Bryn Mawr Classical Review 1998.08.08
Mervin R. Dilts, George A. Kennedy, Two Greek Rhetorical Treatises from the Roman Empire: Introduction, Text, and Translation of the Arts of Rhetoric Attributed to Anonymous Seguerianus and to Apsines of Gadara. Mnemosyne Supplement 168. Leiden: Brill, 1997. Pp. xxvii, 249. ISBN 90-04-10728-2. $113.75.
Reviewed by Harvey Yunis, Rice University
Word count: 1473 words
The two treatises contained in this volume, both most likely from the first half of the third century CE, offer detailed, technical instruction on how to construct effective speeches in the milieu of declamation in the high Roman empire. Other than this generic common ground and basic contemporaneity, the two texts are not specifically related to each other. Apsines' treatise is rather more practical, that of Anon. Seg. rather more concerned with the genealogy of his rhetorical advice. Both texts are all business, making not a single reference to the contemporary world outside declamation. After its initial publication in 1840, the treatise assigned to Anon. Seg; so named from its discoverer, Séguier, Marquis de Ste. Brison, was placed along with Apsines' treatise in omnibus volumes of the Teubner Rhetores Graeci series, which contained what had been up to now the most recent and best edition of these texts (Rhetores graeci, vol. I.2, ed. C. Hammer ex recognitione L. Spengel, Leipzig, 1894). Neither text has previously been translated into any language. Apart from basic information in handbooks, the scholarly commentary devoted to the treatises is mostly to be found in scattered nineteenth-century publications. Even those proficient in Greek would be frustrated by the highly technical nature of these texts, which are crammed with arcane vocabulary and obscure references to the theories and declamations of the Greek rhetorical schools of the second and third centuries. For all these reasons these texts have been read by few apart from specialists in ancient rhetoric.
That should now change. Eminently suited for the task, Dilts and Kennedy have brought these treatises into the modern world. Classicists now have at their disposal better texts and fuller, more informative apparatus critici than were previously available, basic annotation demystifying the treatises and connecting them to the rest of the rhetorical tradition, and a new, rational system for citing Apsines. Interested students and scholars of other disciplines who do not read Greek, chiefly historians of rhetoric, literature, and education, will now be able to make effective use of these primary sources of the professional rhetorical world of the eastern Roman empire. They can rely on Dilts and Kennedy as trustworthy editors, translators, and guides. This volume joins several other recent publications that have brought a new degree of accessibility to primary texts of the high imperial rhetorical tradition: G. Ballaira's edition of Tiberius, De figuris demosthenicis (Rome, 1969), M. Heath's translation of and commentary on Hermogenes, On Issues (Oxford, 1995), D. A. Russell and N. G. Wilson's edition and translation of Menander Rhetor (Oxford, 1981), C. Wooten's translation of Hermogenes, On Types of Style (Chapel Hill, 1987).
Hammer's texts were intelligent and serviceable, but the current ones are better. In one place the editors have discovered a significant clause previously overlooked (Apsines 5.22 = 278.7 Hammer). Viewing the significant divergence between the two primary manuscripts of Apsines as largely the result of meddling by scribes in antiquity, the editors choose the most logical reading that either source presents. On the basis of new collations and previously suggested emendations, they have introduced many small changes of wording into both texts, producing treatises that read more smoothly and exhibit greater coherence than do the texts in Hammer's edition. The translations, which make use of plain English while conveying the impression of a technical work, are highly accurate. I discovered only two errors of any consequence. At Anon. Seg. 141 in a discussion of the style of emotional narrative, λελυμένην τοῖς κώλοις means not "dissolved by colons," which verges on incoherence, but "broken up by clauses." At Apsines 1.75 a motion introduced by Hyperides after Chaeronea that makes τοὺς ἀτίμους ἐντίμους is not one that makes "the dishonored honored," but one that "enfranchises the disenfranchised" (ἔντιμος = ἐπίτιμος, as Dem. 59.104).
The editors provide modern citations for the abundant references and allusions to classical literature, predominantly Attic oratory. Where the texts refer, often in an abbreviated way, to subjects of contemporary declamation, whether fictitious, historical, or conventional (e.g. "Scythian nomads founded a city and become sick; someone advises them to return to their former way of life" [Aps. 2.15], "Pericles introduced a motion to lay waste Attica; Nicias introduces a motion to begin from Eleusis" [Aps. 2.18], "reply to those swearing not to marry and being brought to court" [Aps. 5.21]), the editors explain and offer parallels in the extant literature where these are available. The editors achieve remarkable success in explaining the technical vocabulary and conceptual apparatus that notoriously plague this genre. Apart from basic concepts such as the parts of a speech, the technical vocabulary is not fixed in the tradition, but each rhetorician deploys his own variation of inherited terms, seldom in a manner that makes the meaning of such terms self-evident. Out of the dozens of such explanations presented by the editors, only one seems unsatisfactory. Anon. Seg. 30 attributes to Alexander, son of Numenius (a chief source for this treatise), the view that rhetoric is a "stochastic" art. This does not mean that rhetoric "'aims' at a particular effect in each speech." Rather it refers to the notion that rhetoric is a conjectural art, one that involves flexibility and approximation, as opposed to an exact art or episteme that never varies. This sense of a stochastic art, which goes back at least to Chrysippus (SVF II.19, cf. also Philodemus Rhet. B.6, 30, 39 Longo Auricchio), is precisely what allows Alexander to argue that the rules of rhetoric are not invariable but must be bent according to the demands of the occasion.
The Apsines represented in this volume, the one whose name appears at the head of the treatise in the manuscripts, has traditionally been understood to be Valerius Apsines of Gadara, mentioned in Philostratus and the Suda and born probably in the late 2nd c. CE. The identification is discussed and defended by Kennedy in this volume (pp. xv-xvi). In an article recently published in the American Journal of Philology (119  89-111), Malcolm Heath has argued that the manuscript attribution to Apsines is incorrect. Beyond the inherent unreliability of the manuscripts in such matters, suspicion is raised by the problematic way in which Apsines himself is mentioned in several places in the two manuscripts (listed and discussed by Dilts, pp. xxiv-xxv; nota bene: on p. xxv, three lines from bottom of main text, "39.11" is a crucial misprint for "156.21"). Heath's argument is too complicated to summarize and is as he admits, highly conjectural, but one aspect of his argument is of more general interest and clearly affected by the work of Dilts and Kennedy. (Although Heath mentions that he had access to the text of Dilts and Kennedy, he does not otherwise cite the new edition and it is not clear that he had use of it in its entirety.)
Heath draws inferences about the existence and authorship of various rhetorical treatises from extremely brief, obscure descriptions, ascriptions, and references in Syrianus and other ancient students of rhetoric. But such inferences can be sound only if the treatises were passed down essentially intact and with very little borrowing of material among them. Consider an abstract statement of the problem. Ancient student S attributes treatise T to rhetor R, mentioning its discussion of narrative N and declamation D. We possess a treatise that contains a discussion of narrative N and declamation D, but the manuscript ascribes it to rhetor A. Is the manuscript mistaken? Is the treatise we possess not really treatise T by rhetor R? We should deem the manuscript to be mistaken and the treatise to be treatise T by rhetor R only if we have reason to believe that the discussion of narrative N and declamation D occurred only in treatise T by rhetor R. That is precisely the kind of thing that the work of Dilts and Kennedy shows to be unlikely (cf. also H. Rabe, RhMus 67  321-57). First, beyond the common stock of subjects for declamation mentioned in this treatise and several others, they point out the significant overlap of specific passages between the Apsines' treatise and other parts of the rhetorical corpus (cf. p. 195 n. 251, p. 209 n. 280). The same phenomenon occurs also in Anon. Seg. (cf. p. 9 n. 28, p. 69 n. 242), whose treatise can be characterized as a pastiche of earlier rhetorical ideas. Second, on the basis of undeniable stylistic differences Kennedy suggests (pp. xvi-xvii) that the last chapter (on epilogue) in Apsines' treatise was probably added by someone other than Apsines, which may account for the problematic references to Apsines noted above. These features of shared and added material render Heath's inferences inherently unreliable: he has implicitly assumed too little contamination and too much neatness in the transmission of rhetorical subject matter for the attributions of the authorities he cites to be informative.