This well-produced book, the title of which may seem to be a little misleading, doesn’t contain a selection of epigrams from the ‘Anthologia Latina’, but an already existing collection formed by #90-197 in the first part of A. Riese’s ‘Anthologia Latina’ (Teubner, Leipzig, 1894 2, section ‘carmina codicis Parisini 10318 olim Salmasiani’, book XIV in an earlier anthology of XXIV books, according to Riese) = #78-188 in the 1982 Teubner edition of D. R. Shackleton Bailey, whose numeration Kay adopts. His careful, cautious and well-informed introduction (pp. 1-25) deals succinctly with the following aspects of the collection, over 650 lines of Latin epigrammatic poetry written on a number of themes and occasionally using Vandal prosody, and more rarely Vandal grammar:1 authorship (unknown; Kay thinks all the poems are likely to have been written by the author of the metrical ‘praefatio’, who he thinks is likely to be responsible for the collection2), nature of the epigrams (none likely to be inscriptional) and arrangement,3 date (the sixth century is plausible) and origin (Vandal Africa), historical and literary context, transmission of the text (four main mss., all derived from a common archetype, the earliest being the famous ‘Salmasianus’ [=
The Latin text stands on pp. 39-64 without apparatus criticus or translation, which are to be found in the commentary of each poem: the translation stands at the very beginning of the commentary and the apparatus is found in the lemmata of the commentary. This I find both strange and unpractical: see the unclear six and a half lines of apparatus criticus which form the lemma of 112,4 on p. 190. A clear and properly placed apparatus would have been easy to compile; true, it should then have been more complete than it is. That would have been no bad thing; thus, for example, if the reader had known that 171,6 ‘rabidam’ (sc. ‘gulam’) stands as ‘rauidam’ in A, he could have more easily guessed that the right reading is probably ‘auidam’ (Baehrens), for that is the adjective which one expects in a line about the maw of a cat which eats a magpie: ‘nam claudunt auidam cornea labra gulam’.
Each poem or coherent group of poems has its own introduction, in which Kay sheds light on the main issues raised by the poem he studies, giving plentiful, accurate and up-to-date information about a great range of subjects, cultural or (to a lesser degree) grammatical. No one who has used Kay’s commentary on Martial XI (London, 1985) will be surprised by his impressive erudition in every possible kind of ‘Realia’, ranging from the ink and sex of the cuttlefish (96) to the classical ancestor of backgammon (182-185, equipped with an introduction of nearly four pages, in addition to the commentary itself, about the game). One is all the more grateful to Kay as he uses this erudition to illuminate the social, cultural, historical and sometimes prosopographical background of so many poems. Reading Kay’s enlightening commentary gives a glimpse of everyday life in (presumably) sixth century Carthage. Moreover each epigram is carefully set in the relevant literary tradition. Kay is also keen on detecting word-plays and nuances.4 In spite of his exertions and frequent success in explaining ‘prima facie’ obscure epigrams, some items of the collection remain quite mysterious and will repay further study: see e. g. the scoptic poem #180 about a dwarf called Bumbulus or even #128, which Kay thinks is about a celestial globe. The part of the commentary which deals with the linguistic and critical issues is good, but not as good as the antiquarian part (this was already the case in Kay’s commentary on Martial XI). Before commenting on the Latin text and the linguistic and critical commentary, it can be said that the translations of these sometimes difficult epigrams are very useful and generally accurate. Here are some cases where the translation and occasionally the commentary seem to me to be at fault: 82,1 (about the Judgement of Solomon) ‘prolemque negando’, “by denying her child”. I think ‘negando’ is here intransitive and I take ‘prolem’ with the following ‘conseruat’ only. — 99,8 ‘cum’ “now that” (see 106,22 and 24; 161,4). — 113,4 ‘alterno tingere membra lacu’ “to dip your limbs in another pool”. If Kay had seen the meaning of ‘alterno’, he would not have impugned Shackleton Bailey’s correction ‘hic’ for ‘hinc’ at the beginning of the line. — 116,9 ‘Solus. . . probas’ “you are all that is needed to prove”. For ‘solus’ = “especially”, see my note at Val. Fl. 8,143. — 120,10 ‘uatem te poterat reddere ligneum’ “was able to turn you into the wooden poet”. I understand “would have been able to turn you, who are wooden, into a poet”. This mistake is part of a larger misunderstanding: in spite of ‘lusit’ and ‘struxit’, Kay connects ‘te’ with ll. 11-12, which must rather be referred to the subject of ‘poterat’. — 128,1 ‘sinuatur’ “is encircled”. The meaning of the line is obscure to me. — 137,8 (‘adterit adsiduo pene fututor’) ‘hebes’ “with dulled sensation”, rather “till exhaustion” or the like. — 166,8 ‘grauidus bellis’ (the Trojan horse) “pregnant with war”, rather “full of warriors”, according to a not infrequent poetical usage of ‘bellum’. — 167,3 ‘uotum. . . meretur’ “he gets what he desired”. That is the expected meaning, for which the Latin would be ‘potitur’ (with attested accusative and long second syllable). — 172,3 ‘concordant multa’ (adv.) “match. . . exactly”, that is ‘cuncta’, a conjecture which Kay rejects. — 180,3 ‘longis … in armis’ “amidst long spears”, rather “armed with …” (cf. Juv. 13,168 quoted by Kay himself). — 188,11 ‘metarum tendunt circumdare cursibus orbes’ “they strive to complete orbits with circuits of the turning posts”, rather construe ‘metarum’ and ‘orbes’ together, “run round the circular turning posts”, if the text is sound.
The Latin text printed by Kay may well be the best available, better than those of Riese and Shackleton Bailey, whose text and conjectures Kay sometimes successfully challenges (cf. 102,8; 173,5), sometimes adopts — generally rightly, a conspicuous exception being 180,3, where sed ‘ratio est, si stas’ (‘mixtus’ A) is poor (read ‘sed ratio est : latitas’? in any case ‘sed ratio est’ must be followed by a juxtaposed sentence, as Watt saw). But Kay’s text is not consistently superior; better judgement and more expert verbal scholarship would have ensured a better text. Kay has left the next editor5 a lot to do. His textual notes strike a note of self-confidence which I find sometimes unjustified. Thus, though translating ‘cuncta’, he defends 172,3 ‘multa’ against ‘cuncta’, arguing that “replacement of ‘cuncta’ by ‘multa’ in the mss would be less likely than the reverse, and the inference must be that they preserve the truth”. The affirmation is fanciful and the inference wrong. If he had been aware how easily forms of ‘pretium’ and ‘species’ can be interchanged (cf. Statius Silv. 5,3,115 ‘specieque’ M/’pretioque’ Saenger, rightly), he would perhaps not have defended with inadequate “parallels” the transmitted text in 122,1 ‘Haec (sc. mala) poterant celeres pretio tardare puellas’ against Watt’s ‘specie’. The same remark applies to 188,7, where Schrader’s ‘carceribus’ for ‘cardinibus’ is probably right (as ‘cardine’ for ‘carcere’ at Val. Fl. 3,499).
Kay seems to me to defend not a few corrupt or suspect readings through strained exegesis or inadequate arguments and parallels: 81,8 ‘placuisse deo’ “has pleased God” (quoting inter alia CE 2016,4, where ‘placuit deo’ = “God decided to”); 88,3 ‘tamen’ (against Timpanaro’s ‘tantum’); 91,6 ‘hunc. . . hunc’ (against Munari’s ‘nunc. . .nunc’); 93,6 ‘factum’ (against L. Mller’s ‘furtum’); 94,6 ‘saucia’; 103,7 ‘reparator’; 106,10 (about May personified) ‘expoliat pulchris florea senta rosis’ (extensive corruption? I suggest ‘extollit pulchras per loca senta rosas’); 117,4 ‘cum’ (against Baehrens’s ‘nec’); 121,5 ‘distinguit’; 141,2 ‘ora’ (against Baehrens’s ‘corda’) and 3 ‘diffundat’ (against Shackleton Bailey’s ‘defundat’) ; 143,4 ‘lumina tendat’; 145,4 ‘infami’ (‘infirma’ and ‘insana’ have been conjectured); 150,1 ‘pretium si poscere nosses’ (‘noscere posses’ [‘posses’ Zurli]?); 160,3 ‘unumquemque suum referunt pomuscula sucum’ (against Baehrens’ and Hagen’s ‘unum quaeque’, quoting Verg. Ecl. 7,54, where he seems to ignore that ‘quaeque’ has been suspected); 157,1 ‘Dispersit. . . tempora’ (I suggest ‘Diffindit’, cf. 156,2 ‘tempora fissa’; Verg. Aen. 9,589, for ‘tempora’ is not ‘cerebrum’); 169,3 ‘hic’ (against Riese’s ‘his’); 173,3 ‘quem’ (against Courtney’s ‘qui’); 188,10 ‘cogitur’ (‘nititur’ would give good sense). It is only fair to say that in the above-mentioned cases Kay duly quotes rejected conjectures. More or less important textual difficulties may completely escape him: 87,1-2 (on the Chimaera) ‘Ore leo tergoque caper’ (similar passages quoted by Kay suggest ‘Prima leo medioque caper’, ‘tergoque’ being perhaps due to ‘TERGEmino’ in the following line); 100,8 (on a pantomime actor) ‘inlustrat uerum’ (‘uersum’ Baehrens; Kay’s note explains everything except ‘uerum’); 107,4 (about Achilles’ heel) ‘in membris tincti dant sibi fata locum’ (that’s exactly what fates don’t do; Oudendorp proposed ‘non tincti membri ; I suggest ‘non mersum tincti’ or ‘inmunem tincti’); 118,2 ‘infami’ (‘infamis’ Baehrens); 148,3 ‘ut iussum nosset tolerare magistri’ (read ‘ut iussum posset celerare magistri’, “so that he might quickly execute his master’s command”); 151,1 ‘animo’; 156,2 ‘disco tempora fissa gerens’ (‘gemens’?); 181,4 ‘Discordat multum’ (‘minime’? Maehly and Baehrens, who corrected l. 3, saw the illogicality of the passage, which escapes Kay). He adopts some “economical” conjectures which produce gibberish: thus his text of 112,4 is ‘haec (sc. balnea) reddi poterunt, Phoebe, uapora tuis (sc. flammis)’; ‘haec radii tepeant, Phoebe, uapore tui’ would I think be a more plausible wording and not too remote from the paradosis. Other examples are 139,5-6 and 184,1 (where he mistakes the conjecture ‘talis’ > ‘talus’ for the otiose ‘talis’ “such”). He seems to me to have chosen the wrong variant ‘pectore’ for ‘corpore’ at 93,3 (I suggest ‘quot, licet. . . colligit’ in ll. 3-4; Kay’s corrected text is flat and calls for such a word as ‘quot’).
Most of Kay’s emendations (84,1 ‘templo’; 101,3 ‘[quam] *superaërius'”[how] incredibly high”; 109,6 ‘actoremque’; 168,5 ‘Vitae’, which he doesn’t put into his text, instead of ‘paucis’; 174,3 and 5, interchanging ‘cum’ and ‘dum’ because ‘cum’ would be ambiguous in ‘peterent cum mille carinis | tangeret et classis etc.’) do not appeal to me, but 88,6 ‘si’; 124,1 ‘mensam’ and 131,2 ‘e Ioue’ (‘de Ioue’ would avoid hiatus) may be right. Among Reeve’s conjectures published by Kay ‘eduri’ 120,9 and ‘quondam Quintilis’ 106,13 (building on Shackleton Bailey’s nice correction) are quite good, but his ‘templi’ 84,1, though less offensive than Kay’s ‘templo’, is wrong (for ‘piis. . . templis’, see TLL X,1. 2241,9 sqq.). Though the linguistic commentary is a business-like piece of work and contain many helpful and learned notes, some difficult passages (90,2; 148,4; 171,7 ‘uitalis semita cessit’ [‘cessat’ Riese]) and phrases (103,5 ‘Thebarum moenia saepsit’ “the meaning is ‘Thebas moenibus saepsit'” [internal object accusative]) deserved more thorough comments.
Four useful indexes (“names and places”, “authors and works”, “Latin words”, “general index”) follow the commentary. Some miscellaneous criticism or suggestions: say “Saumaise”, not “de Saumaise” (passim), as one says “Wilamowitz”; there is no such scholar as “Mercer” (p. 139): “Mercerus” or “Mercier” is correct. — 96,2 ‘candens piceum sepia claudit onus’: perhaps ‘condit’, a word often interchanged with the corresponding form of ‘claudo’. — 123,1 (rightly marked as corrupt by Kay) perhaps ‘His cessit Veneri praelatae gloria (‘gloria’ Baehrens) formae’ “thanks to these apples Venus got the glory of being preferred as most beautiful”; 127,4 the expected word is ‘tortor’ (‘ignoret
All in all classicists will be very grateful to Kay’s pioneering commentary and especially to the antiquarian erudition which enables him to illuminate this often neglected corner of Latin poetry.
Notes
1. The literary quality of these often interesting epigrams ranges (in my judgement) from very low to good. Note the beautiful line 157,4 ‘semper Apollineus flore resurgit amor’ (about the hyacinth).
2. The ‘aliter’ which introduces the epigrams following the first of two or more dealing with the same subject seems to have led Riese, p. xxi, to think that they were ‘quasi e certamine poetico orta’. This idea, known to the students of the Greek Anthology, might have been mentioned and discussed. Further work on the genesis of the collection and its components seems to be needed.
3. Note that the first epigrams which follow the ‘praefatio’ are Christian and that the last of the collection allegorically equates the circus to the sky. The rest is or seems to be profane, sometimes very profane.
4. I suspect there is an extensive double-entendre in 97,5-6 (‘De eunucho’), ‘Coniugibus cautis placita est monstrosa uoluptas; | fidus enim est custos qui sine teste datur’ 1) “cautious wives like monstrous pleasure (given by a eunuch), for they can rely on a guard provided without a witness/deprived of testicles”; 2) “cautious husbands like (such) a monstrous favori (= ‘deliciae’), for they can rely on a guard (for their wifes) deprived of testicles”. It would perhaps be silly to think that the author also played on ‘custos’ and Greek kusthos = ‘cunnus’ (for a pun on kusthos and ‘costos’, see Ausonius Epigr. 82,5 Green).
5. Apparently L. Zurli (cf. Kay, p. viii).