Bryn Mawr Classical Review

BMCR 2014.06.29 on the BMCR blog

Bryn Mawr Classical Review 2014.06.29

Walter Watson, The Lost Second Book of Aristotle's Poetics.   Chicago; London:  University of Chicago Press, 2012.  Pp. xii, 304.  ISBN 9780226875088.  $45.00.  

Reviewed by Richard F. Hardin, University of Kansas (


Professor Walter Watson, retired from teaching Philosophy at Stony Brook University, is deeply schooled in the entire Aristotelian corpus, but is here writing “a book intended for educated readers in general” (60). Inevitably, because he wants to situate the Poetics in the context of Aristotle’s whole work and methods, he must dwell on technical matters, but for the most part anyone interested in the first great work of literary theory will find The Lost Second Book of Aristotle’s Poetics accessible as well as valuable. Even anti-Aristotelians will appreciate the book, if only in finding new sources of thought in their disagreements.

The book builds on Richard Janko’s Aristotle on Comedy (1984), which argued that the supposed abstract or epitome of Poetics Book Two, in the manuscript known as “Tractatus Coislinianus,” is actually based on the lost second book, an opinion much debated over the previous century. The “Tractatus,” thought to have been written in the sixth century, is chiefly valued for what it says about comedy, a subject largely neglected in Aristotle’s surviving text. The epitome is reprinted in three pages of this book, with Watson’s phrase-by-phrase translation. Although its authenticity has probably been questioned more often than accepted, it has contributed to literary theory, for example in Northrop Frye’s adaptation of its threefold character cast of comedy—buffoon, eiron, and alazon—in his Anatomy of Criticism. Philology alone has failed to settle the matter of authenticity, but Watson believes that philosophy can augment philology through support from what he calls his “symbolon argument.” The term refers to the ancient Greek practice of guaranteeing contracts by giving each of the parties the broken half of a durable object such as a vertebra, each half of which was called a symbolon. The task, then, is to prove that the epitome of Book Two fits not just with what is said in Book One but with Aristotle’s general practices and body of work.

Accordingly, Watson proceeds succinctly through the philosopher’s program for the arts and sciences, from the Categories and other treatises in the group known as the Organon to the various fields of study, theoretical and practical. “Poetics is the last of the sciences,” Watson explains, “because of its dependence on all the others. It derives its scientific form from the Posterior Analytics, the theory of causes by which it is organized from the natural sciences and philosophy, the moral distinctions by which it discriminates characters from the Ethics, and the political order that it presupposes from the Politics” (43). In the ensuing discussion of poetic imitation, worth reading in itself, Watson offers a list of expectations raised about Book Two based on Aristotle’s method of discourse generally and on Book One in particular. For example, “Since the question of whether tragedy is more achieving of the end than epic poetry is left unanswered, we can expect it [given Aristotle’s method] to be answered in Book II” (81). A central problem has been that the “Tractatus” begins by defining poetry as either nonimitative or imitative, whereas, in Book One, it is said to be only imitative. The solution is that Book One “traces the genesis of poetry and its species,” while Book Two “begins from poetry as already in existence, and considers the ends which it realizes” (87). Another problem is that all the parts of comedy—plot, character, thought, etc.—are called “matter,” whereas in Book One they were “forms.” The solution: “In relation to the differentiae of imitation with which Book I begins, the parts are all forms taken on by these as matter, but in relation to the ends of poetry, from which Book II begins, they are all matter for the realization of these ends” (94). The parts of the symbolon are shown to fit; the symbolon argument comes to an end. One might ask whether the “Tractatus” might be the work not of Aristotle but of someone who understands the philosopher as well as Watson does, so the thought, but not the authorship, could be Aristotelian. But from a purely philosophical point of view, Watson seems to have realized his aims.

Watson’s background at the University of Chicago, in the time of Richard McKeon, has made him no stranger to the Chicago school of literary theory, echoes of which permeate this study. A familiar theme of the school is its twofold division of poetry or literature into mimetic and didactic, defined according to the aims of the text. The “Tractatus” uses slightly different terms, with nonimitative poetry classed as either educational or historical, rather than didactic, both having instruction as their purpose. Rasselas, The Divine Comedy, and Paradise Lost are, for Watson, educational poems. A historical poem is Tennyson’s “The Revenge,” about the last fight of that warship, which he contrasts with Shakespeare’s imitative poem, Henry V. The latter “does not depend on the actual occurrence of the victory at Agincourt, but on the universals present in the play itself, which make the outcome of the battle probable whether it occurred or not” (118-19). Some nonimitative poetry, such as encomia, may resemble rhetorical texts, the difference being that the end of the poetic encomium “lies in the poem itself, whereas the end of rhetoric is to be effective with respect to an audience” (123).

The mention of “universals” touches on one of the great ideas in Book One, that of what W. K. Wimsatt, no Chicagoan, called “the concrete universal,”—the principle that poetry is more philosophical than history although it shares history’s individuality and concreteness. This topic is central to the chapter on “Educational Poetry.” The chapter on catharsis in the Poetics (in tragedy “the removal of fears through their actualization”; in comedy the removal of pleasure and laughter through pleasure and laughter [147, 179]) is involved with an interesting concept in the “Tractatus,” that of symmetry, meaning the balance between the emotions to be removed and the emotions that do the removing. Watson provides a number of examples of symmetry contributing to the success of tragedy. Our enjoyment of Oedipus’ inquiry and discovery is balanced by the fearful outcome of that inquiry. The audience takes pleasure in the kind of man Hamlet is, but this is balanced by the fear that, being that sort of character, he will fall victim to tragedy. “The endless fascination of Cleopatra balances the claims on Antony of the struggle for supremacy in Rome” (162).

Someone expecting that this book will seriously investigate and expound Aristotle’s theory of comedy may not find that it satisfies that interest. The longest chapter in the section on comedy, and by far the longest in the whole book, is a discussion of the laughable, dwelling mostly on humor rather than comedy. One wants to hear a serious discussion of comic plot, to which the book gives less than a page. Is it true, as many have said, that good comedy sacrifices character (always flat) for an intricate plot? Is it possible to have themes in comedy, or must comedy adhere to its festive origins, free of all care? How does comedy stand in relation to the values of the society for which it was written? A major disappointment for anyone interested in ancient comedy must be the utter neglect of Plautus’ and Terence’s comedies, which are, after all, (1) our principal examples of post-Aristophanic comedy, and (2) based, to varying degrees, on lost Greek comedies of that period. If anything discloses the weakness of the “Tractatus” on comedy it will be seen in the discussion of character. The buffoon, the eiron, and the alazon or braggart are supposedly the “only three characters . . . identified as the characters of comedy” (237). Yet this does not begin to exhaust the characters of New Comedy known through the Romans—senex iratus, senex generosus, meretrix, uxor dotata, and so many more—none of which consistently play the part of braggart, ironist, or fool.

Watson has been writing on these subjects for many years, and his obvious expertise lends him the confidence to make Johnsonian declarations like “the virtue of diction is to be clear without being commonplace” (43). He compartmentalizes ideas so that, say, no text could be both educational or historical and imitative. How can one be certain which of these really represents the author’s intention, or whether a long text like War and Peace might not be both nonimitative and imitative? Are the forms or kinds of literature fixed with Aristotle, or do they continue multiplying and morphing over time? That question needs asking because Watson is prone to look for literary forms today that continue those known to antiquity, claiming, for example, that satyr plays have their modern equivalent in the Western, because “Satyrs dwell on the lawless frontiers of civilization” (137). Is comedy truly imitative in the way that tragedy is, and therefore grounded in the real and probable? And can it really be the case that comedy’s catharsis is the removal of pleasure and laughter? The limitations of the chapters on comedy mar an otherwise absorbing and informative book, and mean that we must still turn to Lane Cooper, Leon Golden, Janko, and M. S. Silk for more thoughtful reconstruction of the material at the heart of the lost Book Two.

Watson’s interest in fitting the Poetics within the rest of Aristotle’s corpus explains his addition of an appendix on the order and provenance of the works. He argues that the work of Andronicus is generally correct but requires some modification based on the histories of Strabo and Plutarch.

Read comments on this review or add a comment on the BMCR blog

Read Latest
About BMCR
Review for BMCR
Support BMCR

BMCR, Bryn Mawr College, 101 N. Merion Ave., Bryn Mawr, PA 19010