Andrew Ford
The fact that the term “aesthetics” was only introduced into philosophic discourse by Alexander Baumgarten in 1735 is regarded by some historians of the subject as a sign that the ancient Greeks were not much interested in questions about the nature of beauty and the arts. (For them, ta aisthētika would have suggested sense perception in general.) Other historians treat this fact as a factoid, a historical curiosity that need not stand in the way of assuming that the ancients responded to works of art in ways much like ourselves and confronted the same problems as we do when we try to think about beauty and art. The problem with this reasonable assumption is that it is hard to amass many actual texts that deal directly with the subject: most ancient discussions of issues we would define as aesthetic are embedded in works devoted to quite other topics, especially politics, ethics, or metaphysics. Thus Kristeller’s much-quoted denial that the ancients had anything comparable to modern aesthetics remains a challenge: “ancient writers and thinkers, though confronted with excellent works of art and quite susceptible to their charm, were neither able nor eager to detach the aesthetic quality of these works of art.”1
The present chapter hopes to throw light on this question by turning from the Greeks’ philosophies of art to their literary criticism. The daily, practical business of commenting on poems, interpreting them, and evaluating them implied the existence of standards for judging literature. This was especially so in the “agonistic” musical culture of the Greeks, which regularly set one poem against another in a competition to see which was the “finest” (kallistos: Ford 2002, 272–293). Already in the Archaic period, the most prestigious performances of Homeric epic took the form of competitions among professional reciters (rhapsōidoi), and the Athenian contests in tragic, comic, and dithyrambic poetry are well known; amateurs exchanging songs at drinking parties often turned to singing games. All such events raised the question, whether explicitly or implicitly, of what criteria to use when declaring one poem “finest” of all. I propose to study closely two texts from Classical Athens in which the question of standards is discussed. The pair is notable for suggesting that works of verbal art ought to be judged on their own terms. Taken together and set in context, the texts suggest that, in practice if not always in theory, the ancients recognized an aesthetic dimension to literature to a far greater extent than is sometimes allowed.
In criticism, the possibility of a literary aesthetics emerges most clearly in the question of whether poets have any autonomy in the sense of immunity to certain kinds of objections that may be laid against their work. In the texts to be examined, the specific form this question takes is whether poems have any value that is unconnected to social, moral, or political values. Claims for the autonomy of the aesthetic, of course, were widely asserted and roundly challenged in Romantic and modern theory (see Abrams 1985). But critical practice is concerned less with questions about the status of the work of art in itself – whether it is a thing in its own right with values of its own – than with the poets’ right of appeal against certain strictures on their work. In practice, as I will argue, poetry could in certain contexts enjoy a rather robust concept of autonomy.
To focus on critical practice rather than poetic theory is not, of course, to sidestep the formidable problems with establishing the aesthetic as a separate realm of experience or philosophy. For “literary criticism” is as difficult to define as aesthetics. What makes literary criticism literary? Is it that it has a special kind of object in view that calls for special standards to be fairly evaluated? How, then, could we set such objects apart, without circularity, from other human activities and productions we may wish to judge? Or is literary criticism a kind of approach that may be taken to many and perhaps all kinds of verbal expression? In that case, how could the foundations of such an approach be established without question-begging distinctions between reason and taste? Like “aesthetics,” “literary criticism” has no precise counterpart in ancient Greek or Latin, and so we must be careful when we read our texts not to assume that “the literary” was recognized as a distinct and autonomous area of human activity. Indeed, the term “literature” itself must be used in quotation marks since,pragmatically, the Greeks tended to think mainly of poetry when they thought of high verbal art and to discuss the issues that concern us primarily in relation to poems. If we are willing, however, to bracket these problems of definition, attending to certain passages debating how to critique a poetic text is a promising area in which to discover whether the ancients were willing to recognize a value in works of art that was not reducible to political, ethical, or other non-poetic values.
Two short but dense texts center this study. The first is a passage from Aristotle’s Poetics that is remarkable as the only explicit statement in ancient criticism that poems are to be judged by the laws of poetry and not by those of another art. The passage is hardly unknown, but tends to be ignored or interpreted weakly. I will offer a strong reading and explain how it can be accommodated to Aristotle’s other stated views about poetry. The second text is meant to suggest that Aristotle’s was not an isolated position in the critical tradition. It is a scene from Aristophanes’ Frogs, produced in 405 BC. In the play’s parodic poetry contest, Euripides comes very near to articulating an aesthetic defense of his art along lines very like Aristotle’s. Though the idea is quickly passed by, the resemblance to Aristotle suggests that his was not an idiosyncratic view and that a certain autonomy for works of art was already imaginable in the fifth century. It must be conceded that claims for poetic autonomy are not often sounded in the Greco-Roman critical tradition, and I shall offer reasons for why this is so.
First, it will be necessary to sketch the intellectual background to these texts, in particular to highlight a new conception of “art” (tekhnē) that rose to prominence over the fifth century BC. The key point here will not be – as it was for Kristeller and others holding that the ancients lacked a distinct conception of the aesthetic – an alleged failure on the part of the Greeks to distinguish “fine art” from “craft” (both of which could be designated by tekhnē); rather it will be the conceptual depth given by fifth-century speculative thought to the notion of tekhnē as productive activity which proceeds in accordance with specifiable principles toward a definite end. In this perspective, an artisan who would produce a work worth commending as “fine” (kalos) was bound to observe the principles of his or her tekhnē and was culpable for infringing them. At the same time, however, the proliferation of highly developed and specialized arts/crafts in Classical Athens made it obvious that each one had aims and methods of its own; this could suggest that it made little sense to hold an artisan accountable to principles belonging to one art when she or he was practicing another. In this way, the Classical period set the stage for a notion of poetic autonomy in its root sense, one that insisted on poets’ respecting the laws (nomoi) of their own (aut-) discipline, but that limited their exposure to demands of other arts/crafts. Aristotle’s strong assertion of poetic autonomy amounted to granting poets immunity from prosecution for violating laws that come from outside poetics, most notably the art of politics.
Large-scale claims have been made to explain why a separate category of the aesthetic was somehow, intellectually or culturally, impossible to conceive before Baumgarten and the emergence of the fine arts in the eighteenth century. Such theses are worryingly overgeneral, and yet Ancient Literary Criticism, the excellent anthology by Donald Russell and Michael Winterbottom (1972 ), does show that Greek and Roman writers give short shrift to aesthetic considerations when they critique their poetic traditions. From the earliest discursive remarks on Greek poetry, surfacing in the sixth century BC, through the voluminous rhetorical and pedagogical works of the imperial age, the preponderant concern in ancient discussions of literature is one they would have called ethical and we would call political: the one question most often asked of works of verbal art – and often the decisive question in assessing their worth – was whether they were likely to make a person hearing or reading them a good citizen: did they help, or at least not hinder, their audiences’ becoming and being just, temperate, pious, and wise in their dealings with others and with the gods? The prevailing ethical tendency in ancient criticism was complicated (and made more interesting) by the early realization that poetry afforded a powerful kind of pleasure that, in itself, need have nothing to do with forming character: its ability to “enchant” (thelgein) the soul is taken for granted already in Homer and is a commonplace thereafter. This hedonistic strand, however, is easily reconnected with ethical concerns: already in Homer and Hesiod the quasi-magical power of poetry to delight is associated with social goods – with distracting minds from debilitating distress or ensuring that the glorious deeds of ancestors are preserved for later generations to emulate (cf. Halliwell 2011).
At first sight, it may have seemed a slight modification in traditional views when poetry was seen as one among many productive arts in the fifth century BC; but the conception of tekhnē itself was shifting at the time and this opened the door to evaluating works of verbal art not against the usual moral and social norms but with an eye to their intrinsic properties, such as linguistic correctness or formal perfection. A prime stimulus in this development was the powerful new model of art provided by Hippocratic medicine: rejecting supernatural and occult causes, the healing art could boast that it improved human life through the methodical application of unaided reason to natural phenomena. This technical, rationalist, and progressive idea of art was carried over into the study of language and its products by sophists like Protagoras, Prodicus, and Gorgias and by anthropologists/natural scientists like Democritus. Their fundamental contribution was to insist that poetry was a tekhnē as they understood “art,” a productive enterprise that could be analyzed to discover the way of proceeding that would best ensure success (Ford 2002, 161–175). Although affinities between the making of poetry and artisanal production can be traced very far back in Greek and pre-Greek literature (Nagy 1990), it was only in the fifth century that poetry was spoken of as a tekhnē in the sense of a methodical “art” rather than a “craft” that might depend on an inborn knack or inspired ability (Ford 2002, ch. 6). Many of those professing the “modern” approach to poetry were concerned with rhetoric, and their focus on the role of language in persuasion brought poetic linguistic devices more into view. They tended to regard poetry simply as speech (logos) with meter added as an enhancement (meter being the most obvious of the liberties in expression poetry was traditionally allowed). This suggested that its special hold over its audiences could be traced to intrinsic, formal properties. Once poetry was studied in this way, one could pronounce upon the “correctness” (orthotēs, related to “orthodontist”) of a poem in technical, linguistic terms without reference to its truth or social utility. An aspect of such enlightened fifth-century inquiries into poetry, then, was to raise the possibility at least of evaluating it on its own terms.
In an important if limited way, these studies created a space for a non-political, non-moralizing form of poetic criticism: if poetry is viewed as language elaborated to impress an audience, an expert has grounds for confining his appraisal of a work to the question of whether it is technically “correctly” (orthōs) composed, whether its message had been designed and executed in the way most likely to win acceptance by the audience addressed. The question of whether its message harmed or helped the city could be, with all due allowance for its importance, set aside as a separate question belonging outside the specific field of inquiry; in much the same way, the art of rhetoric established its technical status without committing itself to particular ethical or political positions. In this way and to this extent, fifth-century science and rhetoric opened a path to aesthetic appreciation of poetry by separating the knowledge of how something was said from what was said; the implication was that it was possible in some important sense to do the former “finely” (kalōs), whether or not the latter was thought fine. A technical assessment of poetic language was not the only way to evaluate poetry, of course, but its advocates placed it on a very high level: “The most important part of education,” Protagoras asserts in Plato’s dialogue of that name, “is being exceptionally clever about poetic language [peri epōn deinos]; that is, to be able to perceive what in poetry is said correctly [orthōs] and what not, to know how to distinguish between the two and be able to give an account of oneself when questioned” (Prot. 339A).
For cultural conservatives, these new insights into the power of poetic language made it only more imperative to scrutinize poems’ moral and social effects. As Plato’s Gorgias dramatizes, even those who could establish themselves as technical experts in using language effectively could not claim exemption from society’s demand that their science be exercised for the benefit of the community. There was, in short, a tension between the way scientific critics might use the word kalōs to praise formally perfect works of art as “finely done” and the meaning this flexible word had in popular criticism, which was ever ready to take extrinsic values into account and to connect the attractiveness of satisfying form to moral or social fineness. Ultimately, this is the tension between the dulce and the utile as standards for judging poetry, and Aristotle’s Poetics stands out in this tradition for articulating the enlightened point of view that the artisanal status of poets entitles them, in some contexts, to immunity from political and moralizing critiques.
Aristotle’s Poetics addresses not only the nature of poetry, its genres and their powers, but, as he says at the beginning of the work, “a number of other topics that belong to this field of inquiry” (Ch. 1, 1447a11–12).2 One of these related topics is discussed in the 25th of its 26 extant chapters, entitled “On Problems and Solutions.” The terms refer to a popular poetry game in which one player posed a challenge (proballesthai) to a piece of poetry, a “problem” that another player had to solve (lusis), usually by proposing an interpretation on which the cited passage could be defended as “correct” (orthōs; Ch. 25, 1460b24; cf. 1460b13). Chapter 25 proposes to “theorize” (1460b7) the rules for such a game, and is introduced as follows (1460b6–21, my paragraphing):
On the subject of challenges [problēmata] that may be put to poetry and how to meet them [luseis], one can discern [theōrein] their number and nature by reflecting in the following way. Since the poet is a representer of things [mimētēs], just like a painter or other image maker, he will necessarily in any given case represent one of three kinds of things: things such as they were or are now; or what people say and suppose [was or is]; or what should be.
And the language in which these things are propounded admits archaisms, metaphors and other deviations from normal speech; but we grant this to poets.
A further consideration is that correctness [orthotēs] in the art of poetry is not the same as in the art of politics, nor in any other art. In terms of the art of poetry, there are two kinds of mistake [hamartia], a mistake in terms of the art or an incidental mistake. For if the poet [chose to represent a horse correctly, but erred through] incompetence, the error belongs to the art itself. But if there is something incorrect about the object chosen – for example, a running horse with both right legs thrust forward at the same time – the error pertains to a specific art, such as medicine or whatever you wish; it is not a mistake that belongs to the art of poetry itself.
The sentence I have italicized is the most explicit claim for poetic autonomy in antiquity. The question is how far Aristotle thought this immunity to criticism extended, for to distinguish the principles of poetic composition from those of politics or medicine seems to carve out a very large sphere of autonomy: one can readily see that a doctor’s complaint, for example, that Homer gives the wrong recipe for restorative beverages would be regarded as a quibble (see Plato, Rep. E-406A for a criticism of Il. 11.630 in such terms); but Aristotle also mentions the art of politics, which in his view subsumed ethics; hence to say that problems raised against poetry by experts in politikē may be dismissed as “incidental” to the art is considerable autonomy indeed.
In the nineteenth century this passage was taken as an unqualified endorsement of aesthetic autonomy:
Aristotle asserts that poetry is to be measured not by a moral but by a purely aesthetic standard. All important in his eyes is the perfection of the imitation, the shaping activity of the artist. Poetry must be judged by its own laws, its own basal assumptions, and errors that are errors only according to some alien standard are faults [incidentally], and accordingly excusable.
(Carroll 1895, 18)
A few modern commentators have agreed that “the independence of poetic orthotēs [correctness] suggests the existence of purely aesthetic values” (Lucas 1968, 235), but most recent Aristotelians would say this goes much too far. A main reason is that the statement is at odds with what seems to be Aristotle’s ethical orientation in the rest of the Poetics (to say nothing of the assumptions of ancient literary criticism as a whole). After all, it has been urged, ethics is the stuff from which tragic plots are made – with good or bad characters meeting success or disaster according to their moral choices (6.1449b36–50a3). Indeed, the very emotions that the tragic art is designed to arouse, pity and fear, are conceived by Aristotle as having a cognitive basis: they arise from a judgment that a particular situation is worthy of pity or fear (13.1453a3–7). As a result, a poet will have to take ethical matters into account to make a play that will engage his audience’s emotions. As Amelie Rorty puts it: “Although Aristotle focuses on the formal elements of tragedy – on the best way to structure plots, his is not an aesthetic theory. The pleasures and insights of tragedy do not rest solely or primarily in their purely formal properties, in the elegance and structural tension of balance” (Rorty 1992, 2; cf. Nightingale 2006, 40).
Those who hold further that Aristotle conceived that the pleasure of poetry is one of moral learning are particularly prone to resist any suggestion that poetic criticism might be insulated from political or ethical concerns.3 Among these, Stephen Halliwell and Malcolm Heath have most directly confronted Ch. 25 and tried to minimize its implications by stressing the qualifications Aristotle goes on to make. On Halliwell’s reading, Aristotle says not that “correct standards for poetry are not the same as those in politics or in other arts” but only offers “the proposition of non-identity of the values in question” (1989, 339–341; cf. 2011, 210–222). Heath argues similarly that in saying “correctness is not the same thing in politics and poetry” Aristotle “does not mean that they are unconnected but that they are not co-extensive” (Heath 2009, 469). These attempts to find some overlap between poetic correctness and ethical values seem strained as paraphrases, and in any case run up against the clear trend of Aristotle’s argument in the chapter which, as will be seen, never curtails the status of poetic principles as the ultimate determiners of poetic correctness. Aristotle does indeed allow, implicitly here and explicitly in the Politics and Ethics, that outside the game of “problems” poets may have to bow to the authority of other arts and yield to their recommendations. The key consideration, however, is not some ineliminable link between poetry and politics or ethics but the contexts in which critical discussion takes place. What Poetics 25 declares unswervingly is that, when adults discuss poems and debate their merits among themselves, ethics and politics will have to take a back seat to poetics.
Let us take up, then, Aristotle’s argument as it develops in what all agree is a difficult and corrupt text at points. In the quoted passage, Aristotle sets out his method: to distinguish three kinds of challenge that may be made to a poem and to suggest, invoking principles developed earlier in the treatise, a line of defense for each. One kind of objection concerns what the poet says, the “objects of imitation”; but Ch. 2 had established the principle that poets can represent people who are better, worse, or like us (1448a1–6); the payoff in Ch. 25 is that, if it is objected that a poet has represented something impossible, an Aristotelian can reply that poetic representations need not be realistic but may portray the way things were, ought to be, or even are thought to be. We need not delay over the second kind of “problem,” those concerning language or “the medium” in which poets exercise the imitative art, as Ch. 1 explained. Defusing objections to poetic expressions will take up the most space in Ch. 25 (1461a9–1461b9), but does not involve aesthetic issues since Aristotle roots this license in convention without any appeal to the imperatives of the poetic art: “for we grant these things to the poets” (1460b13). It suffices to observe that grammar police must not harass poets if they do what their art tells them to do – “season” their mode of expression to make it more pleasant (1449b25, 1460b22 with Janko 1987). The third principle (italicized above) is the one that will mainly concern us. This is to distinguish between an error (hamartēma) in a passage of poetry that violates one of the principles of the poetic art and errors in terms of some other field of knowledge. In the game of problems, Aristotle rules that errors in respect to other arts are only incidental, while an error with respect to the poetic art may be pronounced “incorrect,” not orthōs (1460b23).
This important distinction is illustrated with examples from the visual arts. In a passage that is corrupt but tolerably clear, Aristotle says that if a painter or sculptor portrayed a running horse with both right legs thrust forward (which he wrongly assumed was impossible), this would be technically a mistake to an expert in the art of equitation but only an incidental error in terms of the arts of painting and sculpting, whose ends, presumably, are producing convincing representations (1460b16–21, cf. Janko 2011, 522). A little later he gives a second example from the visual arts – painting a female deer with horns (1460b30–32): as the author of the Parts of Animals, Aristotle sees a technical error here, but says it is less of an error than painting one that cannot be recognized (Halliwell 1989, 333).
These examples suggest that the each of the mimetic arts works according to rules inherent in the form and these are not to be overridden by demands arising from other areas of expertise. This position would seem to commit Aristotle to entertaining arguments that, in the poetic arts, distortions of reality or even untruths can be defended if they communicate the poem’s effect more fully or more powerfully. He goes a considerable distance in this direction when he turns back from painting to poetry to spell out what counts as an error with respect to the art itself (1460a22–61a9). Aristotle says that, against charges that what a poet has represented is impossible, one can reply that the impossibility is “correct” if it contributes to the telos of the work, its purpose or “end” (1460b22–28). The important role of a poem’s end or function (telos or ergon) is implicit from the first sentence of the Poetics: Aristotle says one must define each particular kind of poetry (each genre) to determine the “potential” or “capacity” (dunamis) it has, for only if we know this can we decide whether a given poem is such as can be called “fine” (kalōs ekhei, 1.1447a10). For example, the art of tragedy aims to produce “a sort of katharsis of pity and fear” in the audience, and this end entails that “the tragedy which is most excellent [kallistos] in respect to the art” will show a change from good fortune to bad (1453a22–23). Tragic poets who compose plots in which the good end happily and the evil suffer may win contests, but their works forfeit endorsement as kalos because the kind of pleasure (hēdonē, 1453a36) they afford more properly belongs to a different genre, comedy, which has different ends (1453a30–39).
In Ch. 25, Aristotle illustrates the principle with an example from epic, Achilles’ solo pursuit of Hector in the Iliad in which the hero restrains the Greek warriors from joining in the pursuit just with a nod (Il. 22.205–206). This scene had already been noticed in Ch. 24 as implausible but effective; Aristotle finds it commendable because it contributes to the epic’s telos by increasing the audience’s sense of the marvelous (to thaumaston); the marvelous is by definition pleasurable for Aristotle, who remarks that it is especially at home in epic where the recitative mode can make an absurdity less obtrusive than it would be if seen on the dramatic stage (24.1460a11–17). In Ch. 25, Achilles’ pursuit of Hector shows that, in quest of providing or augmenting a pleasure appropriate to his genre, the poet may violate probability.
The distinction between incidental and technical errors will map out a rather large area of autonomy; and one could characterize this as aesthetic autonomy to the extent that the “end” of the mimetic arts is not to convey facts past or present or to provide models of ethical conduct but to produce those emotional responses the art in question is designed to elicit. To be sure, the poet’s autonomy is not absolute, and Aristotle goes on to put limits on it. The poet does not have license to flout notions of the possible for no reason: ideally, he says, one ought to make no mistakes, and so a poet can be justly blamed for importing an impossibility if it was possible for the poetic end to be achieved no less strongly without the mistake in the particular art (25.1460b26–28). There is no reason, however, to infer from this some obligation on the poet’s part to remain attached to the real world (e.g. Halliwell 2000, 110), for this limitation on the poet’s autonomy does not come from the authority of any other art but from the requirements of the poetic art itself: the art determines the end of each genre, and everything in the poem should further that end, “either as contributing to the end in itself or by making another part of the work more striking” (1460b25–26). The poet’s obligation to his own art, which directs him to produce certain emotions in the audience, trumps demands that he be realistic, but imposes on him the obligation not to insert gratuitous irrationalities – not because they are irrational but because they are gratuitous, that is, they dilute the force of the work. (Cf. 1462a18–62b13 on the superiority of tragedy to epic because tragic unity is “not watered down” with irrelevant episodes.)
Having won for poets permission to portray impossibilities, Aristotle then defends them against charges that what they say is untrue (1460b32ff.). Reverting to his observations opening the chapter, he says sometimes poets can be exculpated as exercising their right to portray how things ought to be or how things were (1461a1–4). An additional possibility is to say the poet has represented things the way people say they are (1460b35). Here Aristotle instances matters dealing with the gods: if a poet be faulted for following popular belief and portraying jealous and malicious gods, for example, the reply Aristotle recommends is, “That’s what people say.” This casual waving away of charges of moral impropriety must seem inadequate to those who expect the Poetics to dismantle Plato’s censoring of divine immorality in Homer and Hesiod in the Republic; but Aristotle’s only concession to this tradition, which can be traced back to Xenophanes (cf. 1461a1), is to allow that “it might be better that some such stories were told another way.” But he wraps up his point simply by repeating what he considers a sufficient reply to such charges – “that’s what people say” (1460b36–61a1). It is more important that a poem engage the beliefs and even prejudices of its audience than that it be philosophically or theologically correct.
Just as poets are not bound to confine representations of the gods to the morally commendable, they need not limit representations of human action to the ethically admirable. In meeting ethical objections to what characters do or say, Aristotle urges considering not only whether the thing is contemptible in itself, but also the agent, the context, and the agent’s purposes; one would therefore not blame, for example, an evil done “so that a greater good may occur or a greater evil avoided.” There is, to be sure, a limit on justifying representations of unethical behavior, as in the case of absurdities: “Criticism of irrationality and depravity is right when they are unnecessary and they serve no purpose” (1460b26–28, tr. Halliwell), Aristotle says, pointing to Menelaus in Euripides’ Orestes as an example of gratuitous wickedness (1461b19–21; cf. 15: 1454a28–29). Once again, Aristotle’s qualification leads some to infer that, even if ethical and poetic correctness are not identical, ethical judgments ought to inform artistic ones (e.g. Halliwell 1989, 342; Heath 2009, 469). It is certainly true, as said above, that an audience’s ethical concerns are not put to sleep when watching a play; but Aristotle does not espouse a view that art must be as far as possible morally uplifting, any more than he had said art ought to strive for realism. Again, the telos rules: because the art of poetry dictates that the “finest” form in each genre must be that which best conduces to its end, gratuitous evil is objectionable not because it is evil but because it is gratuitous, that is to say, the evil neither contributes to the end in itself or makes another part of the work more striking. When irrationality or depravity are gratuitously introduced, a valid objection may be raised, but it will be based on the art of poetry, not on ethics or politics. It says, in effect, that the Orestes would work just as well if Menelaus were a benign character; one could keep the plot and have the essential emotional effect. James Hutton (1982, 18) comments well: “the characters are required to be ‘good’ not for the sake of affording models of conduct, but because the fortunes and misfortunes of vicious characters would not elicit from the normal spectator or reader the response proper to this kind of poetry.” For this reason Aristotle had previously (in 13.1454a10–13) commended early poets for discovering, even if by chance rather than art, that only a limited number of families supplied suitable myths for tragic plots, for example the house of Laius or of Atreus with their generations of parricide, filicide, infidelity, and cannibalism. He thus recommends stories of houses beset with evil doings and sufferings as what the art demands, as making for “the finest tragedies in artistic terms” (13.1453a12–23).
At this point Aristotle concludes his defenses of the autonomy of poetry based on his distinction between objections pertaining to the art of poetry and those of other arts.
He then takes up a series of linguistic solutions, which, as I have noted, is less germane to our inquiry. One may remark, however, that the prominence of linguistic issues in Ch. 25 supports the connection argued for earlier between the sophistic study of poetic language and assertions of poetic autonomy. In this light, Poetics Ch. 25 appears as an attempt to give a theoretical basis to the possibility, opened up by fifth-century specialists in poetic language, of judging verbal art in technical rather than political or moral terms; because poetry is an art, its products can claim an essential autonomy, a right of appeal to “the art of poetry and not another art.”
Aristotle’s defense of poetic autonomy, however, does not entail that poetry is completely independent of politics and ethics. To say that poetry, like any art, has its own end (telos) is not to say that poetry is autotelic in the sense that it is an end in itself. No ancient ever declared that art was “radically cut off from the domains of morals, utility, religion and the like” (Porter 2010 , 166; cf. Shiner 2009, 161–163); an assertion of absolute freedom was bound to sound odd to the Greeks, and autonomy in this sense has all the earmarks of a Romantic aspiration. Aristotle is quite clear in his Politics (Bks. 7–8) that the art of poetry should be ruled by that of politics in some contexts, such as fixing school curricula or regulating public entertainments. The reasons are spelled out in his Nicomachean Ethics and its notion of “architectonic arts”: every art aims at some good, an end for the sake of which everything in the art is done; but some ends call for combining several arts, and in these cases some artists will fall under the command of others according to their contribution to the ultimate end. So when an architect tells carpenters, sculptors, and stonemasons what to make, the individual artisans have autonomy in the techniques they choose to apply to their materials, but the arkhitektōn, or “master-builder,” prescribes the end of their art (the house or the theater). In the theater, the poet is the “architect” since he makes the plot, the “soul of tragedy” for the sake of which other artists, such as scene painters (Poet. 1450b17–20, 1453b1–11) or actors (1404a23–24), exercise their arts. (The principle obtains even though Aristotle was realistic enough to know that, in practice, star actors pushed poets around: 1451b33–52a1 and Rhet. 3.12 1413b9–12.) But in the city, the poet’s art may be guided and controlled by the art of politics, the architectonic art par excellence that knows “the good for man” (NE 1.2). The principle that correctness in the art of poetry is not the same as in the art of politics thus has a flip side, which is that political “correctness” will prevail in public uses of poetry, including censorship in some contexts. (In Plato, too, the statesman has the right to censor imitative artists because his art is superior to theirs: Rep. 342c; cf. Euthyd. 291c.)
Because poetry is a social activity and has come to exist in cities, the poetic art may be made subordinate to political laws. Politics is entitled to determine which activities are to be permitted in the polis and with what limitations in order to promote the happiness of its citizens by equipping them to engage, insofar as they are able, in excellent activities (Pol. 1.2; 7.13, esp. 1332a7–21; NE 1.2.4–6). Some would therefore conclude that literary aesthetics is impossible, for it is “inherent in the nature of politics as Aristotle conceives it that there could not be a human activity that is not answerable to politics” (Heath 2009, 470 with n.3). Yet the principle of the poet’s autonomy remains uncompromised in an essential way: it is important to note that the poet is subject to the statesman because of the architectonic supremacy of politics, not because the poetic art is fundamentally ethical or political in nature. One must never confound the principles governing the poetic art with those of its architectonic superior. Such authorities can in theory force the poet to make bad poetry (e.g. by insisting that tragedies have happy endings), but what they cannot do is make bad poetry good; expert though they be in politics, they are powerless to change the rules of the poetic art.
Ultimately, the autonomy of the poet derives from Aristotle’s conception, inherited from the fifth century, that poetry is an art among others and, like them, has its own ends and methods. This means that, while poetic autonomy outlines a sphere in which laws we may call aesthetic operate, the principle itself is not rooted in aesthetics, since it applies to other arts equally. Aristotle’s defense of autonomy does not rely on positing a distinct realm of aesthetic experience; what he insists on is the distinctness of the arts: a doctor’s practice can be regulated by political experts, but they cannot redefine the end of his art nor the best methods for pursuing it.
It is therefore unnecessary weaken the proclamation of poetic autonomy in Poetics Ch. 25 so as to reconcile it with Aristotle’s conception of hierarchically nested arts and with the dominant ethical/political strand of ancient criticism. Conflicts among these agendas disappear if we consider the contexts in which this autonomy will be honored. The game of “problems” envisioned in Ch. 25 will not be played out in public spaces – Aristotle is clear that the principles of the art of poetry will not necessarily prevail in the theater and distinguishes between evaluating poetry “in terms of the art itself and in respect to its audiences” (1449a9). While audiences in their “weakness” (1453a34) may acclaim technically inferior works, and while political scientists may deplore or seek to control poetry’s role in the city, knowledgeable adults may gather and talk about poetry in terms of the art itself. The problem game was a long-lived social entertainment with roots reaching back to symposia of the intellectual elite (Pfeiffer 1968, 69–71) and was a way for groups to negotiate a sense of “correctness” while observing “exacting rules of courtesy and self-control” (Johnson 2010, 12). Needless to say, even as putatively wise poets were being put to the test for correctness, it was the performers who were being evaluated for being “good at give and take” (epidexioi: Rhet. 2.1, 1381a29–35); they put their literary culture in play to demonstrate their forthright freedom of speech (parrhesia) and willingness to challenge authorities and to show ultimately their restraint and willingness to submit to norms that they agree to impose on themselves. As sympotic order was an imitation of civic order, the “correctness” of poetry in such contexts was an imitation, a copy with a difference, of “correctness” in the polis (Ford 2009).
Although the 25th chapter of Poetics is usually under-interpreted, it must be acknowledged as giving a strong defense of the poet’s autonomy, one that detaches the “end” of the poetic art from reinforcing ethical or social values. But a problem remains as to why this view seems so isolated in the critical tradition. The picture becomes more filled in if we turn from Aristotle to Aristophanes and show that the philosopher was not expressing an idiosyncratic position of his own but deepening a line of thinking about poetic judging that had already been on display in Athens a generation earlier.
Aristophanes’ Frogs is a comic masterpiece providing a precious glimpse, albeit through a fun-house mirror, of self-consciously sophisticated literary criticism in the Classical age. The play devotes its second half to a duel between poets, as a traditionalist Aeschylus squares off against a modernist Euripides. For their critical pronouncements Aristophanes drew heavily on the high-flown discourses of Protagoras, Prodicus, and Gorgias – in other words, on classical criticism at the highest level (Segal 1970 ). Along with this perspective came the supposition that poetry is an art: Euripides is Aeschylus’ rival in the art (antitekhnos, 816); it is in artistry that he claims to be superior (831). Their contest as a whole can be seen as a reductio ad absurdum of the “modern” proposition that technical correctness in poetry is something measurable (Ford 2002, 152–157). In the debate’s scrutiny of poetic language (1119–1413), Aeschylus and Euripides use such terms as “error” (hamartanein 1135, 1137, 1147) and “correctness” (orthotēs 1181, 1131) to attack each other’s verse, and when Euripides charges that there are 20 errors (hamartiai) in the first three verses of Aeschylus’ Choephoroi (1131), it recalls Aristotle’s report of Protagoras, who faulted Homer for making two grammatical blunders in the first line of the Iliad alone (Poet. 19.1456b15–19).
Alongside these technical sallies is a great deal of discussion of the poets’ piety, morality, and politics; indeed the ultimate decision between them is made on the basis of “which one gives better advice to the city” (1421–1422). As a whole, then, Frogs may be said to reflect the norm of the ancient critical tradition, confounding poetry’s formal qualities with moral and social ones and giving the latter the trump hand. Early on in the contest the disputants agree that the purpose of poetry is to improve the audience. Asked why poets deserve admiration, Euripides volunteers, “For sophistication and wise advice [nouthesia]; and because we make men better in cities” (1009–1010). One might conclude that any aesthetic defense of poetry is thereby written off (Dover 1993, 318), but it may be unwise to single out any critical pronouncement as central to the play. In any case, we can recall the model of Protagoras again to see that to agree that poetry should make people better need not entail rejecting critical sophistication about language. Plato’s Protagoras prized “correct usage” (orthoepeia) as “the most important part of education” and yet assumed that poets should provide wise advice (nouthetēsis: Prot. 325E–326A). Similarly in Frogs Euripides could agree with the idea, likely to have been widely voiced in Athens, that poets shape the character of citizens without conceding much: everything depended on how one defined goodness (Ford 2002, 197–201), and the play makes very clear that Euripides’ idea of a well formed citizen was the very opposite of the fire-breathing warriors Aeschylus claims to have nurtured: sophisticated, talkative, sensitive, and unathletic, the pupil formed by Euripidean poetry looked a lot like an aesthete (cf. 945–67, 1069–1073). His politic concession, then, should not make us miss two rather aestheticizing moments in his self-defense.
The play’s carnivalesque freedom makes it vain to seek in either poet a consistent critical position, but Aristophanes does draw a comic contrast between Aeschylus, standing on the whole for traditional poetry and traditional virtues like courage, and Euripides, championing modernism and the enlightened critique of convention (Hunter 2009, ch. 1). Accordingly, it is Euripides who focuses on formal elements in attacking Aeschylus, such as his diction (phrasis) or his prologues (1119ff.). In this light it seems significant that, when the two come to blows over specific passages, it is Euripides who comes near to claiming a poet’s autonomy.
After a general exchange of insults the contestants get down to cases when Aeschylus adduces his fire-breathing Seven Against Thebes as an example of how he made the Athenians better. Calling it “full of Ares” (in a tag quoted from the sophist Gorgias: 1021), he boasts that the play made the Athenians burn to fight for their country. Dionysus interjects that this was a shame, since the play also emboldened the Thebans (Athens’ formidable opponents during the Peloponnesian war: 1023–1024); this is only a wisecrack, but it points to a vulnerability in virtue-oriented criticism: one can readily agree that the Seven arouses patriotic fervor more than other plays, but that is a separate question from whether it is good for the city.
Aeschylus then contrasts his own “profiles in courage” (Jeffrey Henderson’s nice rendering of 1040) with the “whores” Euripides has brought on stage: “By god I never created Phaedra’s and Sthenoboea’s, harlots, or ever portrayed a woman in a state of passion” (1043–1044). With a hint of prudishness, Aeschylus deplores two Euripidean characters notorious for illicit lust, Phaedra (for her stepson Hippolytus) and Sthenoboea (for Bellerophon, a guest in her husband’s house); both women were rejected and accused their beloveds of rape with disastrous consequences. Euripides’ reply to this charge is suggestive: “And what harm, you villain, did my Sthenoboea’s do to the city?” (1049). Granting the premise that poets should make citizens virtuous, Euripides denies that representing morally depraved behavior must, in itself, corrupt an audience. This passing defense (reiterated in substance at 1064) is noteworthy for agreeing exactly with Aristotle: the philosopher would put it that portraying a heroine behaving badly is not per se an error in poetry, but may be correct if, in context, it contributes to the “end” of the play. Euripides, then, claims a certain autonomy for the poet, but without offering an aesthetic defense; to claim that there is value in aesthetic experience he would have had to go further and argue that watching a Sthenoboea might make men better citizens. But this is comedy and Aristophanes only lingers on an issue long enough to get a laugh and set up the next one. Nonetheless, the naive moralizing assumption that the behavior of a play’s characters is carried directly over into the audience’s lives is challenged, and the absurdity of Aeschylus’ answer – “Your Bellerophon’s made the respectable wives of respectable men commit suicide” (1050–1051) – is perhaps a hint at the misguidedness of his critique.
The implication of Aeschylus’ charge is that these lustful, suicidal heroines can lead to “copy cat” behavior, to which Euripides offers a different, equally suggestive defense: “But the story I told about Phaedra was already established, wasn’t it?” (1052, tr. Henderson). In his previous response Euripides denied that representations of depraved behavior are blameworthy per se; he now entertains the premise that it would be harmful if people imitated (in its root sense of miming) such behavior, but replies that he didn’t make up the story. If he’d made up the Phaedra myth, he implies, he might be culpable, but poets have license to avail themselves of the traditional stories. Once again, Euripides defends himself along lines recommended in Poetics 25: Aristotle would agree that, confronted with Phaedra as a problematic character, the obvious defense a poet might make is to say, “that’s the way they tell the story.”
To this, Aeschylus’ retort is that the story may well be traditional, but one must consider its effects in shaping civic behavior. He closes the exchange by assuming the demeanor of a schoolmaster: the poet should conceal what is wicked, he says, not stage it; for what teachers are for the young, poets are for adults (1054–1056). It seems that Aeschylus’ perspective is endorsed at the play’s end when he wins the contest because the city needs the kind of men his plays produce. At the same time, there is more than a little push-back against this view, both in the Euripidean remarks underscored above and in the general portrayal of Aeschylus as a puritanical scold.
It is easy to recognize that Aeschylus adumbrates a pedagogical position that is given philosophical depth in the Republic when Plato argues that poetry in schools should be censored because of its power to shape moral character in the young (380A–C, 390A–B). It is less easy to see, but no less true, I hold, that the position assumed by Euripides is the same one that is argued comprehensively and rigorously by Aristotle in Poetics 25. The Frogs, produced for a mass audience, suggests that fifth-century technical criticism had already begun to formulate defenses poets might make against schoolmasters and other institutionalized “experts” in poetry who would interfere with their exercise of their craft. The question therefore arises, why do almost all the texts we classify as literary criticism assume a moral-political point of view? They ask, with Aristophanes’ Aeschylus, whether it is good or bad for people to hear such things as the poets say, and this is because, again like Aeschylus, they tend to think of audiences as children. It is not only Book 3 of the Republic that worries over how young citizens should be taught poetry, but also the Protagoras, whose high-level critics perform before an audience of young men and who interpret Simonides’ poem on virtue (aretē) by expecting it to conform to conventional morality. A good many of the familiar classics of criticism, from Horace’s Ars Poetica to Macrobius’ Saturnalia, are written with an eye toward young men as consumers of poetry – often enough the actual sons of the writer and the addressee. The common agenda is, to quote a title of Plutarch’s, “How a Young Person Should Listen to Poetry,” a treatise written with a view to the education of Plutarch’s son and the son of Marcus Sedatius. Plutarch explains in his preface that the impetuous nature (phusis) of the young (the neos) makes them enjoy poetry more readily than philosophic discourses (14e–15c). The treatise accordingly proposes to make reading poetry propaideutic to philosophy (15f) by showing how anything offensive in poetic portrayals of gods or human virtue may be reinterpreted or explained away (16d). Ultimately, Plutarch is leaning on Plato here, and thinks of poetry in terms of how it will affect what Plato called their “tender and highly impressionable young souls” (Republic 377E, cited at “How a Young Man” 3e–f).
The pedagogical approach to literature is abundantly preserved in the record because it was so useful to teachers, those great consumers and producers of books about literature. Works like “How a Young Person” (or Republic 3 for that matter) furnished teachers with methods and examples for combining erudition in poetry with an ethical and political stance that would appeal to the fathers who were paying tuition. In this tradition, Poetics Ch. 25 is exceptional in being for grownups. It is addressed to a select audience whose demands of poetry are quite different from what prevails in the classroom, or in theater; its audience is comparable to the select company the tragic poet Agathon invites to a symposium to celebrate his theatrical victory privately (Symp. 174A). In such a group, withdrawn from active engagement with civic duties and public business, one may apply a mutually agreed on conception of “correctness” as one discusses poetry. Here something like an aesthetic dimension of literature can be respected if one is willing to insist, as Aristotle did, that what is dictated by the art of poetry may trump objections raised by ethics or politics.
In a nutshell, the difference between Aristotle and the tradition exemplified in Plutarch is between an interest in what is technically correct in poetry (what the principles of art suggest will most effectively realize its nature and give it the best chance of succeeding) and an interest in correcting statements that offend against morals lest they corrupt the young. Aristophanes confirms that, despite the pedagogical slant of so many of our critical texts, the proposition that poetry is to be judged according to its own standards rather than by those of another art was far from unthinkable and was not recognized by academic philosophers alone.4
This essay has argued that readers of the Poetics should recognize its strong assertion of the poet’s autonomy in Ch. 25, even if that autonomy was limited: it was not attached to a fully conceptualized literary aesthetics and did not offer any basis for resisting political interference with the use of poetry in civic life. Despite these limitations, Aristotle’s position suggests ways to rethink the possibility of literary aesthetics in the face of certain anti-aesthetic axioms that have taken root since the demise of deconstruction. For many academic critics today, aesthetics stands convicted as a mystified concept whose only use is to lay claim to a privileged social position; in its place, the discussion of literature in professional journals is dominated by political criticism (which claims to know the good for a community), with some help from ethical approaches (which claim to know the good for man). The academic debate taken as the point of departure for this chapter, whether there is such a thing as ancient aesthetics, is from this point of view an historicizing way of asking whether these methods have a right to exhaust the discussion of literature.
Poetics 25 first reminds us that political critiques of poetry are talking about something else than poetry. On a certain view of politics, there is nothing else to talk about; but Ch. 25 urges giving poets a hearing as poets by concentrating on the art of poetry. The exiled dissident artist Gao Xingjian would agree: “To demand that writers act as the conscience of the society can only strangle literature” (2012, 11). Having had first-hand experience with that political expert Mao Zedong, Gao insists that, “Once literature is contrived as the hymn of the nation, the flag of the race, the mouthpiece of a political party or the voice of a class or a group, it can be employed as a mighty and all-engulfing tool of propaganda. However, such literature loses what is inherent in literature, ceases to be literature, and becomes a substitute for power and profit” (2000).
In addition, political critics who are aware of how political power can insinuate its ideology as the correctness to which all must conform may reflect that, if literary studies are to be an intervention in politics, it would be a strategic error to abandon claims for artistic autonomy. If it is thought that literature does its greatest service when it speaks truth to power, it can only do so if it has first secured immunity to political criticism and the right of suppression it entails. An artistic critique aimed at the heart of a political system can only be produced and heard if many within that system are saying, with more or less good arguments, that the arts must be allowed to say what they please. Finally, this study has suggested that an important corollary of the poet’s autonomy is that critics can thereby defend their own autonomy as readers, their freedom to find new interests and new conceptions of self and society. Whatever the political regime, a claim for autonomy is a claim that a place must be reserved where grown-ups may be allowed to hear and judge, in terms that they have negotiated for themselves, whatever may be said in art.
The basis for any study of ancient criticism remains the primary texts, many of which are shrewdly selected and translated by D. Russell and M. Winterbottom (1972). A. Barker (1984–1989) performs a similar service for texts on ancient music, which was closely allied to poetry. A valuable overview is The Cambridge History of Literary Criticism, Vol. 1: Classical Criticism, edited by G. Kennedy (1989). Ford (2012) tracks the rise of literary theory from Homer through Aristotle. The collection of essays Ancient Literary Criticism, edited by A. Laird (2006), is an excellent guide to further research, not only for the 20 “classic” essays he reprints but for his judicious and up-to-date appendices listing further reading.