Bryn Mawr Classical Review

After many years of neglect, French-speaking scholarship has now repeatedly¹ grappled with what may be the most illegible piece of classical literature, one which nobody can read without a proper commentary and which even then makes very difficult reading. I imagine some scholars may eventually, after repeated perusal, be able to read and understand Pindar or Thucydides' Greek without any external help, but I doubt if that can ever happen with Lycophron's 1474 cryptic iambic trimeters.

However, there is evidence that this poem was read and explained in ancient schools, namely Statius' well-known line carmina Battiadae latebrasque Lycophronis atri (Siluae 5,3,157). Statius tells us that his father used to explain Lycophron's enigmas to the sons of the Roman elite. Hurst (p. XLV) wrongly quotes this line as 5,2,157 and fails to note that atri is a humanistic correction for the only authoritative manuscript's reading ari, instead of which arti has also been suggested (Baehrens). The use of both words is quite unusual. Hurst might have referred to B. Gibson's 2006 commentary on Siluae V, which has a good note on the passage (see also my forthcoming edition of the Siluae). Hurst, who has much to say that is interesting about Lycophron's obscurity, fails to notice that Statius' line is apparently the earliest testimony to the application of a word meaning "obscure" to Lycophron or his poem.

The literary part of the introduction, even if it is open to criticism, seems to me much better than the one dealing with the textual transmission, which raises serious issues. Hurst convincingly (in my judgment) argues for only one Lycophron, not two, and he fully discusses the views that have been suggested on the well-known final prophecy, reasonably concluding it is not interpolated. He has too little (I believe) to say on dialects and on metrics. I for one regret that he prefers (p. XXXVII) the reading ἄκρης (which I think is a mere Perseverationsfehler due to the preceding ἄρχῆς), rejects the variant ἄκρας and comments on the "jeu sur les sonorityés" introduced by what he considers an artificial ionicism. Nothing is proved by the fact that the phrase ἄκρης ἀρ' ἀρχῆς may have become standard among Byzantine writers (Eustath. Macremb., Christus Patiens). Hurst prints the transmitted reading πρύμνης ἀρ' ἄκρας l. 515 and ἐξ άκρας σκοπῆς l. 714: why not print the hyperionicism here as well? Is it because there is no variant? Then, when there is a variant, why not choose the one which is corroborated by the transmitted reading in 515 and 714? Oddly enough Hurst mentions Bachmann's correction ἄκρης in l. 515 but not in l. 714. Bachmann 1830's policy, criticized by Hermann (Opuscula, V, p. 235-236), is (I believe) utterly wrong but more consistent.

After the chapter on the textual transmission and the editor's principles there is a short and interesting literary essay the title of which is "Pourquoi l'Alexandra?" Hurst notes how much the recent revival of interest in Lycophron's poem is linked to a change of literary taste. He and others seem to think that Lycophron can be compared with other purposely obscure poets such as Mallarmé and Saint-John Perse. I venture to disagree, for I think Mallarmé and Saint-John Perse may be called obscure, but they are poets. Contrary to what many seem to think nowadays, obscurity by itself doesn't make good poetry or good literary criticism. We must be aware that Lycophron wanted his readers to ponder over his enigmas and understand them intellectually, while Mallarmé and Saint-John Perse's readers are supposed to enjoy their poetry without necessarily understanding its literal meaning. This difference seems to me to be essential. Hurst might have dwelt more extensively on the comic and scoptic elements in the poem (for example ll. 90 sq. on Paris are excellent and delightful scoptic poetry).

Hurst offers on p. XLIII-LV a sketch of the history, viz. the circulation, of the text. He then goes on to discuss the textual tradition. Apart from eight papyri, Hurst relies on nearly the same manuscripts as Scheer, who, in an 1879 paper, discerned two families, A (XIth c.), B (Xth c.) plus A's copy V (presumably XIth.), and C (A.D. 1282 for the part containing the poem), D (XIIIth c.), E (XIVth c.). Hurst adds "Scorialensis gr. R I 18" = M, a ms. written A.D. 1255 and brought to Hurst's notice by the late Irigoin. Hurst (p. LXXII) says it clearly belongs to the second family but shares some readings with what Hurst calls "la catégorie des deteriores de Scheer". He uses it in view of its age and (a strange reason!) its "probable origin". Hurst only sporadically quotes V and M. The description of the mss., which unusually follows the exposition of their division into two families, ends with these reassuring words (p. LXXV): "all the other mss. have been systematically collated by Antje Kolde. As I said before, no clue emerged which could have made it possible to challenge Scheer's classification". On p. XLII n. 1 the reader is informed that Hurst derives his knowledge of these mss. from a list of 136 rubrics "of varying import" established by the Pontifical Institute of Mediaeval Studies of Toronto.² In the same note Hurst writes that a history of the manuscript tradition would be interesting but is beyond the scope of this edition for which "we have been content with collating the mss. and all the published papyri in order to offer an up-to-date version of the poem". Nowhere in the introduction or the siglorum index does Hurst enumerate the despised deteriores, but a few deteriores are individually mentioned here and there in the apparatus, e.g. the mysterious "Seld. 17 (s. XIV)" (p. 82, l. 1438), which is quoted for a good reading.³ Elsewhere Hurst uses the symbol "d", though he implies that these mss. have been collated. If the deteriores are what their name means and Hurst's above-mentioned statements imply, how is it that P. Oxy. 4428 (IIIth. c.) offers a reading, κελτου (l. 189), which "supports the reading of the deteriores" (p. LX)? The truth may be that those deteriores are recentiores which may have retained wholly or partly true or traditional readings absent from the few mss. used by Hurst. The same papyrus confirms a very significant good reading formerly known through the Etymologicum Magnum, φάρωι (l. 154), to which the shallow τάφωι has been substituted in the mss. used by Hurst. The apparatus at l. 154 implies that the reading of the papyrus originates from the Etymological Magnum!

The page pertaining to the editor's principles (LXXVI) is packed with contradictions and more or less problematic statements. He says it would be vain to draw a stemma, "which no former editor has done", but proceeds with sketching it. After saying that he has relied on Lycophron's "manner" to establish the text and consequently has not mechanically followed any of the two families, he informs his reader that he has attempted to stick to the first family. Though the deteriores do not fare well with him, he preferred some of their readings as "more in conformity with Lycophron's poetics". "I ask (he says) the reader to consider that the resulting margin of error probably (!) does not exceed the margin which would result from the mechanical application of a bias towards one branch of the manuscript tradition".

A rich apparatus of testimonia has its place above the apparatus criticus. Both are far from being always clearly and carefully written. Critical units are not separated when they pertain to the same line and, since they are packed together, they too often strain the reader's eyes and his attention. Why do mss. BD not stand in the critical unit for l. 51? The apparatus of testimonia at l. 97 is very hard to unravel. The apparatus criticus at l. 1126 attributes to the unfortunate Hermann the monstrous conjecture νούνυμνον. One may wonder a while what "πεπασμέναι Scheer cf. Π" (l. 1138) means, till one realizes that Π, which should refer to a papyrus, is all too often a misprint for "P" = antiquior paraphrasis. "478 post 479 poni monuit Canterus" and "1184 post 1185 collocari monet Canter" are bewildering. The reading which is printed at l. 1295 is not the one which comes first in the apparatus. Wilamowitz' conjecture at ll. 1436-1437, divided into two critical units with the omission of one fundamental word, is difficult to guess.

Fortunately, the establishment of the text, though not very satisfactory, is better than one might have expected after reading page LXXVI. The printed reading at l. 369 is ὡς φθιτῶν θέμις, but the truth lies in ms. B, ἡ θέμις φθιτῶν, which points to ἥ θέμις φθιτῶν, Kinkel's correction, mentioned by Hurst and confirmed by the Homeric phrase which is first to be found in Il. 9.134. The imperfect (ms. E and recentior paraphrasis) is necessary at l. 1463. Hurst, who prints the ugly prodelision ἡ 'χερουσία at l. 90, keeps the very problematic article l. 93 without noticing the difficulty of τριπλαῖς, which he translates "des trois belles". Hermann very attractively substituted to the article the necessary θεαῖς, which was lost before the next word beginning with theta. The article τόν is a mere stop-gap. On p. LXXVI-LXXVII Hurst claims to free his reader from the editor's arbitrariness through enriching the apparatus with many conjectures: actually they are too few. Still one must be happy to have them since they are not so rarely very probable or even certain (the future, conjectured by Holzinger at l. 1467, deserves to stand in the text). Lines 365-367 as printed by Hurst are neither Greek nor meaningful and the note on p. 155 fails to grasp the difficulty:

...μυρίων τέκνων
Ἑλλὰς στενάξει πᾶσα τοὺς κενοὺς τάφους
οὐκ ὀστοθήκαις χοιράδων ἐφημένους.

Even if one supposes a lacuna after l. 366, line 367 remains problematic: "ils ne sont point gisant dans leurs cercueils de récifs" (Hurst) is, even by Lycophron's standards, absurd: one would expect οὐκ ὀστοθήκαις χοιράσιν δ' ἐφημένους. Hurst attributes the essential δ' to Bachmann 1830, but I find it in the text of Meursius 1597, in that of my 1601 copy of Canter's edition, in the text of Kinkel's 1880 and Scheer's 1881 editions without anything in the apparatus indicating that it has been added (nor does Bachmann say he has added it): is it only the vulgate or is it also the transmitted reading or at least that of Scheer's primary mss.? Canter translates non urnis sed scopulis insidentia. The reading χοιράδων (a blatant solecism, if taken with the participle) is very probably a mistaken anticipation of the same form l. 373. Lycophron's Alexandra may belong to those classical texts which can still yield easy corrections: see line 1295:

ἔχθρας δὲ πυρσὸν ἦιραν ἠπείροις διπλαῖς.

Hurst prints Wilamowitz' ἦραν, "ils ont levé le flambeau guerrier (?) entre deux continents", but it is clear to me that what we need is ἧψαν. The phrase πυρσὸν ἧψαν, figurately used, is already found in Pindar, Isthm. 3 (4),61 (43) and illustrated by such Latin parallels as Florus, epit. 3,14,1, primam certaminum facem Gracchus accendit. How can the genitive stand in the second half of l. 1361? I think the last two words must be a local dative. Neither (I believe) can the accusative stand with a verb meaning "sleep in" at l. 1354: ...ἔνθα Τυφῶνος δάμαρ κευθμῶνος αἰνόλεκτρον ἐνδαύει μυχόν.

I suggest αἰνόλεκτρος ἐνδαύει μυχῶι, "where Typhon's dreadfully wedded (cf. 820) or dreadfully bedded wife sleeps in the recess of her hiding place". The correction of the hapax ἐνδαύει into ἐνναίει would solve the grammatical problem but seems implausible. Hurst's conjecture in l. 1438 implies a strained interpretation of the word βουτρόφοις and (more importantly) leaves 1436-1437 in a very problematic condition. The corruption appears to be concentrated in 1436-1437: line 1438 is very probably immune and Hurst's conjecture is no progress on former suggestions. Other proposals by Hurst may be seen at l. 520 (an obvious change of a majuscule into a minuscule, already in Canter's text), 869 (a barely apt neologism), 1212 (not better than the transmitted text).

The translation, which is understandably not more elegant than the difficult Greek, is useful but might have been much more helpful if it had been more accurate and closer to the original. More or less essential Greek words are left untranslated (e. g. ἄρτι l. 16; μυχούς l. 44; ταρχυθεῖσαν l. 369; αἰγίλιψ l. 1325; πάλαι l. 1326; μακρά l. 1451) or are mistranslated (e. g. ἄφλαστα "poupe" l. 26; "beaupré" with "poupe" for κόρυμβα l. 295; ἀρχαῖς "forces" l. 1437, through confusion with ἀλκαῖς?). The reader who faces the difficult ἔλυσε χρησμῶν στόμα (l. 4) is not helped by the extremely loose rendering "lâché le flot de ses oracles": why not "freed her mouth from its oracles"? For that seems to be the meaning of the Greek. Lines 1356-1358 are misconstrued (the first word goes with the last: they form a nice "Sperrung").

Readers of the translation will find extremely useful the commentary which runs more than 200 pages and is mainly due to Antje Kolde (cf. p. 85 n. 1): it provides them with almost all the mythological and literary information they need. Scholars, however, may miss a commentary adequately dealing with grammatical, lexicological and textual issues. There is a short bibliography at the end of the introduction. It is not very accurate: the illustrious K. O. Müller becomes C. G. Müller: his illustrious adversary G. Hermann becomes H. Hermann and his review of Bachmann's edition is quoted "H. Hermann (1834), Opuscula, Leipzig". Oddly enough the reader is informed p. XCII that the abbreviations of Greek authors' names are taken from "le plus récent dictionnaire grec, à savoir celui de Franco Montanari, Vocabolario della lingua Greca, Milano, (Loescher), 1^re édition 1995, 2^e édition 2004": what about the DGE? Montanari's dictionary is not used in France and it cannot be considered a reference work. References to secondary literature are not always clear or consistent: contrast e. g. "PCG" with "Austin-Kassel" ("Kassel-Austin" is usual): "Wilam. Comment. 7 cit." at l. 1436 with "Wilamowitz (1883) 7" at l. 1437. It seems useless to prefix "L." to every Byzantine lexicon. The book ends with an index of proper names in French.

However useful the literary introduction, the translation and the commentary may be, the recensio is clearly insufficient and the emendatio, the apparatus and even the translation fall short of expectations. A modern edition of the text with full, clear and accurate apparatus based on further investigation of the manuscript tradition and displaying critical acumen is still badly needed: we also need, after Holzinger's 1895 commentary, a modern one aiming at and reaching high philological standards. But this book fills a gap in the Budé series and, even if one may still look for a more scholarly achievement, one may welcome another⁴ edition of this abstruse poem from the pen of Hurst, who has been busy with Lycophron for decades and is one of the few living scholars who can claim to be very conversant with him.

Notes:

1. Hurst mentions a French literary translation by P. Quignard (1971) and annotated translations by G. Lambin (2005), P. Hummel (2006), C. Chauvin and Chr. Cusset (2008).

2. The list is available on http://pinakes.irht.cnrs.fr and has 142 items.

3. Did Hurst mean Oxford, Bodl. Libr., Arch. Seld. B. 18? There is no Lycophron in Seld. B. 17! Bachmann 1830 (from the collations of J. Potter) already attributed the above-mentioned reading to a "Seld." "Cod. Pal. 40" (l. 1199), "Neapolitanus I E 22" (l. 1403), "Auctar. T I 14" (l. 1438) may be worth checking.

4. The bibliography mentions an Italian edition (1991, with M. Fusillo and G. Paduano) with translation and notes and one in modern Greek (2004).