Sometime in the seventh or eighth century ad an anonymous Latin grammarian working in France produced an alphabetical list of nouns about whose genders his contemporaries might have been uncertain, with their genders. It included entries such as Corbes gen. fm., ut Cato: ‘corbes messorias’, ‘Corbes (‘baskets’) is feminine, as Cato says corbes messorias’ (entry C23), and Cubitum corporis gen. mas., mensure¸ autem gen. neu., ‘Cubitum (‘elbow’, ‘cubit’) as part of the body is masculine, but as a measurement it is neuter’ (entry C24).
Who would want to read this work today, and why? That is not a rhetorical question; it is at the heart of any evaluation of the book under review, which offers a new edition and study of the work based on a novel answer to this question. Previous editors have catered for readers interested in the ancient form of the Latin language and in traditional grammatical doctrine, on which the writer evidently drew extensively. Such readers are in fact interested in this work; for example, entry C24 is quoted in the Thesaurus Linguae Latinae as part of a discussion of grammarians’ views on the gender of cubitus/um (s.v. 1274.32–4). Consideration of their needs led previous editors to try to reconstruct a coherent, sensible treatise and present it in correct, intelligible Latin.
But another group of potential readers was arguably badly served by those coherent, corrected editions: scholars interested in late and medieval Latin and in the linguistic understanding current in Merovingian France. Merovingian scholars, working before the Carolingian Renaissance, had limited access to ancient texts and little training in writing Latin to the classical standard. And as one can see from the manuscripts of this work (which are relatively close in time to the original author and frequently agree on non-standard features), the work as originally composed was far less intelligible and correct than the version presented in previous editions. S.Y. therefore tries to reconstruct the treatise as its author produced it, complete with abbreviations, misspellings and numerous errors of Latin, of grammatical doctrine and of quotation. The result is that despite using exactly the same manuscript sources as her predecessors, she often produces strikingly different readings. For example, her entry C17 follows the manuscripts to read Clemam, alli ‘ciemam’, ut Volumnius: ‘stridenti dabitur patella ciem[e]ae’, ‘Clemam (name of a plant, fem. acc.), but others call it ciemam, as Volumnius says stridenti dabitur patella ciemae.’
Clema is an extremely rare word, and ciema (otherwise unattested) could only be a medieval variant of clema, not an ancient one. If these are the original author's words, he must have used corrupt texts. And we have no independent evidence for the fragment of Volumnius, so perhaps its ciemae was a corruption not of clemae but of something else. Previous editors have therefore emended to cyma ‘cabbage sprout’, a more common word attested both as third-declension neuter and first-declension feminine. H. Keil (Grammatici Latini V.574.1) rendered this entry as Cyma, alii cymam, ut Volumnius ‘stridentis dabitur patella cymae’, ‘Cyma (‘sprout’, neut. acc.), but others use cymam (‘sprout’, fem. acc.), as Volumnius says stridentis dabitur patella cymae.’
The most recent previous editor of this work, F. Glorie (Corpus Christianorum series Latina 133A, p. 762.117–19), reconstructed this passage as: Cyma <generis neutri, et «cymata»>; – alii «cyma» <generis feminini, et «cymas»,> ut Volumnius: «stridentis dabitur patella cymae», ‘Cyma (‘sprout’) is neuter, and the plural is cymata; others make cyma feminine, and the (acc.) plural cymas, as Volumnius says stridentis dabitur patella cymae.’ This is lovely, but is it what the Merovingian author wrote? Such entries indicate that S.Y. is right to say that Glorie's edition is too speculative, and they suggest that her own edition is a reaction to it. It is a well-argued reaction: in four pages of commentary on this entry (pp. 114–17) S.Y. makes a good case against the emendation to cyma, which has been used in one form or another by all editors since V. Le Clerc in 1849.
Another traditional emendation is rejected in S.Y.'s entries C29 and C30, which she prints as: Cavum se per se dicas gen. neu. est. Caverne¸ autem gen. fem. scribuntur, ut ‘cavi parietes, cavae speluncae, ample cavernae’. Notandum quod ‘parietes’ gen. mas. dixit, ‘Cavum (‘cave’) is neuter if you say it by itself. Cavernae (‘caverns’), however, are written in the feminine, like cavi parietes, cavae speluncae, amplae cavernae. It should be noted that he used parietes in the masculine.’ This passage appears in the manuscripts as a single entry, and previous editors have connected its parts by adding cavae. Keil's version (Grammatici Latini V.574.20–3; similarly Glorie, p. 764.153–6) is Cavum, si per se dicas, generis neutri est; cavae cavernae (Verg. Aen. II 52) autem generis feminini scribuntur, ut cavi parietes, cavae speluncae, amplae cavernae. notandum quod parietes genere masculino dixerunt, ‘Cavum (‘cave’) is neuter if you say it by itself, but cavae cavernae (‘hollow caverns’) are written in the feminine, like (other adjectival occurrences such as) cavi parietes (‘hollow walls’), cavae speluncae (‘hollow caves’), amplae cavernae (‘large caverns’). It should be noted that they used parietes in the masculine.’ This makes far more sense than S.Y.'s version and explains why the manuscripts treat this passage as a single entry; of course that is no proof that it is what the author wrote, but this time the commentary (pp. 135–6) does not really discuss the pros and cons of the emendation.
Occasionally the editions differ even on the central point of the entry, the gender. S.Y.'s entry C33 reads Callum gen. mas., ut Cicero: ‘ipse labor callum obduxit’, ‘Callum (‘hard skin’, acc.) is masculine, as Cicero says ipse labor callum obduxit.’ Here the quotation does not prove anything about the word's gender, and that is because it is a misquotation. Cicero (Tusc. 2.36) wrote ipse labor quasi callum quoddam obducit dolori, ‘work itself provides a certain insensitivity to pain’, treating callum as neuter, and other grammarians used this quotation as an example of use of the word (attested in both genders) as neuter. Previous editors have therefore emended the text; for example, Keil (Grammatici Latini V.575.3) says Callum generis neutri, ut Cicero ‘ipse labor callum obduxit’, ‘Callum is neuter, as Cicero says ipse labor callum obduxit.’ These previous editors are no doubt right about what was originally in the grammatical tradition on which the author of this text drew, but S.Y. (pp. 141–2) makes a solid case that the corruption/confusion does not post-date the composition of this text: the author wrote ‘masculine’ rather than ‘neuter’.
As these examples suggest, S.Y.'s detailed commentary (419 pages, discussing a text of 44 pages including generous apparatus) is invaluable; it offers the only thorough study of this work ever undertaken. The introduction (76 pages; in Italian, like the commentary) is also excellent, being much clearer and more helpful than anything previously written on this work. Whether the edition will be much used is more doubtful; I am not sure that the scholarly world is quite ready for such a novel approach, though perhaps soon we will be. A translation is not provided, presumably because, since the meaning of entries is often genuinely doubtful, discussing their meaning in the commentary instead of offering a translation allows for more nuanced interpretations. Despite the justice of that argument, the lack of a translation makes the contrast between Keil's easily intelligible text and S.Y.'s Merovingian Latin even starker. But the difficulties are far from insuperable, because S.Y.'s Italian is very clear, well written and to the point.
