Scholia Reviews ns 6 (1997) 23.

E. Wolff (ed. and tr.), Dracontius: Oeuvres. Tome IV. Poèmes Profanes VI-X; Fragments. Paris, Les Belles Lettres, 1996. Pp. xv + 235. ISBN 2-251-01398-9. No price supplied.

Gregory Hays
Thesaurus Linguae Latinae, Munich.

'I haven't really been keeping up with the Dracontius bibliography,' a colleague lamented recently. To an outsider it might have sounded a trifle arch (rather like complaining about how rusty your Persian has gotten) but she had a point. Vandal Carthage's greatest poet has indeed been attracting more attention lately, and a new Budé edition is one indication of that. The Christian poems were edited by Claude Moussy and Colette Camus in two volumes (1985; 1988). The profane pieces have fallen to Moussy's students, Jean Bouquet and Étienne Wolff. Bouquet's vol. 3 (1995) contained the Orestes and Romulea I-V and an introduction written jointly with Wolff. The latter now offers us Romulea VI-IX, the De mensibus, and the De origine rosarum, plus two exiguous fragments.

I begin with the text. Aside from brief citations in florilegia, the Romulea are transmitted in a single late and untrustworthy manuscript (Naples, Bibl. nat. IV E 48, s. XV/XVI), while Mens. and Ros. are preserved only by quotation in Bernardino Corio's Historia di Milano. Recensio is thus over practically before it starts, and the editor's main task is emendatio. By my count, Wolff's text differs from Vollmer's 1905 MGH edition in 49 places -- not a great divergence given the state of the text. Leaving aside his own conjectures (of which more below), I count 14 cases where Wolff preserves the reading of N (or, once, obelizes) against Vollmer, and conversely 13 cases where he prefers to emend against Vollmer and the manuscript (four of these are cases where Vollmer had obelized). In 9 instances Wolff and Vollmer both reject the transmitted reading but accept different conjectures. (Two divergences -- 9.142 iram; 10.239 iugalis -- appear to be typographical errors). On the whole I should say the honors are about evenly split (in many cases there is not a great deal to choose), and most of Wolff's choices are at least defensible in the context of a reading edition like the Budé.

In two points, however, Wolff's edition shows itself significantly inferior both to Vollmer and to Diaz de Bustamante's 1978 edition. The first is the apparatus. This is generally accurate, though not flawless (at 10.279 read 'distnixerat n'; at 10.336 the emendation eiecto is incomprehensible because Wolff fails to report that electro stands in the margin). Unfortunately, it is also littered with junk (much of it contributed by the industrious Bährens), and inflated still further by absurd 'vote-counting' entries like 'indue N imbue Bücheler Duhn Gualandri Diaz Grillone.' Here Wolff would have done better to emulate the clarity and concision of his colleague Bouquet in volume 3.

The second problem is the excessive indulgence Wolff displays toward his own conjectures. No less than eleven of these are placed in the text (Rom. 7.5; 7.6; 7.35; 9.29; 10.71; 10.212 ter; 10.256; Mens. 3; Ros. 12); one (Rom. 9.201) appears in the apparatus. Of these, only quod at 10.71 seems to me an improvement. At 7.35 Naiades . . . Oreadas belongs in the apparatus, but the transposition is hard to explain and the false quantity in Oreadas is probably Dracontian. At 10.212 Wolff's cum makes petunt impossible to construe, his deletion of cui produces an implausible asyndeton, and arietis is unacceptable, not on its own account (ariet- scans as an anapest elsewhere in Dracontius), but because it makes the preceding pellis unmetrical. Wolff does not buy Vollmer's hypermetrical inaurata, but I am at least tempted: the epithet is standard, the metrical parallel at Aen. 6.602 a strong one, and the anomaly helps explain the corruption. At Mens. 3 the transmitted nives is impossible, but Wolff's nixus is hard to swallow. Perhaps in uvis? (Cf. Vollmer p. 441 for similar clausulae, p. 360 for the loose construction with in).
In turning to the translation and notes, one has to cope first with a problem of format. Wolff's thesis, on which the present work is based, apparently included a line-by-line commentary of the sort sensibly used by Moussy and Camus in the first two Budé volumes. In the work's metamorphosis from thesis to Budé, however, this commentary has been transformed into a set of discrete notes keyed to the French translation. Who gains by this? Not the general reader, since the notes are primarily aimed at Latinists -- nor the Latinist, who has an extra obstacle placed between himself and the notes. (Orientation is not exactly furthered by the Budé practice of arbitrarily printing some notes as footnotes and others as endnotes).

That said, the notes are generally worth the search. Wolff displays a thorough familiarity with Dracontius's corpus, and a subtle feel for his often eccentric Latin. He does not gloss over difficulties, and is not afraid to confess incomprehension when it is warranted. He is also good about drawing attention to echoes and adaptations of earlier poets, notably Statius and Claudian, but without exaggerating their significance (cf. the general analysis at vol. 3.57ff.). There is less discussion of Dracontius's mythological sources, but that is probably prudent, given the murkiness of the whole issue. The views of previous scholars are judiciously assessed throughout, and Wolff is careful to report plausible interpretations that differ from his own. The first two volumes were criticized for ignoring secondary work not written in French; Wolff does better than his predecessors here, though it is surprising to find no mention (e.g. in the discussion of mime on p. 189) of David Bright.[[1]]

A book like this stands or falls on detail, so I include some observations on individual points (in some cases references are to the corresponding note):

6. 59 erat cui castra voluptas] not 'il était dans le camp du plaisir,' but something like 'pleasure was his parade-ground' (for the conceit cf. Ninus in Phoenix of Colophon fr. 3 Powell 'whose sword was goblets and his spear a jug').
6. 70] populos = people, 'les gens.' This usage is poetic (cf. Ovid, Met. 6.179 with Bömer ad loc.) and common in later Latin.[[2]]
6. 118] adoptatae really must go with noctis.
8.59f. quis semita nulla tenetur / obvia dum veniunt] The phrasing is obscure, but probably 'against whose opposition no course can be maintained' rather than 'dont aucun chemin ne contient la progression'?
8.76 pastore propinquo] not 'à côté du berger' but 'at the shepherd's approach.'
8.285 mentes armabat in iras] The phrase should be taken in the obvious sense (cf. the Vergilian armare in proelia). Wolff's interpretation ('armait son âme contre sa colère'), though no doubt 'plus intéressant', is oversubtle and does not sort well with 291 below, where Telamon is explicitly described as succensus in ira.
8.537-9] The lines are not merely 'difficiles' but unintelligible; I suspect 537 is corrupt.
8.555] The Latin cannot possibly bear the construction Wolff puts on it (but what does it mean?).
8.558] Read deo?
8.562] The text is sound.[[3]]
8.577 ff.] Claudian Rapt. 3.263 ff. is the model, not just a parallel.
8.591] The cheeks are Antenor's, not Priam's.
8.628] I do not see how mors can be a caterva, even metaphorically. Read catervas with Duhn.
8.653] ThLL is right to gloss fugax here as inconstans; per castra should be construed with clade Pelasgum.
9.26] The lectio difficilior rule does not apply to emendations.
9.34] spectant is not 'l'équivalent de patiuntur' but equals exspectant -- this is the sort of treatment corpses can look forward to from Achilles.
9.216 veniat tantum pensabitur Hector] Not 'aussi cher que tu l'auras estimé' but concretely 'for a sum equivalent to his weight.'
9.221] Vollmer did not take Dardanides as a plural but as a generic singular (= ThLL's 'pro plur.').[[4]]
10.116] the peplum belongs to the lilies, not Cupid (for the metaphor cf. Pervigilium Veneris 20f.).
10.174f.] For the motif ('god's advent calms waves') cf. Kenney on Apuleius, Met. 4.31.4.
10.183] The mock-epic detail (cf., e.g., Iliad 1.46) was worth a note.
10.278 'Illo iam gressus,' dixit, 'convertite, tigres: estis opus, mea turba, deo'] Wolff has completely misunderstood this passage: Dionysus is inviting his followers to take a detour to Colchis, not to turn back. Translate 'Turn your steps in his [Cupid's] direction, tigers! a god requires your presence, my followers!'
10.505f.] digna is sarcastic and modifies corona ('the kind of crown she deserves').
10.543] Diaz's suggestion is impossible: the snakes in fr. 2 are figures on a flag.
10.577 male conceptis] not 'pour faire du mal' but simply 'unnaturally.'
10.588] Wolff misses the paradox ('Polynices's brother -- and foe').
Mens. 14] Wolff is probably right against Vollmer and Housman; for the word-order cf. 22 below.
Mens. 20] I suspect pastoral aesthetics, rather than sobriety, is at issue: 'the joyful peasantry is more becoming when stained with must.' But the expression is certainly awkward however taken.

The verdict on volume 4, then: the text marks no significant advance on Diaz, while Vollmer will remain the standard edition. But Wolff's extensive notes constitute a welcome addition to Dracontian scholarship: in effect, they constitute the first commentary on these difficult poems. And the translation will be invaluable to future interpreters, even (perhaps especially) in the places where it does not quite convince.


[[1]] D. Bright, The Miniature Epic in Vandal Africa (Norman, Oklahoma, 1987).

[[2]] See further H. Roensch, Collectanea Philologa (Bremen, 1891) 169f.; L. Callebat, Sermo Cotidianus dans le Métamorphoses d'Apulée (Caen, 1968), 169.

[[3]] On this construction cf. R. Renehan, Studies in Greek Texts (Gottingen, 1976) 60, with references.

[[4]] Cf. J. Wackernagel, Vorlesungen über Syntax vol. 1 (2d. ed.; Basel, 1950) 93f.