Tuesday, October 11, 2011


David Mankin (ed.), Cicero, De oratore Book III. Cambridge: Cambridge University Press, 2011. Pp. xii, 348. ISBN 9780521596572. $38.99 (pb).

Reviewed by Andrew R. Dyck, Los Angeles

Version at BMCR home site


David Mankin will be known to readers of this journal as the editor of Horace's Epodes (Cambridge Greek and Latin Classics, 1995).1 He now adds the first rhetorical work to appear in the series. This is also the first complete English commentary on De oratore 3 since Wilkins (1893), the learned commentary published by the Winter Verlag in Heidelberg covering only 3.96-230 in English (v. 5, 2008), the earlier part in German. Mankin explains in the Preface that v. 5 appeared too late for him to be able to use it apart from Renting's list of corrections to Kumaniecki's apparatus.

The Introduction provides the kind of background, historical and literary, needed by those approaching this text for the first time. At the beginning Mankin includes some remarks on the 'instigator' of De orat., Q. Cicero, but more might perhaps have been said about his career, his choice as dedicatee and his relations with his brother. Mankin is sensitive to Cicero's position as an Academic Skeptic who may have wished to withhold judgment on controversial matters; he also discusses the antecedents of the dialogue form and provides helpful guidance on the historical background and careers of the dramatis personae.2 He isolates the unique elements in the presentation of the 'quarrel' of philosophy and rhetoric in this work, the result of which is to raise the stakes for Cicero's Roman readers. His most original contribution in the Introduction is to argue that the diction and rhythm of the characters' speeches may be meant to evoke, for Cicero's older readers, the manner of the speakers themselves. The Introduction concludes with a brief discussion of the textual transmission.

Mankin has devoted considerable thought to his text, which is generally an improvement on that printed by Kumaniecki (a table listing divergences from his predecessor would have been useful). Briefly stated, the direct transmission consists of the codices mutili, copied twice in the mid-ninth century, then lost (=M), and the codex Laudensis discovered at Lodi in 1428 and copied four times before its disappearance (=L); in addition, there are testimonies, i.e. quotations embedded in the MS traditions of other authors (=T). Mankin continues the upgrading of L and, especially, T and takes account of Cicero's rhythmic preferences (see e.g. on 27 tribuatur vs. tribuitur). Though generally conservative, Mankin sometimes champions conjectures, especially ones he cites from an unpublished MS by one L. T. Brown (it would have been helpful had Mankin indicated the scope and location of this work); Mankin offers no conjectures of his own.

On occasion one wishes Mankin had shown greater boldness. Thus at 26 Lambin's 'alii <laudandi alii> uituperandi', an easy saltation error, surely should have been adopted (Mankin ad loc. admits it is 'attractive'). Again at 36 Mankin perhaps should not have retained the transmitted confirmaret, rather than adopt the early conjecture conformaret (as do most editors), in spite of references to the 'strengthening' effect of practice and reading at 44, 48 and 88; the imagery drawn from sculpture in our passage and the parallel at 200 ('hic nobis orator ita conformandus est . . .') surely make a convincing case for conformaret. At 51 Mankin is doubly cautious in enclosing 'quam te in uitia' in both square brackets and cruces, evidently because he diagnoses the words as a 'corruption of a gloss'. But the decisive point is whether or not the words are intrusive; if they are, they belong in square brackets, and cruces, which are for (potentially) genuine, but corrupt matter, are unwanted (there is further confusion in the note ad loc., where, by an apparent typo, the questionable matter is cited as 'quam te inuita'). At 52 Mankin retains 'tantummodo' of L (the only witness here) even though it is unparalleled in the 'non modo . . . sed' construction; but one can make a strong case for Bake's 'modo': surely 'tantum' and 'modo' were variants that came to stand side by side in L.3 At 78 we read 'quid est quod aut Sex. Pompeius aut duo Balbi aut . . . M. Vigellius de uirtute hominum Stoici possint dicere . . .?' Mankin ad loc. is right that Stoici in this position seems abrupt; and hominum seems an otiose limitation of uirtus; none of the solutions so far proposed is satisfactory. Perhaps read '<modo> hominum Stoicorum' ('in the manner of the Stoics'), whereby modo will have dropped out before hominum, which was then taken to limit uirtute, and the case of the following word adjusted to make it the subject of possint. Again in 92 surely Manutius' 'multo' is needed for 'multum' to match Cicero's practice elsewhere with a comparative (in spite of the different usage of other authors); and D's omission of et after undique and before collectis is likely to be right, its insertion here an easy dittography since c and t are confused elsewhere in this tradition (cf. Mankin on 99 apropos of terram/cerram). In 140 Brown's dis<c>endum surely merited adoption. Mankin daggers quasi conuersione in 190, where conuersione is surely intrusive from conuersiones in the previous line. If that is so, then the shape or sense of conuersio will be no guide to the word that it ousted. To judge by the context, one would expect either 'flumine' (cf. Brut. 325 or N.D. 2.1) or possibly torrente (cf. Quint. 10.7.23); this would be parallel to perpetuitas = 'the connected quality (of speech)'. At 211 we read 'aliud dicendi genus deliberationes, aliud laudationes, aliud iudicia, aliud sermo, aliud consolatio, aliud obiurgatio, aliud disputatio, aliud historia desiderat'. Bornecque wanted to substitute contentio for consolatio to effect a contrast with sermo (cf. Off. 1.132), a suggestion that 'merits consideration' (Mankin). I suspect, rather, that after the three categories of speeches we have three sets of pairs and that 'aliud contentio' has fallen out by saltation before 'aliud consolatio'.

On the other hand, Mankin is perhaps too quick to accept Lambin's 'humanitate dignam scientiam' for L's 'humanitatem dignam scientia' (94): his argument against the transmitted text is that 'Crassus would be conceding that Greek rhetors "possess" . . . the Roman quality of "humanitas"'. Now here 'humanitas' has the sense 'culture, civilization', which belongs par excellence to the Greeks; cf. e.g. Flac. 62 'adsunt Athenienses, unde humanitas, doctrina, religio, fruges, iura, leges ortae atque in omnis terras distributae putantur'; nil mutandum. Sorof's deletion of the first sentence of 157, followed by Mankin, seems implausible: what is perhaps needed is the insertion of 'si translatio' at the beginning of the sentence (si was proposed by Brown, translatio by Lambin), an easy saltation error before similitudinis.4 Again at 178 Mankin deletes et before 'in diuersam partem' on the grounds that it must mean 'also'; that is, however, not necessarily an obstacle, as Mankin himself shows on 193; but surely in 178 et coordinates the two adverbial expressions sensim and 'in diuersam partem'.

Mankin has equipped the text with a parsimonious critical apparatus and detailed commentary. The latter helpfully explains technical terms and background from philosophy and rhetoric and is particularly strong on fine points of Latinity and Ciceronian usage. It is also occasionally leavened with wry wit, e.g. on 96: 'At Petr. 90.1 people hearing Eumolpus reciting a poem throw rocks at him; Italy had not yet discovered the tomato'. The following are a few of the relatively rare instances where one can query his interpretation: On 42 Mankin says that the reference to Cotta as Catulus' sodalis should probably be taken as 'crony' (though he admits that the two might have been 'colleagues' in a priesthood); in fact, the word probably is technical for joint membership of a sodalitas; there were, moreover, various such bands, by no means all of them priesthoods; see further Powell on Sen. 45. In the note on 74 Mankin declares that Cicero was 25 when he defended Sextus Roscius, but he was probably, rather, 26; cf. Quint. 12.6.4. It would perhaps be more natural to take the asyndeton at the beginning of 75 as adversative, rather than either explicative or in an occupatio (the possibilities raised by Mankin). Mankin suggests that the alliteration 'tantum tribuisse . . . temporis' (85) may express Crasssus' '(feigned?) annoyance', for which he cites Skutsch on Enn. An. 104 ('o Tite, tute, Tati . . .'), who, in turn, cites Oedipus' famous alliterative reproach to Tiresias (Soph. OT 371). Perhaps, however, with its quasi-spitting, the alliterative t expresses (in Ennius and the OT) not annoyance but contempt. Our passage, on the other hand, is likely to exhibit what Skutsch (ad loc.) calls 'natural' alliteration (there is no other hint in context of annoyance on Crassus' part).

Section 110 is difficult. Surely the philosophers continue as subject (from 'dicunt' in 109); no need to follow Brown in inserting isti, which would, in any case, not sufficiently clarify a change of subject. It is the philosophers who are striving to 'recover [their] lost possession', i.e. the hypotheseis ('illud alterum genus'), to which they cling by a fringe (lacinia), as described in the sequel. The only change to the transmitted text needed in this section, besides those made by Kumaniecki, is that of alterum (before uero) to alii (sc. philosophi, in contrast to Philo); some reader/scribe evidently thought a contrast to the preceding 'alterum genus' was needed, but in fact the topic continues to be the hypotheseis (as treated by most philosophers).

The gen. after plus (131) is usually described as a partitive, rather than 'defining/of the rubric' (cf. Gildersleeve-Lodge §369). The notable encomium of Epaminondas ('haud scio an summum uirum unum omnis Graeciae': 139) passes, surprisingly, without comment (reference might have been made to Tusc. 2.59). It is not clear why Mankin thinks the verb in the phrase 'uerba nouantur' otiose (on 154); cf. 152 'uerbum . . . nouatum'. On 157 Mankin describes Mur. 35 and Planc. 15 as 'two notable descriptions of the Comitia Centuriata'; but in both passages Cicero speaks merely of comitia, and in the latter passage the comitia tributa are the relevant body (since election to the curule aedileship is at issue in the case). The ista in the phrase 'in uulgari ista disciplina' (188) seems more likely to be contemptuous than 'refer[ring] to the whole company', as Mankin supposes. Again hac in the phrase 'inflexione hac forti ac uirili' (220) perhaps has first-person deixis in the sense 'used by us orators' (as opposed to actors), rather than, as Mankin suggests, that Crassus 'has stood up . . . to demonstrate'. In the same section, perhaps manus etc. should be taken as in apposition with gestus, rather than as a case of anacoluthon. On 226 Mankin seems to confuse Stroebel, the editor of Inv., with Marx, the editor of Rhet. Her.

If I have dwelt overlong, more recensentium, on small points of disagreement, it should be stated emphatically that this is a most welcome addition to the green-and-yellow series. I expect to return to it again and again for help in interpreting this difficult text.


1.   BMCR 1996.7.19 (C. Bannon).
2.   When Mankin speaks of Cicero being 'quaestor for Sicily, where he collected evidence for the orations (the Verrines) that would set him on the path to the consulate' (p.28), the unwary may conclude that Cicero gathered his evidence against Verres while he was quaestor in 75, but, in fact, by that date Verres had not yet served as governor.
3.   Cf. R. Ku+hner and C. Stegmann, Ausf├╝hrliche Grammatik der lateinischen Sprache, II (Hannover, 1966) 60; see also 60, where 'quiuis qui minus' could also be alternative readings juxtaposed in L.
4.   I cannot argue the point in due detail here, but cf. M. McCall, Ancient rhetorical theories of simile and comparison (Cambridge, MA, 1969) 107-11; D. Innes, 'Cicero on tropes', Rhetorica 6 (1988): 316.

No comments:

Post a Comment