When, towards the end of the first century after Christ, the rhetorician Dio of Prusa called Aristotle the figure ‘from whom they say that criticism [kritike] and the study of language [Grammatikē] took their origin’ (53.1), it was not the Poetics which he, or those whose judgement he was citing, had in mind. Dio mentions in this context the ‘many dialogues’ in which Aristotle had discussed and praised the poems of Homer, and the phrase indicates that it was the work On Poets, among others, to which he was referring. The three books of On Poets, and the six or more books of Homeric Problems (presumably not in dialogue form), were in fact the two chief works in which Aristotle's ideas on poetry were disseminated in the ancient critical tradition; while the Poetics, originally produced for use within the philosophical school, never became at all readily available or widely known. Our own view of Aristotle the literary critic is therefore bound to differ substantially from that of antiquity; and it is the more inward voice of the philosophical theorist of poetry which the Poetics, if incompletely, allows us to hear. But it is a useful first step in a fresh approach to the surviving text to consider briefly what can still be discerned of its relation to the lost works on poetry which Aristotle intended for a more public audience.

The great challenge for any interpreter of Plato's views on poetry is to appreciate why he is so uncompromisingly hostile towards it. That he should seek to subordinate poetic to philosophic measures of expression and understanding is not in itself surprising. Philosophy has long had a need to keep poetry in its place - as Plato, alluding to the ‘ancient quarrel’ between the two, was among the first to tell us (Rep. 10.607b). But what is striking in Plato's attitude is that even when he comes to acknowledge a usefulness in poetry - its role in educating the young, in civil celebration, in persuasion of many sorts - he is not content (as is, say, Aristotle) to grant its virtues, unstintingly, while nevertheless delimiting their scope; rather, he regards poetry at all times and in all its uses with suspicion, as a substance inherently volatile. He recognises that human society is not possible without some form of poetry, but discerns in this fact a mark, so to speak, of our fallen state. Many philosophers have measured their distance from the poets; but Plato would put them beyond hierarchy altogether; would banish them - at least, would banish those he confesses to represent poetry at its greatest - from his ideal society.

Poetry as performance: the example of Ion

We shall not appreciate the reasons for Plato's hostility towards poetry unless we bear in mind how poetry would typically reach the public in Plato's day. In a modern culture our most frequent direct contact with the literature deemed important in our society (and in the West this would of course include the very poets on whom Plato targets his attack) comes either through private and (at least potentially) reflective reading, or in the context of the classroom; and is supplemented in the case of drama by visits to the theatre, to see actual performance.

This chapter will survey a variety of intellectual movements in the time of the Roman Empire which are related to criticism, complementing what has been said in the previous chapter about neo-Platonism and extending the range into the writings of Fathers of the Church before Augustine and Jerome and the Latin grammarians and rhetoricians of late antiquity. Some of this material provides background for Medieval criticism, to be discussed in volume II; some developments are also of interest in that they seem to foreshadow critical issues of the twentieth century.

The search for meaning through interpretation

Greek intellectuals of the Roman Empire discussed epistemology and adumbrated semiotic and hermeneutic theories of interpretation. There was, however, wide difference of opinion among them. At one extreme lies the tradition of Pyrrhonian scepticism, which had begun at the end of the fourth century BC and influenced the Academic school of the Hellenistic period in its controversy with Stoic belief that certainty could be obtained from sense perception. Among surviving works of the Classical period, scepticism is best represented in Cicero's Academies. The fullest exposition of a radical scepticism is, however, found in the writings of Sextus Empiricus in the late second century of the Christian era. Sextus rejected sense perception as a basis of knowledge, and with it all claims for certainty in logic, physics, and ethics. The sceptical principles set forth in Sextus' Outlines of Pyrrhonism are applied in detail in Against the Dogmatists to undermine the validity of teaching in grammar, rhetoric, mathematics, astronomy, and music.

Histories of criticism in early Greece are usually based on surveys of those relatively few passages where the Greek poets speak about themselves and their poetry. Although this chapter will comment on many of these passages, the publication of a new history offers an opportunity to go beyond this fragmentary evidence by considering it in the wider framework of the society or societies for which this poetry existed. In what follows, the primary evidence is not restricted to whatever the poets say about themselves and their world: rather, it embraces the context in which they say what they say. The task will be to describe the social function of early Greek poetry and to present a picture of the traditional thought-patterns that shape the very concept of poet and poetry. It is through these thought-patterns that early Greek poetry defines itself and the poet as well, making it ultimately possible for critics of later times to talk about poetry.

The very notion of ‘critics’ and ‘criticism’ can best be seen in the post-Classical context of a great period of scholarship, in Hellenistic Alexandria. The Alexandrian concept of krisis, in the sense of ‘separating’, ‘discriminating’, ‘judging’ those works and those authors that are to be preserved and those that are not, is crucial to the concept of ‘canon’ in the Classical world. Literally, kanōn means ‘rod’, ‘straight-edge rule’, then by synecdoche a ‘standard’, ‘model’. The Alexandrian scholars who were in charge of this process of separation, discrimination, judgement, were the kritikoi, while the Classical authors who were ‘judged worthy of inclusion’ within the canon were called the enkrithentes, a term that corresponds to the Roman concept of the classici, who are authors of the ‘first class’, primae classis.

The name ‘Silver Latin’ is often given to the literature of the first century of the Christian era and is generally understood to imply its inferiority to the Golden Age of the late Republic and Augustan era. Analogy with the five Hesiodic ages, in which the silver age was both later and less worthy than the golden, suggests the cliché of decline. To what extent did the Romans of the early imperial period feel that they and their contemporaries were a falling away from the previous generation? We will see that the change in form of government, by denying opportunities for significant political speech, trivialised the art of oratory. But was there any such external constraint on poetry?

Modern critics have reproached Silver Latin epic and tragedy with being ‘rhetorical’. Certainly it is clear from Tacitus' Dialogus that men thwarted from political expression transferred to the safer vehicle of historical or mythical poetry both the techniques and ideals of public oratory. But just as no one suggests that Juvenal's satires were poorer compositions because of his apparent rhetorical skill, so rhetorical colouring in the higher poetic genres of tragedy and epic is not necessarily a fault. We would judge the individual composition primarily by its internal coherence: but Roman critics like Quintilian measured a work by its conformity to the characteristics of its genre and defined those characteristics by a canon, which by his time consisted largely of late-Republican and Augustan writers. Thus for classicising critics of the first century ‘different’ meant worse, while the creative artists who achieved significant poetry or prose did so largely in reaction against a norm they could not usefully imitate.

Greek criticism of the first four or five Christian centuries presents a rich and diverse picture. It is not, however, one that can be complete in itself. Many of the basic concepts derive from Hellenistic or even earlier writings. The great elaborations of the rhetorical theories of types of issue (staseis) and types of style (kharaktērres, ideal) which we see, for example, in Hermogenes and his commentators, are firmly grounded in what was inherited from Hermagoras and Theophrastus and those who built on their work in the earlier period. The many attempts to discuss the relationship between literature and morals are still, in the main, a response to Plato. Furthermore, there was an important bilingual literary public - Romans who knew Greek, not Greeks who knew Latin, for these were few - and Latin literature had reached its classical acme and become a subject of study in its own right. There was thus a need to compare and contrast, to study the process of imitatio, but also to treat the two literatures as in an important sense one. In the late first and early second centuries the union seems particularly strong. Quintilian uses Dionysius' On Imitation as the source for his list of recommended reading in Greek, and the model for his corresponding advice about Latin (10.1). Aulus Gellius, reporting or embellishing the conversation of the elegant and pretentious academic circle of Antonine Athens, compares Virgil with his models in Pindar or Theocritus or Homer, and the comic poet Caecilius with his exemplar Menander.

If the Greeks were first in Europe to create and record literature, to develop literary genres and define their natures, and to evolve critical systems for describing and prescribing forms of rhetoric and poetry, the Romans, paradoxically, scored a different first. They were the first cultural community to inherit literary models - those set up for them by the Greeks - before they began to compose their own literature. It might be claimed that they practised literary criticism, however rudimentary, before they practised literature, for they were faced with questions of what to imitate and how. The emergence of a relatively developed, highly imitative, national literature in the third century BC has some analogies to the appearance of new criticism and national literatures in the Renaissance; in both cases critical theory, adapted from the prototype literature, helped to mould form and content and in both cases formal education in grammar and rhetoric provided norms for literary expression. But Renaissance writers in the vernacular had a richer tradition of native poetry on which to draw than did the Romans, richer lexical resources, and greater ambition for literary originality.

Livius Andronicus, Naevius, and Ennius

Of Rome's three earliest poets, two were professional teachers and both were culturally as much Greek as Italian. The few relatively secure facts about Livius Andronicus are representative of Rome's literary beginnings. He came from Tarentum in Greek-speaking south Italy, which had been captured by Roman forces between 280 and 275 BC. He probably knew little Latin when he first came to Rome, but from time spent in instructing young Romans in Greek acquired sufficient command of Latin to produce a translation of the Odyssey, not into quantitive Greek hexameters, but into archaic, accentual Saturnian verse.

Inherent in any literary criticism are assumptions about the nature of language and about what constitutes valid interpretation. This chapter will set out briefly what some Greek philosophers have to say on these and related subjects, but it should be recognised that we are often viewing their thought on the basis of modern assumptions about the implications of what they say, rather than entering into their own epistemic system. Until the fourth century BC we should probably grant that the Greeks saw language as a natural map of reality; they sought more often to understand reality with the tool of language than to try to understand the nature of language itself. Yet the discussion can provide a substratum of thinking which lies beneath the literary criticism of Aristophanes, Plato, and Aristotle, and a background for the discussion of the theories of Hellenistic philosophers in chapter 6, or for later Greek hermeneutics in chapters 10 and 11. Readers whose interests in criticism are not theoretical may, however, prefer to skip to chapter 3 and continue at this point with the applied literary criticism of Plato.

Early Greek hermeneutics

The Homeric poems already reveal a society grappling with interpretative problems. Nestor in Iliad 1 is represented as a wise old man whose insight, based on experience of situations and people, allows him to interpret and reconcile opposing views, and whose words ‘flowed sweeter than honey’ (249). Odysseus in both poems is astute and in the Odyssey often veils his thought, either as a form of self-protection or to test the attitudes of others.

Literary prose first emerges in the middle of the fifth century BC in writings in the Ionic dialect, including the Histories of Herodotus, then in the Attic dialect in the oratory of the Siciliạn Gorgias and the Athenian Antiphon, and is seen at the end of the century in Thucydides' History. Oratory and history, throughout antiquity, remain privileged prose genres, to which is added as a third the philosophical dialogue, developed by Plato in the fourth century. Except for a very few references to the literary epistle, ancient critics ignore all other prose forms as sub-literary. History, with its mixture of narrative and set speeches, may be thought of as corresponding to epic in poetry, the dialogue to drama; epideictic oratorỵ, seen in the display speeches of sophists, such as that attributed to Lysias in Plato's Phaedrus, or in funeral orations or speeches at festivals, has some relationship to lyric forms, such as the hymn, but the chief poetic antecedents of deliberative oratory are found in the debates in epic.

Aristotle claims (Rhetoric of Aristotle 1.1404a20 - 39) that prose became artistic in the first instance by borrowing stylistic features from poetry; a more accurate statement might be that analogies were created to the effects of poetic sound and rhythm, seen in the so-called Gorgianic figures of isocolon, paronomasia, homoeoteleuton, and the like. These are flagrantly indulged by Gorgias himself and found in more restrained form in Thucydides and elsewhere. As pointed out in chapter 1, section 3 above, ‘prose’ assumes the prior existence of ‘poetry’.