action. (Wordsworth was in this sense profoundly Aristotelian.) A ‘thing’ is for Aristotle never inert: it always implies its action and its power. Can it be that this central analogy for the dynamis that in many specialised aspects runs through all things as through a single hierarchical order is not the organic figure drawn from plants, fishes, and animals, but the human dynamis? ‘For the actuality of Nous [intelligence, intuition] is life.’26
When Aristotle looks at tragedy he wants to find out the form of tragedy; that is, in his terms, what tragedy of its own nature comes to be. The pre-Socratic philosophers had tried to account for everything in terms of the distribution of chaotic matter by mechanical causes; Plato and Aristotle, each in his own way, had moved away from that position. For Aristotle, action and power, motion and form are the dynamic modes which everything discloses. The higher we ascend in the order of the cosmos, he believes, the more purely the motion expresses the form that is its end; and the highest form must be pure activity. At the human level, it may be, he sees tragedy – and perhaps all art – as pure act in the human psyche. Tragedy as the end of a certain aspect of human dynamis; and, if the tragic action flows from the moral centre of man, tragedy will also tell us something profound about man. Aristotle’s theory of the origins and growth of art starts by identifying two causes ‘deep-rooted in the very nature [of man]’,27 shows how the main literary-dramatic forms emerged and identified themselves, and concludes that tragedy ‘when it had gone through many changes, stopped when it had realised its own physis - its integral nature’.28 When he deals with tragedy – and we must remember that he is simultaneously thinking of epic and tragedy under the heading of those literary kinds that arise out of ‘serious’ people and subjects – he has to find a way of thinking positively from the given end; he has to be able to infer accurately the action (energy) that the end realises. The action is human, the energy is human, the tragedy is human; there can be no other assumption, no other analogy. The action is plotted and prepared by a maker, a poet; it is acted out, brought into physical existence, by actors in the theatre (it is the actors, not the poet, who do the mimēsis – but the poet can also be an actor and often was); and the action is traced out and realises itself before an audience (though the tragic effect can come about through reading). Aristotle is very much aware of the complex web of relationships between poet, actors, performance and audience, and of their interactions; he knows all about the egotism of actors too, the silliness of audiences, and the way poets can be deflected by sensational appeals to vulgar taste. It has often been said that, even among the small group of superb Greek tragedies that have come down to us, few would meet Aristotle’s specification; but this, I take it, is one of the clear signs that he is not working deductively either from the huge corpus of plays known to him or from his own personal preference (which is known to have been questionable). He is looking for the form of tragedy, and needs to invent a method – a means – of finding that for himself and of disclosing it to others.
His method in the Poetics is brilliantly simple and appropriate. He says in effect, ‘Let us suppose that we want to make a tragedy; how should we set about it?’ Intensely aware of the complex and refined dynamic of tragedy, he is not content to say what tragedy is (as though that were easy anyway), but insists on showing how it works. As he advances, he concentrates on making and doing and acting – and is it not poetry-making, not poetry itself, that is a more serious and ‘philosophical’ business than history-making? The word dei (it is necessary) recurs, particularly in the later chapters, like a reiterated dominant seventh: it is necessary to do this; you must –, you should –. The Renaissance scholars took these utterances for rules, and once into that mood worked out some pseudo-Aristotelian rules of their own; in the end even Corneille, it is said, shook in his shoes at the thought of breaking the ‘rules’, and a little later it was said that because Shakespeare broke the rules, yet wrote passable plays, he must have invented a kind of tragedy that Aristotle – poor fellow, with his limited horizon – had never dreamed of. Such conclusions seem to me less than inevitable. I cannot seriously think of Aristotle giving a master-class in tragedy-writing. But he has thought of what is probably the only way, even now, of seriously and responsibly engaging the critical attention of a student of literature: it is a dramatic device of teaching. He says in effect, ‘Just imagine that you are capable of making one of these things. Just imagine that you are capable of tracing out the right action that will realise itself in the right end, that you are capable of entering into and generating the action which, acted out by others, will ruthlessly bring about this end, an end so profound and momentous that at best we can only catch a glimpse of it. Then we shall see the physis, the dynamis, the telos, the life.’ His dei, dei, dei is the insistent reiteration, within this dramatic supposition, of what matters most: the action, the specific action that needs to be traced out, by what conceivable means, working within the limits of what resources to what end. For every action implies its realisation in an end. But there is no formula to guarantee success, only the poet’s judgement and luck and vision. The fact that no poet ever worked successfully in the way Aristotle ‘recommends’ does not affect the validity of his imaginative scheme. His aim is not practical, but theoretical; yet paradoxically, as far as his aim is ‘critical’ it is intensely practical – it helps us with our doing.
Aristotle knows certain central things from his experience of tragedy: that tragedy happens only to people of a certain kind or quality; that, if part of the horror is seeing a man broken, it must be a strong man (and that is implicit in the pleasure peculiar to tragedy); that if the issue is to do with law and man’s nature, the man must be morally strong – without the strength, what we see is merely pathetic, pitiful, or revolting; that tragedy is to do with the darkest and strongest issues in our experience – life and death and law and responsibility and freedom and necessity. He knows that we can betray ourselves from within, that when we take the law into our own hands we pass from freedom to mechanism and cease to be human, having cut ourselves off from the law of our inner nature; and he knows that a man can know that he is doing this and yet do it, and watch himself doing it, capable even in his fascination of altering or reversing the action. A certain quality of moral awareness is required in the person this can happen to, and a certain degree of strength; and it can happen only over something that really matters, such as the defiance of blood-relationship or some other primordial human bond.
A tragic action correctly traced will lead to the end of recognising at least something about the nature of man, the values that are paramount, the vulnerable centres that we must at all costs preserve – which is the law, our law. Here, it may be, the old debate about what happens according to nature (physei) and what according to law (nomoi) comes into ironic coincidence in Aristotle’s mind when he sees the form of tragedy, when the inner law simply is our nature – not ‘natural law’ or ‘the law of Nature’ but the law of our nature.
‘Tragedy is a mimesis [process] not of men simply but of an action, that is, of life.’ To achieve the precise end, a precise action is needed. We could think of the tragic action as a sort of trajectory traced by a projectile, implying a certain amplitude, direction, velocity, momentum, target, and that in every moment of flight all these terms are implied; and the nature of the projectile matters very much, because it is a man who, being morally strong, makes choices, determines the flight, is not simply propelled, is not a mere victim. Aristotle, I suggest, is showing us the tragic action as though it were a pure abstract motion traced out with exquisite precision, the precision that is needed to impart the force of necessity to an action that can at no point be predicted for certain because it can at any moment be altered or deflected: it will at once feel both inevitable and free. The plot, the sequence of events that specifies the action, Aristotle says, has to be conceived as a schēma, an abstract motion, and you put in the names afterwards; but the schēma is not simply a locus of dramatic points or a flight plan, for the points are not so much intersections in time and space as events, each momentous, crucial, chosen, formative. Yet the tragedy is inside the protagonist and is of his own doing; and, if he did not know, he could have known, perhaps should have known – which is why knowing and not-knowing is crucial to the tragic action. Recognition (anagnōrisis) is not a device of plot-structure, but an essential crisis in the action; and hamartia a mistake rather than a sin, a distinction that was clearer to Peter Abelard and other subtle Fathers than it seems to be to us – harmartia is an ignorant act, and in tragedy (as in ‘The Ancient Mariner’) ignorance is no excuse, for in these matters the plea is made not to a court of external law, but is argued in the inner dialogue of moral choice according to the law of our nature. And these things have to be declared outwardly, presented openly in action, so that they strike us not only with the frisson of horror and pity but with the shock of recognition; we too must be drawn into that intricate web of knowing and not-knowing. And that is the peculiar pleasure of tragedy.
To claim that Aristotle is simply talking about a ‘tragedy of action’ out of poverty, not knowing anything else, and that later dramatists discovered a ‘tragedy of character’ that Aristotle had never considered possible, is a radical misunderstanding of Aristotle’s position. To establish the existence of a ‘tragedy of character’ of comparable force and incisiveness it would be necessary to show that the ‘tragedy of character’ does in fact trace the specific action required of tragedy and that it does so with resources not accessible to the ‘tragedy of action’. The resources for tracing out the tragic action are very few: plot, characters, speech, song, the various techniques of stagecraft and acting. Language is indispensable, speech being one of the principal resources of human action, if also the most ambiguous. But things need to happen, not simply in sequence but in a sequence that implies the whole ineluctable trajectory and the end. The people involved have to be the sort of people that such things can happen to; at least one of them has to be capable of irreversible moral choice and yet capable of making a disastrous mistake in at least one of his moral choices, and it still has to be a moral choice, not just an accident or ‘the will of the gods’ or ‘Fate’. The plot is the sequence of events that in one sense delineates the action (the action which alone can produce the end); and the persons involved in the action delineate the action by being the sort of persons that could initiate such events and have them happen to them. ‘You can’t have tragedy without action, but you can have tragedy that is weak in character’ – that is, without persons who are shown taking strong moral choices. Aristotle cites examples, but this must be at an extreme limit of tragic possibility since the tragedy aēthēs – ‘without characters’ – throws away one of the most powerful and subtle resources for delineating tragic action; for the praxis (action) of the play is defined by the praxis of the persons in the play, and the praxis makes the characters what they are as well as what they are becoming and will become. When Aristotle says that ‘the first principle of tragedy – the soul, if you like – is the plot, and second to that the characters’, he means this quite specifically, not rhetorically: the soul is the ‘form’ of the person, and prior to the body – the plot is the ‘form’ of the tragedy, and prior to the action – the characters are the ‘body’ of the action (will body forth the action) and are shaped by, as well as generating, the action. The person acting does not disclose or externalise his character in action, as though the character existed before the action: the character (in Aristotle’s view) is shaped by his actions, and in tragedy we see the protagonist, as character, being shaped by his choice and his actions.
This is why the notion of hamartia as a tragic or fatal ‘flaw’ is completely wrong-headed in Aristotelian terms, and why to insist upon such a notion erodes the austere purity of Aristotle’s view of tragedy. If the protagonist had by nature a ‘flaw’ that steered him more or less inevitably into a fatal situation, he would be a mechanism and predictable to us, incapable of inducing terror or recognition; he would be repulsive or pathetic merely; he would no longer be a man-of-action in action shaping himself towards his telos in this action, but a man who – having fallen into mechanism – was no longer capable of discovering his ‘form’ in and through action.
And as for katharsis – the word occurs only once in the Poetics;29 in the central definition of tragedy (Ch. 6) to be sure, but so completely unrelated to anything in the introductory chapters that some textual critics have regarded the phrase as a later insertion by Aristotle. The discussions of katharsis in the politics, and of pity and terror in the Rhetoric, are of questionable relevance to the Poetics; and the promise in the Politics to ‘explain this further in my discussion of poetry’ is not fulfilled anywhere in the surviving corpus. So the one phrase – ten words – has accumulated a massive exegetic literature.30 Katharsis, as we know from experience, has its implications, in some sense, for the audience; but is it a technical word at all? And are we prepared to accept that one of the distinctive formative principles of tragedy (some seem to claim that it is the final cause of tragedy) is to be found not in the action but in the audience’s reaction? Gerald Else, as far as I know, was the first to insist that the katharsis occurs primarily inside the action; and Kitto, I think (pace Lucas), settles the matter for good and all. It is the incidents within the action itself (not the emotions of the audience) that are purified, brought into a sharp focus specific to tragedy, by the mimesis, by the presentational action – by the mimesis, not by ‘tragedy’. Events in the area of pity and terror are minutely defined in a kathartic process towards Unity of Action – that is, Purity of Action; and so the pleasure peculiar to tragedy, because of this refinement, is aroused by the quality of the action.31 And comedy has its katharsis too, presumably, in as much as its action needs to be ‘purified’ within its proper area and only so refined will arouse the pleasure peculiar to comedy. And yet, as Kitto says in a wistful aside, ‘There are times when one suspects that Aristotle’s own lectures on the Poetics would be more valuable even than the original text.’
Few of these observations are in any way new. But these are the sort of things, as a vigorous and single guiding view, that I should want a translation of the Poetics to keep steadily in the reader’s mind, in the final choice of each central term, in the shaping of phrases that too often and too easily recall improper connections, in a pungent running commentary that keeps the reader off his comfortable heels, in the rhythms of a speech that might conceivably be coming unguardedly but deliberately out of the intelligence of the man who was affectionately known to his contemporaries in the Academy as ‘The Brain’. To do so successfully, a person would need to be pretty skilful, learned, and lucky.
In this view, the action of tragedy (to think of only one of the ‘kinds’ Aristotle has under his eye) is not a ‘representation’ or ‘imitation’ at all, but the specific delineation, within extremely fine limits, of a moral action so subtle, powerful and important that it is almost impossible to delineate it; an action self-generated that has as its end a recognition of the nature and destiny of man. (No wonder few ‘tragedies’ meet the specification.) In this view, mimēsis is simply the continuous dynamic relation between a work of art and whatever stands over against it in the actual moral universe, or could conceivably stand over against it. So mimēsis is not definable by itself, least of all as a simplistic preliminary to a subtle inquiry. For this very reason, I imagine, Aristotle does not define it except in action, by a variety of uses gradually drawing around the word the limits of its activity – which is ‘definition’ in another but perfectly legitimate sense.
It would follow – and I should be prepared to argue – that the notion of ‘mimetic’ and ‘non–mimetic’ art is a verbal fiction based on a misunderstanding of Aristotle’s use of the word technē. I should be prepared also to challenge R. S. Crane’s statement that there is a ‘Coleridgean method’ of criticism distinct from and diametrically opposed to Aristotle’s; I begin to sense an Aristotle-Coleridge axis in criticism and poetics but am not yet prepared to speak about it.32 And I would affirm my own strong conviction that the method of the Poetics provides – for those who care to explore it – a paradigm for all those critical procedures that seriously seek to discover the nature of what they are examining, that seek to release with accurate definition the energy contained within what precise shaping limits. It seems to me more than possible that what Aristotle has to say about tragedy is absolute, that his account is not limited by the number of examples that he happened to have at hand. It is the privilege of genius to make such discoveries on incomplete evidence and to make durable statements about them.
In trying to discover and disclose the driving energy of Aristotle’s thought in the Poetics, I have addressed myself to making a translation with commentary – and have in the end come upon the clean air of Aristotle’s penetrating imagination and his grave, unwinking intelligence, to find the Poetics a dramatic record of his profound and incisive thinking, contemplative reflection of the highest order with a brilliant method of exposition to match it. Here indeed is theory, theoria, vision. My exhilaration may perhaps be pardoned even if it is not universally shared. Immanuel Bekker, whose Berlin Academy edition has provided the standard system of reference to the whole Aristotelian canon, edited in all some sixty volumes of Greek texts and collated more than 400 manuscripts. Gildersleeve said of him that in company he knew how to be silent in seven languages. Less learned than Bekker and less taciturn, I did not feel that, in the matter of the Poetics, I could any longer – whatever the hazard – be silent, in the one language I know at all well.