Bryn Mawr Classical Review

This work fills a major gap in the scholarly literature on Latin grammar by providing a comprehensive and largely reliable dictionary of Latin grammatical terminology. It represents the fruits of a painstaking lexicographical documentation and description, and will be a valuable tool for students of Latin grammatical literature. However, it could be improved in a number of respects. The reviewer hopes that the following critique and detailed corrections will be of assistance to serious users of this work.

This review opens with an overall presentation of the work analysed here (1.). The next paragraphs focus on: the domain covered (2.); lexicographical aspects, i.e., micro-organization of the lexicon (3.) and translation of the source-terms (4.); philological aspects (5.); integration of (modern) literature (6.). Then follow a list of (minor) corrections (7.) and a conclusion (8.)

1. The author of the work under review presents it in the following terms: 'The subject of the present study is the terminology of the Latin grammarians. This represents a variety of technical language created and developed for the purposes of grammatical description and analysis' (p. XVIII). As such, this Lexicon fills a real gap in the historiography of ancient linguistics and in the lexicography of grammatical terminology. While for Greek grammatical terminology we have V. Bécares Botas's Diccionario de terminologyía gramatical griega (Salamanca, 1985; primarily based on the texts in G. Uhlig's Grammatici Graeci and basically intended as a 'glossing dictionary'), for Latin we had only the (Latin) dissertation by L. Job, De grammaticis vocabulis apud Latinos (Paris, 1893), which was not conceived as a lexicographical work, and which, in addition to covering only a selection of grammatical terms, is outdated in its documentation and by its historiographical information; fortunately, since 1990 we have also had the useful Index grammaticus of Lomanto and Marinone,¹ which renders excellent services as a concordance.

Schad's Lexicon consists, apart from the 'Acknowledgements' (p. XIII) and a list of 'Abbreviations' (pp. XV-XVI), of a rather substantial 'Introduction' (pp. XVII-XXIV) and 'The Lexicon' (pp. 1-430), followed by a list of 'References' (pp. 431-438) and three indices: 'English-Latin' (pp. 439-446), 'Greek-Latin' (pp. 447-452) and a list of 'Prefixed Verbs' (p. 453). Within the Lexicon five extremely useful tables are included: 'Aptotus and the formae casuales' (p. 39), 'Gerund and Supine' (p. 188), 'Impersonalia' (p. 203), 'Neuter/neutralis' (p. 263), 'Intransitive verbs' (p. 264). The Lexicon itself has about 1430 entries. The entries are given in the nominative singular when they are nouns, in the nominative masculine singular in the case of adjectives, and in the first person singular of the present indicative in the case of verbs.

2. Schad has titled her work a 'lexicon' of 'Latin grammatical terminology'. Each of these constituents deserves a careful explanation and justification; Schad's introduction is rather implicit on this matter.

(1) First, the work is more than a lexicon in the sense of 'lexical inventory'; it is intended as a (systematic) dictionary and, to some extent, as a reference-list (if not a concordance) of attestations, with rather extensive quotations, which allow the reader to check how a particular term is used in a particular context (or in a variety of contexts). As a specialised dictionary, Schad's lexicographical product could have benefited to a larger extent from the state-of-the-art in lexicography and metalexicography;² I will limit myself to pointing out, very briefly, some issues that could have deserved more reflection (and could have been straightforwardly implemented in the dictionary):

(a) the need to distinguish between straightforward gloss, explanatory gloss, transposition into modern terminology, and definition; compare the difference between the four following examples:

descisco 'defect from' (p. 125), (straightforward) gloss
detractio/detrectio 'removal', of letters (p. 127), explanatory gloss
dignitas 'intransitive construction' (p. 130), transposition into modern terminology
diastole 'mark of punctuation, indicating separation' (p. 128), definition

(b) the clear marking of the difference between conceptual meaning (or: the rendering of a term by a gloss or by a definition) on the one hand, and applicational range on the other;

(c) the making explicit of the principles used for lemmatization (see below, section 3).

(2) One would also have liked to find some further explanations as to how the term 'Latin' and the syntagm 'grammatical terminology' have to be interpreted. The implication seems to be that 'grammatical terminology' stands here for a type of 'sublanguage',³ and more specifically for the technical (meta)language of (ancient [Latin]) grammar (cf. the passage quoted at the beginning of this review, section 1). It remains unclear, however, how Schad defines and delimits (what) a grammatical term (is), and to some users of the Lexicon it may not be clear on what grounds Schad excludes rhetorical terms, but includes terms referring to poetry, to prose, and to style in general⁴ (see also section 3 concerning the inclusion of particular terms), and why she does not take into account the commentaries of grammarians on literary authors.

As to the adjective 'Latin' in the title, this should be understood as referring to (grammatical terminology used in) texts written in Latin (or, in Latin and Greek, since some of the artes grammaticae are (to some extent) bilingual texts). Several terms in the Lexicon are in fact Greek terms taken over in their morphological shape by Latin authors (and sometimes conserved in their Greek graphical shape within the textual tradition).⁵

Finally, 'Latin grammatical terminology' refers to the terminology of authors (writing in Latin) from Antiquity and the Early Middle Ages. On practical grounds, such a restriction is fully justified (otherwise the Latin/Latinate terminology of writers in Modern Times should have to be included also), but while Schad seems to set the end of the 7th century as the terminus usque quem,⁶ one misses important texts of the 6th and 7th century, such as Isidorus' Etymologiae, the Ars Ambrosiana, or the Expossitio Latinitatis of the Anonymus ad Cuimnanum.

3. As to the micro-organization of the Lexicon, the overall (and maximal) structure of the entries is as follows:

term (or lemma) [in bold]
English translation, definition or explanation [in italics]
references and citations from Latin authors, with information organized according to the following relevant domains: classification, definition, subdivision/characteristics, etymology/explanation of the term, (early/earliest) attestation of term with a technical-grammatical meaning, collocations, possible points of interest
corresponding Greek term
synonyms, antonyms, and related terms
secondary literature
non-grammatical signification/use of the term

As an illustrative example, the following entry (p. 425) may be quoted:⁷

univocus

'homonymous'
Prisc. part. 3.511.12 Turnus quae pars orationis est? nomen . . . cuius est speciei? -ae: significat enim nomen proprium regis Rutulorum et appellativum piscis palustris (498.17).
neut. as noun (sc. nomen) 'homonym'
Prisc. part. 3.462.3 'arma' . . . cum . . . sit homonymon, quod quidam -um dicunt, et varias res significet. 482.22 est . . . ὁμώνυμον vel -um (497.18).
attestation: only Prisc. (index).
Prisc. uses both the Gk. term homonymon and the Latin equivalent univocum which is found in no other grammarian and appears to be Prisc.'s own creation. Amongst the grammatical terms created by the Latin grammarians, compounds such as this are less common than suffixal formations.
Gk.: ὁμώνυμον, Arist. categ. I a I; DT 36.1.
See homonymum.

This is a very informative and efficiently consultable entry. But both the gloss and the commentary could have been more elaborate:

(a) in fact, univocus/homonymon (homonymum) is used for what is predicated univocally of several 'things' belonging to the same genus; what is important, is that we have one and the same 'denomination' used for several (and this without there being involved metaphorical uses, nor homophonous forms);

(b) as accurately inventorized in Bécares Botas' Diccionario,⁹ there were many Latin translations, in the grammatical and philosophical literature, of the Greek term ὁμώνυμος: univocus; idem nomen habens; aequivocus; uninomynus.

The nomenclature of Schad's Lexicon is characterized by two features: (a) the loose (or 'liberal') conception of 'grammatical term', and (b) an analytical description. Users may feel that some entries could probably be deleted, and, in fact, some of the (sub)lemmas should be removed because they are based on an erroneous reading or on an error made in the compilation of the textual documentation.¹⁰ On several other occasions, one can have doubts about the 'grammatical' nature of the term.¹¹

I readily admit that there are cases where one can only be grateful for the inclusion of a term that does not really have a technical grammatical meaning,¹² but it seems to me that (a) a dictionary of grammatical terminology should not include terms used with their common, 'everyday' semantic load (e.g., coepi, necessitas, praeloco); (b) good reasons should be given for the inclusion of lemmas that, although they are learnèd words, cannot be properly said to be grammatical terms (e.g., adiectamentum [p. 18], breviloquium [p. 52], multifidus [p. 254]). On the other hand, if one is willing to follow Schad in her non-restrictive 'technico-lexicographical' approach, one should ask why the Lexicon has poetice 'the poetic art, poetry' and poeticus 'poetic' as entries (p. 304), but not rhetorice and rhetoricus, or why excepto 'with the exception of' (p. 154) is an entry, but not puta.¹³ It is also unclear to me why some pairs of verbs and deverbal nouns are listed as entries, whereas others are not.¹⁴

The 'analytical' treatment of the material has the advantage of convenient and targeted "item-search and consultability", but presents two disadvantages: (1) the reader has to reconstruct classificatory labels that were equivalent for Latin grammarians: e.g. adverbia separandi/separantis/separationis (a single semantic species of adverbs to be 'recomposed' out of three entries [p. 357]:¹⁵ separandi; separantis; separatio);¹⁶ (2) the same commentary is repeated under two different entries.¹⁷

Also, the 'analytical' organization of the nomenclature raises the question of lemmatization. On this matter, Schad's Lexicon could have profited from the vast literature that exists on lemmatization, with application to both dictionaries and concordances. Two important methodological points should be mentioned here:

(a) The structural reorganization of lemmas: e.g., the entry cognitio as it now stands would better be reorganized as either a double entry (cognitio prima; cognitio secunda) or as a single entry given in the form of a binomial: cognitio prima /cognitio secunda. On the other hand, the entry patiendo 'passive' (p. 292) can be deleted: the attestation given there can be grouped with those under the preceding entry patiendi (sub-meaning 2: 'passive').

(b) The alphabetical positioning of lemmas: in Schad's Lexicon we have on the one hand, under G, gens 'race, nation, people' (p. 123; in fact this refers to a semantic category, and it should have been placed under significans/significare gentem),¹⁸ and on the other hand, under P, per obtinentiam (better to be placed under O). Expressions restricted by quasi are classified, correctly in my view, under Q: quasi ad aliquid dictum (p. 337), quasi diminutio (p. 337), quasi diminutivus (p. 337).

4. Scholars interested in the history of ancient grammar, and students of Latin grammar in particular, will appreciate Schad's systematic efforts in offering a translation of the Latin grammatical terms. As already pointed out above, the translations of the entries cover a wide range of 'correspondences between a source-language and a target-language':¹⁹ one finds glosses next to transpositions into modern terminology, and definitions, or paraphrasing descriptions offering some kind of 'essential characterization'.²⁰

The difference between these types (and lexicographical strategies) is not formally (e.g., typographically) marked, nor is this metalexicographical issue discussed in the introduction. The appeal to a particular strategy is often conditioned by the very nature of the term (e.g., glossing dignitas and idioma as 'dignity' or 'idiom' would be hardly helpful, and indeed misleading).

For scholars interested in linguistic (and philological) terminology, as attested and reflected in Latin texts, it might have been interesting to refer to those dictionaries of linguistic terminology²¹ that pay specific attention to 'classical' terminology; one can think here of Springhetti's (Latin!) terminological dictionary,²² of Marouzeau's linguistic-philological dictionary,²³ and of the very encyclopedic Sprachwissenschaftliches Wörterbuch.²⁴

Schad's translations are very useful, and very often the reader can find additional information and further terminological clarification in the commentary given within an entry. I have noted a number of cases where I would propose another (or an additional) translation²⁵ for a term, and a number of cases where I hesitate to adopt Schad's translation.²⁶

5. As to the philological foundations of Schad's Lexicon, the author is quite explicit: 'The textual foundation for the present study is the corpus of the Latin grammarians edited in seven volumes by the nineteenth century German scholar Heinrich Keil. This comprises grammatical treatises of the 1st to 8th centuries, in general following a common tradition and composed along broadly similar lines. (. . .) A more exact list of the works considered in this lexicon can be obtained from the list of grammatical works printed by Lomanto and Marinone (1990: vi-vii) at the beginning of their index (vol. I). This includes all the grammarians in Keil together with a reference to the more recent editions when these are available. The same index (vol. I, p. ix) gives a list of the topics treated by the grammarians. The present lexicon is based on the authors listed in the index with some exclusions' (p. XIX); 'As indicated above, the recent editions of grammatical works, when available, are listed in the Index of Lomanto and Marinone. In this work references are regularly to Keil's edition (volume, page and line) for those works which it contains. The exceptions to this rule are the works of Charisius and Donatus for which reference is made by page and line to the editions of Barwick (1964)²⁷ and Holtz (1981)²⁸ respectively' (p. XX).

Five (relatively) recent editions of Latin grammatical texts should have been integrated, since they offer a trustworthy philological foundation:

(a) De Nonno's edition of the Anonymus Bobiensis
(b) Rosellini's edition of the Regulae of Pseudo-Remmius Palaemon
(c) Schenkeveld's edition of Iulius Romanus' fragments on the adverb in Charisius
(d) Stock's edition of Sergius' commentary on Donatus
(e) Bonnet's edition of Dositheus.²⁹

As to Schad's quotations from the extant editions, these are as a general rule very accurate apart from the occasional error and inconsistency in punctuation. A more serious case of inconsistency occurs with Schad's use of a passage in Quintilian concerning the conjunctions. Under coniunctio, Schad quotes Quintilian's text of Institutio oratoria 1.4.18 as follows (Lexicon, p. 86): [coniunction]es a plerisque dici scio, sed haec videtur ex 'syndesmo' magis propria tralatio.³⁰ But under the entry translatio (Lexicon, p. 408), and more precisely under the sub-meaning 2 'translation' (Gk. into Latin), Quintilian's text is quoted as follows: Quint. 1.4.18 haec (sc. coniunctio) videtur ex συνδεσμῷ magis propria -o.³¹

6. Any translation and/or commentary of grammatical terminology is indebted to the extant scholarly research on terms. Schad's Lexicon contains a useful bibliography of secondary literature (pp. 431-438). These studies have been integrated within the lexicographical description and commentary of entries, but Schad has taken care not to overload the contents of the entries with references to secondary literature. In a number of cases, she should, in my view, have inserted a more explicit reference: e.g., for terms relating to syntax and to (various types of) conjunctions, Baratin's 1989 study³² could have figured more prominently, and under declinatio (as well as under naturalis and voluntarius), Taylor's 1975 study³³ should have been explicitly referred to. Both studies are in Schad's bibliography.

General works, such as Pauly(-Wissowa)'s Realencyclopädie der classischen Altertumswissenschaft³⁴ could have been included in the bibliography (for some entries, the Realencyclopädie could have been referred to); and Schulze's work³⁵ on Latin proper names could have been mentioned in the bibliography (and listed in the literature s.v. agnomen, cognomen, nomen, praenomen). In the bibliography I also miss a number of (more or less recent) monograph studies or collections of articles on the history of ancient grammar.³⁶

There are also a number of specific bibliographical additions to be mentioned:

(a) The classification and the commentary under the entry adverbium should now be revised and rewritten in view of the contributions to the thematic issue of Histoire, Épistémologie, Langage on Priscian's treatment of the adverb,³⁷ and of three detailed studies that are included in the recently published thematic issue "Das Adverb in der Grammatikographie (Teil I)" of Beiträge zur Geschichte der Sprachwissenschaft (vol. 17/1-2, 2007).³⁸ (b) Problems in the description/classification of other parts of speech in Latin grammaticography have been dealt with in a variety of studies.³⁹ (c) Various collective volumes dealing with problems of terminology/terminography have recently appeared.⁴⁰ (d) For general background, the reader of Schad's Lexicon will find useful information in two recently published reference works.⁴¹

7. Schad's Lexicon has been carefully elaborated, printed and proofread. I have only a limited number of (minor) corrections, apart from those already pointed out.⁴²

8. The present review has focused on corrections, additions and refinements to Schad's Lexicon. As a consequence, the merits of the work have not been properly put into perspective. Therefore I would like, in conclusion, to lay emphasis on the wealth of information gathered by Schad, who has provided us with a reference tool that is both indispensable to historians of ancient grammar and extremely valuable for scholars interested in the history of linguistic concepts and terms. Schad's commentaries, though relatively succinct, will be of great help to readers looking for information concerning the origin, the development and the semantic(-referential) range of grammatical terms or to those in search of information on grammatical classifications and subcategorizations, at various levels of description. Schad also deserves our thanks for the inclusion of synoptic tables within a number of entries. ⁴³ It is to be hoped that someday the Lexicon will be made available online, with a web address for improvements, additions, revisions, philological corrections, etc. (all these to be eventually screened and implemented by a webmaster). As such, Samantha Schad's Lexicon will become a dynamically growing reference and working tool for the next generations.

Notes

1. V. Lomanto - N. Marinone, Index grammaticus. An index to Latin grammar texts, Hildesheim, Olms - Weidmann, 1990.

2. For a short presentation of the field of metalexicography, with bibliographical guidelines, see G. Petrequin - P. Swiggers, "La métalexicographie. Contours et perspectives d'une (sous-) discipline", L'Information grammaticale 114 (2007), pp. 7-10. For a systematic presentation of basic notions of (meta)lexicography, see J. Rey-Debove, Étude linguistique et sémiotique des dictionnaires français contemporains, The Hague - Paris, Mouton, 1971 and Le métalangage. Étude linguistique du discours sur le langage, Paris, Colin, 1978.

3. On this notion, see Z.S. Harris, Language and Information, New York, Columbia University Press, 1988.

4. Cf. entries such as poetice (noun) [=poeticê], poetice (adverb), poeticus, prosa, prosa oratio, suavitas, etc.

5. This is the case of, e.g.: ametabolus (p. 31), analogos (p. 33; the attestation cited is Varro 8.55: ἀνὰ λόγον si essent vocabula), antisigma (p. 35), apostrophos (p. 36), barytonus (p. 51; perhaps better: barytonos/barytonus), dasia (p. 107), diastole (p. 122; I would have preferred a lemmatization as diastolê), diathesis (p. 122), dichronus (p. 128; better: dichronos/dichronus), dionyma (p. 132; I would have preferred dionymon; compare: eponymon (p. 152), feronymon (p. 165), synonymon (p. 392)), epectasis (p. 151), historice (p. 196; I would have preferred historicê, especially since there is also an entry historice, adverb of historicus), horisticos (p. 196), hygros (p. 197), hypocorisma (p. 198), hypocorismos (p. 198), hypodigmaticus (p. 198), lexis (p. 236), monadicos (p. 251), monoclitos (p. 251), monophonos (p. 251), pandectes (p. 285; I would have preferred a lemmatization as pandectês/pandektês), paracimenus (p. 286; better: paracimenos/paracimenus), paragogus (p. 286; better: paragogos/paragogus), parasynthetos (p. 286), parataticon (p. 287), paronymos (p. 287; this should in my view have been lemmatized as paranonymon), perilepticus (p. 297), poetice [as a noun] (p. 304; better: poeticê), polyonymon (p. 305), psile (p. 333; better: psilê), syncategorema (p. 392), synonymia (p. 392), synonymon (p. 393), tetraonyma (p. 401; better: tetraonymon), trionyma (p. 401; better: trionymon).

6. 'A self-imposed chronological limit of the end of the seventh century has also led to the exclusion of Bede, the eighth-century author' (p. XIX).

7. Occurrences of the lemma in quotations are not given in full, but indicated by - (followed by an ending, if the case applies): note that the reader has to supply sometimes the correct morphophonological shape of the lemma within an occurrence (e.g.: coniunctio: -es <coniunctiones>).

8. The entry homonymum can be found on p. 196; there the reference to the entry univocus should read 'See univocus' (instead of 'See univocum').

9. V. Bécares Botas, Diccionario de terminología gramatical griega, Salamanca, Ediciones Universidad de Salamanca, 1985, p. 274.

10. aequalis (sub-meaning 1 'similar, sharing a common form', p. 27: the quotation from Dositheus 7.408.15 (sunt -ia nomina (participiis), quae nominativo casu participiis similia videntur, ut 'cultus passus visus', quorum discretio declinatione detegitur) should be read: sunt aeque alia nomina, quae . . .( See G. Bonnet (ed.), Dosithée. Grammaire latine, Paris, Les Belles Lettres, 2005, p. 67). Likewise, simplicus (p. 371: 'occurring singly, absolutely' exc. Bob. 1.562.6 deponentia autem -a verba): the exc. Bob. have, however, simplicia (this is also the text printed by Keil, Grammatici Latini, vol. 1, p. 562, which is the edition used by Schad).

11. This is, e.g. the case of: antecresco (p. 34 'to increase in front, acquire an addition at the beginning'), antefero (p. 34 'to place in front'), circumloquor (p. 62 'to paraphrase'), coepi (p. 63: 'to begin [with a certain letter]'), custodio (p. 105 'to keep'), defendo (p. 114 'to keep'), dico (p. 128 'to say or use a certain word; to call something by a certain name; to create a word'[the latter gloss being in my view a contextualized rendering]), emendatio (p. 149 'emendation, correction'), excepto (p. 154), exiguitas (p. 157 'smallness' [said of diminutive forms]), graeculus (p. 190 'Greek' [of a type of syllable]), habere (pp. 193-195; the verb has its normal meaning 'to have' viz. to have such or such letter/to have such or such ending/to have an accent, etc.), intaminatus (p. 220 'unchanged, uncorrupted'), intemeratus (p. 222 'unchanged, uncorrupted'), internuntius (p. 233 'intermediary'), mere (p. 34 'purely'), necessitas (p. 261 'necessity'),[ Schad comments, quite correctly: 'In both Prisc. and Macr. exc. [= excerpta] a distinction is made between two types of necessity operating in language: semantic or logical and phonetic or practical', but the fact remains that for both authors necessitas has its general meaning of 'necessity; necessary conditioning'.] nitor (p. 266 'lean on'), pereo (p. 295 'to be lost' [The subject can be littera (or the name of a letter/sound), vis or Latinitas]), praeloco (p. 311 'to place in front'), sum (p. 387[Schad translates 'have regular comparison' [of adjectives], but the two attestations quoted correspond to the normal use of esse ('to be, to exist'); see e.g. the second attestation: sunt quae ex positivi imagine per omnes gradus formantur, ut 'fortis fortior fortissimus' (Diom. 1.324.26)]). Note also Schad's commentary on the entry rigidus 'rigid, stiff' [of voiced stops and the letter q, as they receive no aspiration] (p. 351): "This was not created as a grammatical term but taken over from the ordinary vocabulary."

12. This holds, e.g., for entries such as ius, lex, mos, variatio, which offer interesting information on the sociolinguistic and (socio)stylistic views of Latin authors.

13. Also, fortis is an entry (p. 178), but not vividus, although both adjectives occur as qualifying the noun syllaba in the passage of Terentianus Maurus quoted s.v. fortis. [In the list of 'Abbreviations', p. XVI, correct 'Terenianus' into 'Terentianus'.]

14. Cf. corrumpo and corruptio; deduco and deductio; demo and demptio; derivo and derivatio; detraho and detractio; enuntio and enuntiatio; figuro and figuratio; inclino and inclinatio; muto and mutatio; but the entries dubitandi and dubitatio are not flanked by an entry dubito; and we have perfectio, but not perficio; positio, but not pono.

15. The term separativus is not used in collocation with adverbium.

16. See also other semantic classes of adverbs: s.v. demonstrandi/demonstrantis/demonstratio/demonstrativus (pp. 117-118); s.v. hortandi/hortantis/hortatio/hortativus (p. 197); s.v. negandi/negantis/negatio/negativus (p. 261-262); s.v. prohibendi/prohibentis/prohibitio (p. 325); s.v. respondendi/respondentis/responsio/responsivus (p. 350).

17. Just two examples: the commentary (11 lines) s.v. pronomen (p. 327) starting with 'Var[ro] divided words with case inflection' is identical with the one on p. 333 s.v. provocabulum; the commentary (7 lines) s.v. demonstrativus is almost identical with the commentary (p. 347) s.v. relativus.

18. In any event, it seems to me that the entry should contain the term 'significare' (even when the concept of 'meaning' is only implied); compare the entry sensum significans (p. 355: Prisc. 3.277.13 verba sensum significantia . . . accusativa . . . adiunguntur, ut 'amo' et 'desidero, ardeo').

19. On translation(al problems) of ancient grammatical terminology, see P. Swiggers - A. Wouters, "Content and Context in (Translating) Ancient Grammar", in: P. Swiggers -A. Wouters (eds.), Ancient Grammar: Content and Context, Louvain, Peeters, 1996, pp. 123-161; and "Translating Ancient Grammatical Texts", in: D. Cram et al. (eds.), History of Linguistics 1996, vol. I, Amsterdam, J. Benjamins, 1999, pp. 3-11.

20. For a good example of the latter, see the entry idioma (p. 199): 'distinctive property'.

21. For a bibliographical and methodological survey, see P. Swiggers, "Pour une systématique de la terminologie linguistique: considérations historiographiques, méthodologiques et épistémologiques", in: Mémoires de la Société de Linguistique de Paris, n.s. vol. 6: La terminologie linguistique, 1999, pp. 11-49.

22. Aem. Springhetti, Lexicon linguisticae et philologiae, Romae, Apud Pontificiam Universitatem Gregorianam, 1962.

23. J. Marouzeau, Lexique de la terminologie linguistique: français, allemand, anglais, italien, Paris, P. Geuthner, 1933 [several reeditions and reprints]; also useful is F. Lázaro Carreter, Diccionario de términos filológicos, Madrid, Gredos, 1953 [several re-editions].

24. J. Knobloch (ed.), Sprachwissenschaftliches Wörterbuch, Heidelberg, Winter, 1961- (13 fascicles published).

25. I will not comment on Index II: Greek - Latin (pp. 447-451); this index can now be supplemented with the 'Index contrastif grec-latin' in L. Basset et al. (eds.), Bilinguisme et terminologie grammaticale gréco-latine, Louvain, Peeters, 2007, pp. 443-446.

26. The following is a selective list. accidens (p. 8-9): should this really be translated as 'adjective'?; character (p. 61 'character, form of a letter'): perhaps better 'shape'; congregatio (p. 83 'joint occurrence'): I would translate as 'joining'; contextus (p. 96 'coherent structure'): I would translate as 'declensional/inflectional frame' or as 'frame, context, setting'; designatio (p. 126 'designating force'): rather 'reference'; dignitas (p. 130 'intransitive construction'): I would propose 'subject-predicate construction' (corresponding to Greek ἀξίωμα); discretio (p. 134 'separation, distinction, being singled out', of pronouns): rather 'separation, distinction, singling out [by means of a pronoun]'; formula (p. 178, under 2 'form'): add 'scheme'; impositio (p. 206, under 1 'arbitrary imposition of names on objects'): perhaps better 'imposition of names on objects by a decision'; inclinamentum (p. 210 'derivative suffix'): I would prefer a less specific translation, such as 'attachment'; notatio (p. 270 'the explanation (of a term) according to its derivation, etymology'): in the passages from Cicero quoted here, I would render notatio as 'representation'; nuncupo (p. 273 'name'): better 'to designate with a name'; orthographus (p. 283 'concerning orthography'): a more accurate translation would be 'author of an orthographical text/of a work on orthography'; pars (p. 287, under 2 'subdivision'): also 'type'; per obtinentiam (p. 298 'indicating possession'): rather 'according to/through acquisition'; psile (p. 333 'absence of h/aspiration'): given that dasia (p. 107) is glossed as 'rough', I would propose here 'simple/unaffected'; for reciprocus (p. 342) applied to participles in Priscian, Schad proposes 'able to be used predicatively and attributively', but Priscian merely wants to point out the interchangeability (or reciprocal conversion) between (some) nouns and participles, e.g., lector and legens, cursor and currens (See now P. Swiggers - A. Wouters, "Le participe, unité "concrète", (étymologiquement) vraie et problématique", Incontri Linguistici 31 (2008), pp. 101-110 (esp. p. 107 note 3). See also A. Garcea - A. Giavatto, "Reciprocus - Antanaklastos. Pronomi e participi tra grammatici e filosofi", Voces 15 (2004), pp. 43-58.); significantia (p. 361 'meaning, sense'): also 'meaningfulness, significance'.

27. K. Barwick (ed.), Charisii artis grammaticae libri V (with add. by F. Kuehnert), Lipsiae, B.G. Teubner, 1964.

28. L. Holtz, Donat et la tradition de l'enseignement grammatical: étude sur l'Ars Donati et sa diffusion (IVe - IXe siècle) et édition critique, Paris, CNRS Éditions, 1981.

29. M. De Nonno (ed.), La grammatica dell' <<Anonymus Bobiensis>> (GL I 533-565 Keil). Con'un appendice carisiana, Roma, Ed. di Storia e Letteratura, 1982. M. Rosellini (ed.), Ps. Remmii Palaemonis Regulae. Introduzione, testo critico e commento, Hildesheim, Olms, 2001. D.M. Schenkeveld (ed.), A Rhetorical Grammar: C. Iulius Romanus. Introduction to the Liber de adverbio as incorporated in Charisius' Ars grammatica II.13. Edition with introduction, translation and commentary, Leiden, Brill, 2004. Ch. Stock (ed.), Commentarium de oratione et de octo partibus orationis artis secundae Donati. Überlieferung, Text und Kommentar, München - Leipzig, Saur, 2005. G. Bonnet (ed.), Dosithée. Grammaire latine, Paris, Les Belles Lettres, 2005.

30. The entry convinctio (Lexicon, p. 98) reads as follows: 'conjunction'. Quint. 1.4.18 Aristoteles quoque atque Theodectes verba modo et nomina et -es (P, coniunctiones AB) tradiderunt . . .in -bus autem complexus eorum esse iudicaverunt: quas coniunctiones a plerisque dici scio, sed haec videtur ex syndesmo magis propria tralatio . . .a philosophis primum -bus articuli adiecti, post praepositiones. Gk.: "suvndesmo", see s.v. coniunctio.In the list of 'Abbreviations', on p. XV, one should correct 'Quint.: Quintilian, institutiones orationis unless otherwise specified' into: 'Quint.: Quintilian, institutio oratoria unless . . ." [or: institutionis oratoriae libri XII].

31. Several remarks should be made here:

(a) it would have been recommendable to quote the same text uniformly; Radermacher's critical edition (M. Fabi Quintiliani Institutionis oratoriae libri XII. Edidit Ludwig Radermacher, Lipsiae, in aedibus B.G. Teubneri, 1965) gives the text as follows: [ . . . quorum fuerunt] Aristoteles quoque atque Theodectes, verba modo et nomina et convinctiones tradiderunt, [videlicet quod in verbis vim sermonis, in nominibus materiam (quia alterum est quod loquimur, alterum de quo loquimur),] in convinctionibus autem complexus eorum esse iudicaverunt: quas coniunctiones a plerisque dici scio, sed haec videtur ex sundevsmw/ magis propria translatio.(M. Fabi Quintiliani Institutionis oratoriae libri XII, o.c., vol. I, p. 25.)

(b) the variant form tralatio is attested in the Bern (B) manuscript, but translatio is the form adopted in the critical edition, and the one that should have been given in quotation; by internal inconsistency, Schad's Lexicon does not include an entry tralatio, but this is to the benefit of the work.

(c) the Greek part within the quotation should have been given, in Greek characters and correctly, as in Radermacher's edition.

(d) in the entry translatio the parenthetical addition within the quotation should have been '(sc. convinctio)', since Quintilian justifies his literal rendering of the Greek term.

32. M. Baratin, La naissance de la syntaxe à Rome, Paris, Éditions de Minuit, 1989.

33. D.J. Taylor, Declinatio: A Study of the Linguistic Theory of Marcus Terentius Varro, Amsterdam, Benjamins, 1975.

34. And now the online 'New Pauly' (Der Neue Pauly).

35. W. Schulze, Zur Geschichte lateinischer Eigennamen (Berlin, Weidmann, 1904; reprint, with additions: 1991).

36. Ax, W. 2000. Lexis und Logos. Studien zur antiken Grammatik und Rhetorik. Stuttgart: Steiner. Dammer, R. 2001. Diomedes grammaticus. Trier: Wissenschaftlicher Verlag. Desbordes, F. 1990. Idées romaines sur l'écriture. Lille: Presses universitaires. Matthaios, S. 1999. Untersuchungen zur Grammatik Aristarchs: Texte und Interpretation zur Wortartenlehre. Göttingen: Vandenhoeck & Ruprecht. Each of these works could have been put to use within various entries of the Lexicon.

37. M. Baratin - A. Garcea (eds), "Autour du De adverbio de Priscien", in: Histoire, Épistémologie, Langage 27/2 (2005).

38. The issue editors are Aino Kärnä and Stephanos Matthaios. S. Matthaios, "Das Adverb in der Grammatikographie der griechischen Antike" (pp. 13-58); A. Wouters - P. Swiggers, "L'adverbe chez les grammairiens latins de l'Antiquité tardive" (pp. 75-118); L. Visser, "The Description of the Adverb in the Early Middle Ages" (pp. 119-158). The second part, also edited by Kärnä and Matthaios, does not include contributions on Antiquity; it was published as volume 18/1-2 of the Beiträge.

39. On the interjection: F. Biville, "La syntaxe aux confins de la sémantique et de la phonologie: les interjections vues par les grammairiens latins", in: P. Swiggers - A. Wouters (eds.), Syntax in Antiquity, Louvain - Paris, Peeters, 2003, pp. 227-239; G. Graffi, "L'interiezione tra i grammatici greci e i grammatici latini", Incontri Linguistici 19 (1996), pp. 11-18. On the pronoun: S. Carraro, "Osservazioni sulla definizione di pronome nelle Artes Grammaticae", Aevum 73 (1999), pp. 81-91; M. Lenoble - P. Swiggers - A. Wouters, "La structure des artes grammaticae latines. L'exemple du pronom", in: S. Auroux (ed.), History of Linguistics 1999, Amsterdam, Benjamins, 2003, pp. 1-18; P. Swiggers - A. Wouters, "L'analyse du pronom comme catégorie morpho-sémantique", to appear in a volume on Priscian (edited by M. Baratin, B. Colombat, L. Holtz and I. Rosier). On the adjective noun: B. Colombat, "L'adjectif dans la tradition latine: vers l'autonomisation d'une classe", Histoire, Épistémologie, Langage 14/1 (1992), pp. 101-122; M. Negri, "Adiectiuum ed epitheton nella terminologia della grammatica e dell'esegesi letteraria latina. I problemi di un <<doppione>>", in: L. Basset et al. (eds.), Bilinguisme et terminologie grammaticale gréco-latine, Louvain - Paris - Dudley, Peeters, 2007, pp. 285-302. On the noun: M. Lenoble - P. Swiggers - A. Wouters, "Étude comparative des dénominations de catégories grammaticales dans les textes artigraphiques latins de l'Antiquité", in: B. Colombat - M. Savelli (eds.), Métalangage et terminologie linguistique, Louvain, Peeters, 2001, vol. I, pp. 275-291. On the proper 'noun': S. Matthaios, "κύριον ὄνομα. Zur Geschichte eines grammatischen Terminus", in: P. Swiggers - A. Wouters (eds.), Ancient Grammar: Content and Context, Louvain, Peeters, 1996, pp. 55-77.

40. Among those of direct relevance for the contents of this Lexicon I would mention: Basset, L. et al. (eds.) 2007. Bilinguisme et terminologie grammaticale gréco-latine. Louvain: Peeters. Colombat, B. - Savelli, M. (eds.) 2001. Métalangage et terminologie linguistique. Louvain: Peeters. (2 vols.) Fögen, Th. (ed.) 2005. Antike Fachtexte. Ancient Technical Texts. Berlin: W. de Gruyter. Swiggers, P. - Wouters, A. (eds.) 2002. Grammatical Theory and Philosophy of Language in Antiquity. Louvain: Peeters. Swiggers, P. - Wouters, A. (eds.) 2004. Syntax in Antiquity. Louvain: Peeters.

41. Auroux, S. - Koerner, E.F.K. - Niederehe, H.-J. - Versteegh, K. (eds.) 2000. History of the Language Sciences. An international Handbook on the Evolution of the Study of Language from the Beginnings to the Present, vol. I. Berlin: Mouton, de Gruyter. Colombat, B. (ed.) 1998. Corpus représentatif des grammaires et des traditions linguistiques, t. I. Paris: SHESL (Histoire, Épistémologie, Langage, hors-série 2).

42. On p. 244 s.v. loquel(l)a, the commentary should read: 'Prepositional prefixes joined onto other words (loquellae) are distinguished from prepositions which collocate with case forms (casus)'. In the 'References' section, there are a number of errors: sub Alvar Ezquerra, correct: García Fernández . . . Lingüística; sub Alvarez Huerta, correct: Álvarez Huerta . . . García Fernández; sub André 1952, correct: Revue; sub Charpin 1965, correct: Revue; sub Charpin 1977, read: Diffusion H. Champion; sub Cousin 1943, correct: spéciales; sub Di Benedetto 1958 and 1959, correct: Dionisio il Trace; sub Flobert 1981, correct: Revue; sub Kaster 1978, correct: Servius and; sub Lambert 1908, correct: selon les grammairiens; sub Lejeune 1950, correct: Revue . . . Grecques; sub Malkiel 1941, read: -ivu with a long -i; sub Marache 1952, correct: archaïques; sub Panagel, correct: Panagl; sub Robins 1986, correct: Van Hoecke; sub Schuchardt 1866, read: 1866-1868, Der Vokalismus des Vulgärlateins (3 volumes); sub Serbat, correct: médiatif; sub Vendryès, correct: Vendryes; sub Wouters 1994, correct: μετ' ὀλίγον. It might also have been better to give 1959-1960 as the publication date of the fourth edition of Ernout and Meillet's Dictionnaire étymologique de la langue latine (of this fourth edition there are many reprints, such as the 1979 one listed in Schad's bibliography).

43. Personally, I would also have liked to see synoptic tables for the (sub)classification of the various parts of speech, as well as a comprehensive survey of the accidentia. Later editions of the Lexicon could be easily extended with a number of such tables; at the same time, they would make it possible to incorporate more attestations and, in some cases, a reorganization of the structure of some entries.