Bryn Mawr Classical Review

Bryn Mawr Classical Review 2003.06.07

Andrew Ford, The Origins of Criticism: Literary Culture and Poetic Theory in Classical Greece.   Princeton:  Princeton University Press, 2002.  Pp. 376.  ISBN 0-691-07485-2.  $45.00.  

Reviewed by Eustratios Papaioannou, The Catholic University of America, Washington D.C. (
Word count: 3433 words

Andrew Ford has taken on the enormous task of tracing the historical background of critical language and the establishment of criticism as a distinct discipline. Ford (henceforth F.) has executed this task with precision, poignancy, and insightful erudition. Covering roughly two centuries of criticism, from late sixth-century archaic views on song performance to the formation of poetics in the fourth-century, F. retrieves archaic criticism from its modern, dominantly philosophical, historizations. These historizations usually reconstruct pre-Platonic criticism filtered through the agendas of fourth-century philosophers. As F. argues, pre-classical criticism was primarily social as it did not distinguish song from its performance and social setting; thus it cannot be fully understood as mere anticipation of fourth-century poetics. What eventually produced fourth-century poetics was a series of theoretical changes, which F. places in the classical period. These theoretical changes gradually transformed song into poetry (i.e. text that might be analyzed by rules appropriate solely to itself) and the judging of performance into literary criticis. In F.'s words (p. 22), "as it moved from public acts of praising and blaming performances to school lectures or treatises on the optimal form of poetic texts, Greek criticism progressively effaced the social functions not only of song but of criticism as well." Thus "it is first in the fourth-century that we can recognize something like the modern notions of literature as imaginative writing and of literary criticism as a special knowledge of such writings" (p. xi).

Many of F.'s arguments echo familiar themes found within the scholarship of the last two decades,1 including his own work.2 Here, F.'s arguments are combined to form a cohesive narrative as he focuses exclusively on the historical line of thought and practice of criticism that led to fourth-century poetics. He thus manages to uncover previously unnoticed links and reconstruct nicely both archaic and classical criticism. While the general argument may not satisfy everyone with its sometimes forced linearity, the detailed exposure of historical and theoretical links as well as the engaging discussion of texts and contexts will reward the reader with rich insights.

The Book

Rightly separating criticism from aesthetic response, F. defines the former (the focus of his study) as "any public act of praise or blame upon a performance of song" (p. 3). Criticism, therefore, takes place within "critical scenes," which are defined as events during which the performance of song and the judgment of it were complementary social performances (pp. 3-4). This is the criticism that F. surveys in the Introduction and the first two parts of his book (Part I, "Archaic Roots of Classical Aesthetics" and part II, "The Invention of Poetry"). It is a criticism that according to F.'s own definition was preoccupied not with aesthetic or artistic categories, but with social ones. From Homer to performers of the first half of the fifth-century the object of criticism was song performance, not "poetry." Thus "genres" of song were regulated by social function and context in contrast to classical genres, which were regulated by rhetorical criteria of content and form (pp. 10-22). Critical concepts, that were to become fashionable in classical and postclassical aesthetics such as appropriateness (prepon, kairos; pp. 13f.), orderliness (kosmos e.g. pp. 36 and 117), and measure (metron; e.g. pp. 39f. and chapter one on archaic sympotic criticism), as well as less conspicuous concepts such as "figure" (σχῆμα; p. 44) or "purity" (καθαρότης; p. 55), were originally, as F. demonstrates, social and behavioral categories.

In the first two parts F. adeptly recovers archaic criticism from its "classical" interpretations. The following are but a few of F.'s convincing examples: the later sixth-century Xenophanes's rebuke of Homer and Hesiod (often viewed as a precursor of the Platonic quarrel between philosophy and poetry) is discussed in chapter two as part of Xenophanes's general quest for the refinement of singing and social conduct; the late sixth, early fifth-century Simonidean song metaphors (which are usually associated with concepts of mimesis and the perception of poetry as a "craftsmanly" art) are repositioned by F. in chapter four within the archaic conviction that song is the only medium able to grant ever-flowing fame to whom it memorializes; the same archaic conviction persists within the metaphors of craftsmanship that increasingly surface in other performers of the first half of the fifth-century such as Pindar, Bacchylides, or the Thucydidean Pericles (as argued in chapter five).

In parts III ("Toward a Theory of Poetry") and IV ("Literary Theory in the Fourth Century"), i.e. chapters seven through twelve, F. traces a sequence of fifth-century "inventions" that gradually cancel archaic "social" or "functional" criticism and produce fourth-century "formal" poetics, i.e. modern criticism. As F. would have it, classical/modern criticism employs a critical language that measures poetry by formal criteria and grants to artistic products an autonomous aesthetic value. The following are the stages toward Aristotelian formalism in F.'s account:

(a) the appearance of the terms "poetry" and "poet" in fifth-century historical or "anthropological" (p. 144) interpretations of the cultural past (chapter six);3

(b) the conceptualization of language as a "material" entity in the writings of a proto-scientist like Democritus and a "scientifically" enlightened sophist like Gorgias (chapter seven);

(c) the invention of a concept of "literature" by such theorists of prose as Alcidamas and Plato, but primarily Isocrates; here the term "literature" means prose that "gives meaning in a unique way" (p. 249) or a "work of art worth preserving and circulating for close and repeated study" (p. 248); literature, that is, as a distinct formal entity with an artistic value (chapter ten) classified in purely technical terms such as genres (chapter eleven; see also the Introduction);

(d) the advent of the "art of poetry" qua literary criticism as exemplified in Aristotle's Poetics (chapter eleven);

(e) the emergence of the literary critic as a "professional expert" as envisioned, to some extent, in Aristotle's Politics and as realized in Hellenistic, Roman, and Western cultures ever since (chapter twelve).

These "inventions" were contemporaneous to an increase in writing in classical culture (which F. cautiously regards as an important but not all-explaining factor; see pp. 152-157). Furthermore, the first three inventions were concurrent, in F.'s view, with persisting archaic mentalities. For instance, the regulation of poems by social values is visible in the use of poetry in Athenian education during the fifth-century, a period in which trends to objectify and demystify old poems are also manifest (chapter eight). Similarly, Athenian "democratic" judging of dramatic contests was not far removed from archaic concerns for the well-being and cohesion of society (cf. F.'s discussion of homonoia in pp. 277-282). Plato too is, according to F., a literary critic with an eye constantly toward the archaic past, either when he valued allegoresis for the select few (as discussed earlier in the book, pp. 85-89) or when he favored a purified "divine" discourse (pp. 258-261). Indeed, both Plato and Isocrates emerge in F.'s narrative as impure literary critics. Plato was an "irritant" (p. 209) in that he established much of the theoretical framework that occupied Aristotle's discussion of poetry but rejected poetry because of his ethical and ontological fixations. In Isocrates, theoretical formalism was only one aspect of a wider "humanistic" project, since, as F.'s points out, language is in Isocrates's view "a natural endowment that art may perfect to serve human ends" (cf. pp. 257-258).

As might be already apparent, F. deals with heated issues of the history of classical criticism. To the ones mentioned above, I should add Plato's censoring attitude toward poetry in the Republic (discussed in chapter nine) and the quite differing stances toward writing that one encounters in fourth-century intellectualist debates (chapter ten).4 F. displays an impressive knowledge of both texts and modern scholarship and contributes original insights to each text he explores. These would be impossible to discuss comprehensively in a mere review. The following, therefore, is a survey of several avenues that F.'s discussion opens and reflections on F.'s arguments that revolve around Aristotelian formalism.

The Constraints of Criticism

One may glean three recurrent themes in F.'s narrative which could be viewed as a set of constraints within which ancient criticism operated. The first constraint is the interference of the "critic's" self-representation in the formation of critical language and critical stances. The performance of self was greatly embedded in criticism. This is the case, e.g., with those Homerists who resorted to allegory (or, as F. demonstrates in chapter three, repositioned epic poetry into the class of esoteric verse) in order to be viewed as the correct interpreters; similarly Pindar projected a persona "appropriate to the occasion" (p. 115).5 Along these lines, declarations on "literature" by an aspiring intellectual such as Isocrates may lose the theoretical import that F. ascribes to them. I wonder, for instance, if Isocrates's concept-in-the-making of "genre" is part of his self-projection rather than his formalistic theory. F. speaks emphatically, in reference to Isocrates, of "uncrossable lines" between genres (p. 257), but Isocrates seems rather vague and flexible in both his theory and practice of "genres," displaying a fluidity that perhaps served well his self-portrait.6

Gender categories also imposed a constraint on ancient criticism. F. touches upon gender constraints in his Introduction when he states that "publicly pronouncing on song will remain a male prerogative" from Homeric "critical scenes" through the fourth-century (p. 7). Following F.'s sources, one notices that it was not only "male discourses" that dominated the field of criticism but also masculine discourses. This can be discerned, for instance, in the historical scheme which the fourth-century cultural historian Glaucus employed in order to describe the history of music (pp. 140-142: first music = human passions = barbarian East as opposed to later music [Orpheus and Homer] = reason [logos] = Greek West) or in Aristophanes's view of Homer as a "teacher of 'marshaling (taxeis), acts of courage (aretas), and the arming of men'" (pp. 199-200 on Frogs 1036; also p. 144). But masculine criticism is also evident in Gorgias's paralleling of violent, active, and dominant logos with God, Man, and Eros that cause a passive experience (pathos; cf. Encomium of Helen 37-38, 51-52, 58-59 96-99). Similarly Isocrates claimed that ideal discourse requires "a masculine soul" (Against the Sophists 17). If one is to include Plato's interrelated notions of femininity, pathos, mimesis, and tragedy, then gender becomes an element of ancient criticism that should not be underestimated.

The third constraint is the origin of discourse. In most of the authors discussed in the book, origin of discourse is a fundamental aspect of their valuation of poetry, whether the origin is perceived to be the divine (e.g. in Plato) or nature (e.g. in Isocrates). Incorporating the question of origin of discourse may, again, in some cases, lead to different interpretations than those of F. For instance, Simonides's claim that inscription is inferior to natural forces and that it is only "ever-flowing" song that can memorialize those who excel in virtue is viewed by F. as a critique on fixity, stability and an affirmation of flux, continuing life (pp. 101-112 on Simonides's response to the notorious Midas Epigram). Simonides's concern, however, seems to be the origin of discourse. Is not the opposition that he constructs between the fragility of human artifacts and the eternity of forces that originate in the divine or in human virtue?

Aristotelian Formalism and Methodological Presuppositions

The above constraints suggest continuities in classical criticism and it is to the theme of continuity that I now turn. F. often vacillates from affirming to debunking continuity between archaic and classical and (less so) between classical and modern criticism. While most of F.'s argument focuses on historical transition from archaic to classical (primarily Aristotelian) poetics (cf. pp. 5, 8, or 269), the book ends with an affirmation of continuity: "little had changed since archaic times" (p. 296). The sub-title of the book, "Literary Culture and Poetic Theory in Classical Greece," is also put in terms of continuity. Here, literary culture and poetic theory are combined rather than contrasted. If, however, one is to correctly understand F.'s explication of this sub-title (according to pp. 1-5), for Literary Culture one should read: [From] musical culture, for And read: to, and for Poetic Theory read: Self-conscious, formal theorization of poetry. Furthermore, it is unclear if, according to F., the birth of modern criticism is equivalent to the birth of a modern concept of literature (affirmed in pp. xi and 22, questioned in p. 157; see also 229-231 and 248-250). F. is similarly vague as to whether it was Aristotle's Poetics or his view of the kritikos as expressed in the Politics that resembled later (Hellenistic, Roman, Western) criticism (cf. pages 293 and 294 where the contrast is most evident) .

However hesitant F. may seem, one still senses a tendency to turn polarities (which are to some extent imposed by modern scholarship) into historical linearities and vice versa. This is visible in F.'s definition of criticism. In defining criticism as "any public act of praise or blame upon a performance of song" (p. 3), F. has, by default, excluded Aristotelian criticism, which is, in F.'s opinion, "a field of expertise and an independent academic discipline" (p. 270), a "systematic map of all forms of literature ... and a technical account of how each achieves its peculiar effects" (p. 251). There are two possible questions that one might ask with respect to this division of ancient criticism into original, "social" and later, "independent" or "formal": a. Did "social" criticism cease to exist in and after the fourth-century? (F. seems simply to stop looking for "critical scenes" after the fifth-century); b. Was there indeed such a radical shift in critical theory that resulted in the replacement of archaic criticism with "formalism"?

F. regards the developments in terminology -- the transference, that is, of social terms into artistic categories -- as nearly absolute transitions that can be explained through a series of opposites. Thus, for instance, song and poem are polarities (the scheme: one as opposed to and excluding the other), as well as, historical linearities (the scheme: from the one to the other). The list may extend to include archaic vs. classical (passim), public (event) vs. private (aesthetic response) (cf. p. 3), performance vs. text (p. 4), functional vs. formal (cf. p. 8 and passim), event vs. object (cf. p. 9 and 157), dining vs. lecture hall (p. 44), [unconscious (?)] vs. conscious (cf. p. 4) and [blank] vs. Western (passim).

The insistence on the "formalism" of Aristotle's Poetics, as opposed to archaic criticism is not a new matter. It is worth considering, however, how F. perceives formalism and whether this is justified by Aristotelian theory. According to F., formalism is the view that works of arts are precisely works of art, that is, objects contained in themselves that are not to be judged by moral or ontological categories outside of their own artistic criteria. They are objects whose form and structure contain meaning free from any context or performance, objects that cause an aesthetic pleasure which is irreducible to any other function, e.g., ideological, social, moral, or epistemic functions. Thus, in F.'s formalism there exists a series of autonomies that parallel the autonomy of the art-object: the autonomy of aesthetics, of fictionality, of textuality, of criticism, and ultimately of the critic.

If one turns to Aristotle, however, one notices that there exists an unbroken continuum of ontology, epistemology, morality, and aesthetics, a continuum that resists autonomies. In fact, one may argue that the Poetics is an attempt (much along the lines of Platonic anxieties) to include, explain, and thus regulate artistic products within a philosophical continuum. If criticism ever became an autonomous discipline, this would not seem to be Aristotle's explicit objective, however open for interpretation his texts may be.

For Aristotle's ontological enclosure of aesthetics, it may suffice to note here that "nature" (Aristotle's basic ontological category, as described for instance in his Physics) is more central in Aristotle's theory of poetry than "art" -- notably, words with the stem phu are more common than τέχνη (cf. p. 262). Aristotle views nature as poetry's origin (cf. 1448b4-5), as its mode of existence (cf. 1449a15), and ultimately as its telos. The aim of poetry is muthos7 or the "composition" (sustasis) of an action, and this composition is regulated by what is plausible and necessary (cf. 1450a22-23, 1450b29, 1451a12-13 and 36-38). In other words, muthos, the remaking of an action (the only thing that poetry "makes"; cf. 1451b27-32), is bound by ontological criteria (cf. Rhetoric 1357a.34-b.1, 1393a.4-8; Nicomachean Ethics 1139b.22-24, 1140a1-23, b.31-32; Posterior Analytics 96a.15). Poetry, as any art in Aristotle's view (Physics 199a15-17), never creates anything outside what already exists in nature or could exist in natural ways.8 Ontologically speaking the possibility for a fictional world or an autonomous art seems to be denied in Aristotle.

Regarding Aristotle's epistemological enclosure of criticism, it is characteristic that he ingeniously transforms the art of poetry into a quasi-science, an episteme. This is evident in his view that poetry "speaks"9 universals rather than particulars. Universals (in the case of poetry, probability and necessity) are the subject of episteme in Aristotle (cf. Nicomachean Ethics 1140b31-32). To be sure, poetry is not purely an episteme, hence the mallon in Aristotle's formulation. This is the case because, as Aristotle observes elsewhere, art flirts with chance (Nicomachean Ethics 1140a17-23) and tuche cannot be a matter of episteme (cf. Posterior Analytics 87b.19-22) since it is the opposite of nature (Magna Moralia 2.8.2) and does not produce proper praxis (Politics 1323b21-29). Aristotle wishes to ensure that art does not operate by chance (cf. Poetics 1450b32-33 and passim) but that it acquires a function of epistemological cognition bringing it closer to philosophy than history.

Finally, Aristotle's emphasis on action/condition (praxis), found in his definition of poetry, signifies his rehabilitation of tragedy in moral terms. The poetic remaking of praxis according to the universals of probability and necessity is not self-referential; poetry enables the cognition of the universals of praxis, not of poetry itself. This has moral implications because in Aristotle the knowledge of the universals of praxis is fundamental for proper praxis, which in turn is the necessary complement of knowledge (cf. Protrepticus 52.1-10). A significant aside: elsewhere in Aristotle praxis and poiesis are clearly distinct (cf. e.g. Nicomachean Ethics 1140a1-23).10

If Aristotle's theory of poetry does not champion formalism but rather places poetry in the continuum of being/knowing/acting (an argument that could be equally made for Plato or Isocrates), then it becomes questionable whether it was radically different from archaic poetics. If, as F. rightly suggests, the Poetics addressed primarily readers and critics (cf. p. 266), it offered to them a system of evaluation that favored unity of interpretation regulated by the threefold continuum as described above, much like archaic criticism imposed on performers, expectations of permanence, quests for access to truth, and societal norms.11 Ultimately, in both archaic and classical criticism the concern was the totality of being/knowledge/ethics which poetry was to affirm rather than break and which criticism was to regulate rather than disseminate, however polyphonic and open-ended both poetry and criticism were in reality. If there was any explicit auto-nomy in ancient criticism, it was expressed as a movement toward tauto-nomy, the literal rule of the same.

Perhaps a similar anxiety for sameness can be discerned from F.'s posited polarity between Western criticism and its unnamed opposite. To assume a continuity between classical and western modern criticism one risks (a) finding "connections," such as formal criticism, between modernity and its "origins," (b) undervaluing difference and polyphony of both antiquity and modernity (in all fairness, F.'s provides detailed discussions that do indeed expose a multivalent Greek criticism), and (c) neglecting other pre-modern critical discourses that were in dialogue with antiquity, but could be regarded as neither classical nor modern. I wonder, for example, where in a narrative that favors the continuity of "Western" criticism would pre-modern Greek criticism of Late Antiquity and the Middle Ages fit. Indeed, if we were to continue F.'s project into Hellenistic, late antique, and medieval criticism, our understanding of both classical and modern criticism would deepen drastically.

The study of pre-modern Greek literary criticism is evolving, and F.'s contribution makes a substantial and significant imprint. While some of his convictions may be questioned by those of us who view "with suspicion" (as F. notes [p. x]) the search for origins, this eloquent book12 will be an instant complement to any study of the history of criticism. It restores the often ossified, overlooked terms of ancient literary theory through its thorough manner of exposition, and it historicizes the same old debates surrounding what literature is and how we are to discuss it.


1.   Cf. for instance Gregory Nagy's meticulous account of "Early Greek Views of Poets and Poetry" in the 1989 Cambridge History of Classical Criticism edited by George A. Kennedy, Margalit Finkelberg's thoughtful tracing of Greek notions of fictionality in her The Birth of Literary Fiction in Ancient Greece (Oxford, 1998), or Yun Lee Too's stimulating The Idea of Ancient Literary Criticism (Oxford, 1998), a book that in my opinion F. circumvents too lightly; cf. p. 1 n. 2.
2.   Cf., for instance, Andrew Ford, Homer: The Poetry of the Past (Ithaca N.Y., 1992); see p. 5 on the essential difference between Homeric views of poetry and Aristotle's poetics; pp. 13f. on the difference between "song" and "poetry," on the socially/contextually determined concepts of genre in archaic poetry, and on F.'s fundamental thesis (now fully developed in the Origins of Criticism) that prior to Aristotle there was no "autonomous art of poetry" operating separately from the "entire organization of social life." F. situates his new study within the scholarly tradition of Gentili and Havelock, among others (see Origins of Criticism, pp. 8-9).
3.   For the purpose of this review, I include here chapter six (though it technically belongs to F.'s Part II), because the invention of "poetry" is the first important change in the paradigm of archaic poetics.
4.   F. compares nicely Alcidamas, Plato, and Isocrates, yet I wonder if the oppositions involved in their views on writing are more nuanced than F.'s broad argument allows. For instance F. forgoes Isocrates's stress on novelty and creativity in discourse as opposed to the "fixity" ( τεταγμένην ) of writing (not of rhetoric as F. argues in p. 236 on Against the Sophists 12-13); notably, tetagmenos acquires various semantic colorations in Isocrates, whose discourse is characteristically polyphonic. I am also not certain whether in Plato writing is opposed to "a dynamic conversation that occurs and changes in time" (p. 247) or to philosophical discourse that is written in the soul of the student, i.e. is authoritatively interpreted and properly regulated. Writing seems to be problematic for Plato not because of its "fixity, silence, and lifelessness," but because it allows for dissemination outside the privileged few and, thus, misinterpretation; writing challenges the presence of the one and knowing speaker within his discourse.
5.   For rhetoricians fashioning themselves as doctors see pp. 162f.; for sophists turning Homer into a teacher in order to sanction their own claims for teaching traditional virtue see pp. 202f.
6.   His playful Encomium Helen e.g. is a far cry from being purely either an encomium or of Helen alone; as for his terminology, his notion of idea has a characteristically indeterminate meaning.
7.   The aim of tragedy is not a "catharsis of pity and fear" (p. 263); katharsis is rather a side-effect of tragedy; cf. Malcolm Heath's introduction to his translation of the Poetics for the Penguin series (London, 1996), pp. xxxv-xliii.
8.   For this interpretation of the notorious passage of the Physics see H. Blumenberg, "'Nachahmung der Natur': Zur Vorgeschichte der Idee des schöpferischen Menschen," in Wirklichkeiten in denen wir leben: Aufsätze und eine Rede (Stuttgart, 1981) 55-103; F. himself rightly observes that the ancient notion of "making" as related to "poetry" did not imply creation ex nihilo. On the question of the enclosure of mimesis by "natural" ontology see also Paul Ricoeur (La métaphore vive [Paris 1975] pp. 13-61, esp. 51f.), who, while employing similar views of Jacques Derrida ("La mythologie blanche" Poetique 5 [1971] 1-52), affirms the autonomy of criticism established by the Poetics as well as the possible autonomy of fictionality that is inherent in Aristotelian mimesis. In my opinion, both autonomies are potential interpretations of Aristotelian theory (as well as of Gorgian, Platonic, and perhaps even Homeric literary theory for that matter), interpretations that do not, however, account for Aristotle's explicit combination of ontology, epistemology, and ethics. While these interpretations wish to discover liberating elements in Aristotelian theory, they simultaneously assign roles, both too narrow and too segmented, to Aristotle's notions of "nature" and "art".
9.   Poetry does not "represent" universals (p. 267). It should also be noted that the verb λέγω, employed by Aristotle, has a particularly authoritative connotation that relates poetry with episteme; cf. Nicomachean Ethics 1138b19-20 and 1220b28.
10.   Beyond the significance of praxis in Aristotle's definition of poetry, it is also evident that morality plays a central role in the formation of Aristotle's kritikoi and in the formation of their "criticism" as described in his Politics; for the social imperatives inscribed in Aristotle's view of criticism see Too, The Idea of Ancient Literary Criticism, pp. 82-114.
11.   In addition, one may question whether there was the transition from evaluating poetry as "performance" to regarding poetry as "text." The transition from oral to written discourse and the consequent treatment of poems as texts was a change in reality rather than in mentality. Indeed neither did archaic criticism seem to have appreciated song as performance, nor did classical criticism value poems as texts. Simultaneously, the symbolic prioritization of orality as presence (not as performance) was a persistent feature of pre-modern literary theory; it is characteristic, for instance, that in Ps.-Demetrius, On Style, a first-century AD text which summarizes the Aristotelian tradition in Hellenistic rhetorical theory, letter-writing (the written discourse par excellence) is defined by its closeness to oral communication and its corresponding aversion to performance (what Ps.-Demetrius significantly calls the mimetikon; On Style 223-235).
12.   It is no surprise that, in a lengthy and detailed book such as this, one finds typographical errors, primarily quotations in Greek and, to a lesser degree, bibliographical references in German. A full list of those typos is perhaps unnecessary; rather I wish to emphasize my appreciation for F.'s offering the original Greek text, rather than simply his translations which are indeed excellent. An item to be mentioned, however: Simonides's much discussed fragment usually referred to as Psellus 821B Migne (= Migne, Patrologia Graeca vol. 122; quoted by F. in p. 97) is preserved in a text which does not belong to the eleventh-century AD writer Michael Psellos; on the text see now Paul Gautier, "Le De Daemonibus du Pseudo-Psellos," Revue des Études Byzantines 38 (1980) 105-194.

Read Latest
Index for 2003
Change Greek Display
Books Available for Review

HTML generated at 13:28:37, Friday, 03 April 2009