¹ e.g. Lewis, C.T. and Short, C., A Latin Dictionary (Oxford, 1879)Google Scholar; Georges, K.E., Ausführliches lateinisch-deutsches und deutsch-lateinisches Handwörterbuch (Leipzig, 1879–80⁷), 2144Google Scholar.

² Housman, A.E., ‘Draucus and Martial XI 8 I’, CR 44 (1930), 113–15Google Scholar (= The Classical Papers of A.E. Housman [London, 1972], 1166–7Google Scholar). See also Citroni, M., M. Valerii Martialis Epigrammaton liber primus (Florence, 1975), 297Google Scholar; Kay, N.M., Martial Book XI: A Commentary (London, 1985) 224Google Scholar; Vioque, G. Galán, Martial, Book VII: A Commentary (Leiden, 2002), 387Google Scholar. Less certain about Housman's analysis is Howell, P., A Commentary on Book One of the Epigrams of Martial (London, 1980)Google Scholar, 308. The gloss ‘sodomite’, ‘pederast’ is, however, still found in some post-Housman publications, e.g. Pepe, L., Marziale (Naples, 1950), 17Google Scholar; Ernout, A. and Meillet, A., Dictionnaire étymologique de la langue latine (Paris, 1959 ⁴), 184Google Scholar; Ferguson, J., Juvenal. The Satires (New York, 1979), 285Google Scholar; Gernia, M.L. Porzio, ‘Gli elementi celtici del latino’, in Campanile, E. (ed.), I celti d'Italia (Pisa, 1981), 97–122Google Scholar, at 109.

³ Text and translation follow Bailey, D.R. Shackleton, Martial. Epigrams (Cambridge, MA, 1993)Google Scholar, who translates draucus as ‘athlete’.

⁴ On the athletic practice of growing a thick neck see the note in Shackleton Bailey (n. 3) ad loc.

⁵ 1.96.12–13 sed spectat oculis devorantibus draucos | nec otiosis mentulas videt labris (‘[In the bath] he never looks up, but watches the drauci with devouring eyes, and his lips work as he gazes at their cocks’); 9.27.10–13 occurrit aliquis inter ista si draucus | iam paedagogo liberatus et cuius | refibulavit turgidum faber penem: | nutu vocatum ducis … (‘If, as this goes on, some young draucus comes your way, now freed from tutelage, whose swollen penis has been unpinned by the smith, you summon him with a nod and lead him off’); 11.72 Drauci Natta sui vorat pipinnam, | collatus cuï Gallus est Priapus (‘Natta devours the willy of his [her?] draucus, compared to whom Priapus is a eunuch)’: Gallus, viz. a voluntarily castrated priest of Cybele; vocat MSS: vorat Scriverius.

⁶ See esp. Busch, S., Versus Balnearum. Die antike Dichtung über Bäder und Baden im römischen Reich (Stuttgart, 1999)CrossRefGoogle Scholar, 477 n. 37: ‘Wahrscheinlicher ist daher der Begriff drauci dazu verwendet, Männer von einer bestimmten Statur zu charakterisieren, die das Interesse des Ungenannten auf sich ziehen’, with reference to Mart. 1.96.12.

⁷ Athletic prowess was believed to be impaired by sex: Galen 8.451 (Kühn), Σ Juv. 6.73.

⁸ Cf. Mart. 9.27.11–12, cited in n. 5 above. In addition to commentaries cited above (n. 2) see Previ, F. Fortuny, ‘Consideraciones sobre algunos hapax de Marcial’, AUM 40 (1981–2), 111–26Google Scholar, at 117–19.

⁹ The word may also be attested in Juvenal, if raucus at Sat. 11.156 is to be emended to draucus: nec pupillares defert in balnea draucus | testiculos (‘he [viz. the well-behaved slave] does not cart his teenage testicles into the bath’); see Haeberlin, C., ‘Juvenal Sat. XI, 156’, Philologus 50 (1891), 506CrossRefGoogle Scholar, and Verdière, R., ‘Iuuenalianum’, Latomus 11 (1952), 25–6Google Scholar. raucus is usually understood as referring to the boyish tremble of an unbroken voice: see Friedländer, L., D. Junii Juvenalis saturarum libri V (Leipzig, 1895), 505Google Scholar; Courtney, E., A Commentary on the Satires of Juvenal (London, 1980), 509Google Scholar, who both dismissed the conjecture draucus, as did Housman (n. 2).

¹⁰ On Draucus as a personal name, see nn. 21 and 22 below.

¹¹ Viz. acc. pl. draucos.

¹² Franz's reference to CGL 7: 501, repeated by several scholars following in his wake, is in fact to the index to the CGL, compiled by Heraeus.

¹³ Franz, L., ‘Ein Fluchtäfelchen aus Veldidena’, JÖAI 54 Beibl. (1959), 70–6Google Scholar, at 74. Franz was followed by Versnel, H.S., ‘Beyond cursing: the appeal to justice in judicial prayers’, in Faraone, C.A. and Obbink, D. (edd.), Magika Hiera: Ancient Greek Magic and Religion (Oxford, 1991), 60–106Google Scholar, at 83, 103 n. 121. (Cyril's gloss was in fact already signalled by Lindsay, W.M., ‘New light on Festus’, CQ 26 [1932], 193–4CrossRefGoogle Scholar). The form δραύκιν has since surfaced in a Delian defixio (SEG 53.813, 1st–2nd century c.e.): see Jordan, D., ‘Une prière de vengeance sur une tablette de plomb, à Délos’, RA 33 (2002), 55–60Google Scholar, at 59.

¹⁴ Egger, R., ‘Nordtirols älteste Handschrift’, SAWW 244 (1964), 3–23Google Scholar, at 8.

¹⁵ Grassl, H., ‘Neue Gedanken zur Fluchtafel von Wilten-Veldidena’, in Haider, P.W. (ed.), Akten des 6. Osterreichischen Althistorikertages (Innsbruck, 1998), 68–74Google Scholar.

¹⁶ For the sake of completeness, it should be mentioned that draucus may be attested in the Appendix Probi III, where entry 153 reads raucus ñ _ra_cus. Endlicher, the first editor, had already corrected the corrupted entry into raucus non draucus (Analecta grammatica maximam partem anecdota, vol. 1 [Vienna, 1837], 445)Google Scholar, and F. Bücheler (apud Förster, W., ‘Beitrag zur Textkritik der Appendix Probi’, WS 14 [1892], 278–322Google Scholar) followed suit, suggesting that the word order must be reversed: the standard form was draucus, while the condemned one, raucus (≠ raucus ‘hoarse’!), showed a simplification of the unusual initial cluster dr-. Endlicher's conjecture was accepted by several scholars, most recently by Powell, J.G.F., ‘A new text of the Appendix Probi’, CQ 57 (2007), 687–700CrossRefGoogle Scholar, and Asperti, S., ‘Il testo dell'Appendix Probi III’, in Monaco, F. Lo and Molinelli, P. (edd.), L'‘Appendix Probi’. Nuove ricerche (Florence, 2007), 41–64Google Scholar, at 52. Other interpretations abound (see the references collected in Quirk, R., The Appendix Probi [Newark, 2006], 197Google Scholar), but it is important to emphasize that the main objection levelled against draucus by e.g. Niedermann, M., ‘Zur Appendix Probi’, RhM 60 (1905), 458–9Google Scholar and Pisani, V., Testi latini arcaici e volgari (Turin, 1960 ²), 177Google Scholar, namely that the word for ‘pederast’ would be inappropriate in an instructional text, is in fact ungrounded: as we have seen, the meaning of draucus was most likely much more innocent.

¹⁷ TLL V.1.1932.2067; Ernout–Meillet (n. 2), 184; Walde, A. and Hofmann, J. B., Lateinisches etymologisches Wörterbuch (Heidelberg, 1938–1954³), 373Google Scholar; OLD 574; Egger (n. 14), 9; Porzio Gernia (n. 2), 109; Fortuny Previ (n. 8), 117; Leary, T.J., Martial. Book XIV (London, 1996), 104Google Scholar; Henriksén, C., Martial. Book IX (Uppsala, 1999), 2.148Google Scholar; Solin, Η., Die stadtrömischen Sklavennamen (Stuttgart, 1996), 1.615Google Scholar; Watson, P., ‘The originality of Martial's language’, Glotta 78 (2002), 222–7Google Scholar, at 233. The word is not included in M. de Vaan, Etymological Dictionary of Latin and the Other Italic Languages (Leiden, 2008).

¹⁸ See e.g. Adams, J.N., Bilingualism and the Latin Language (Cambridge, 2003), 455–6CrossRefGoogle Scholar.

¹⁹ See Thurneysen, R., ‘Italisches’, ZVS 32 (1893), 554–72Google Scholar, at 562–6; Leumann, M., Lateinische Grammatik (Munich, 1977), 198Google Scholar; Sihler, A.L., New Comparative Grammar of Greek and Latin (New York, 1995), 211–12Google Scholar; Meiser, G., Historische Laut- und Formenlehre der lateinischen Sprache (Darmstadt, 1998), 123Google Scholar. According to Weiss, M., Outline of the Historical and Comparative Grammar of Latin (Ann Arbor, 2009), 163Google Scholar, traho could be derived from the same root as English drag if two aspirated consonants were dissimilated in Latin, just as they were in Greek (*d^hrag^h- > *drag^h-); while this analysis is certainly very appealing, at least there is also the old, no less semantically plausible etymology, according to which traho is cognate with Old Irish pret. ·tethraig ‘receded’. I intend to discuss the sound laws responsible for the development of initial *dr- in Latin elsewhere.

²⁰ Listed by Holder, A., Altceltischer Sprachschatz (Leipzig, 1891), 1303Google Scholar. Whatmough, J., The Dialects of Ancient Gaul (Cambridge, MA, 1970), 158CrossRefGoogle Scholar, likewise cites draucus as potentially Celtic (with a question mark), but as he explicitly says in the preface to his book, his preference had been to err on the side of inclusion.

²¹ Here is an updated and corrected list: Cisalpine Gaul (Como): AE (1991), 857; Transalpine Gaul (Nîmes): CIL 12.3775; Aquitania (Bordeaux): CIL 13.724; Lugdunensis (Champagne-en-Valromey): CIL 13.2564; Pannonia (Vienna): CIL 12.5686, 324; Britain (London): CIL 7.1336, 436a (stamp on a bowl); Algeria: CIL 8.22645, 314 (but the vase is probably from Arezzo). We can possibly identify two bearers of the same ‘barbarian’ name in Rome and Campania. A female name Drauca is attested on a graffito from a brothel in Pompeii: Arphocras hic cum Drauca bene futuit denarii (CIL 4.2193): it has sometimes been connected with the alleged meaning ‘sodomite’ of the noun draucus, which is unnecessary: it is likelier that the prostitute had a foreign name. Draucus in a graffito from Domus Tiberiana (n. 26 in Castrén, P. and Lilius, H., Graffiti del Palatino, vol. 2 [Helsinki, 1970]Google Scholar, 1st–2nd century c.e.) may well have been a slave's name and if so, belongs here, too.

²² D.E. Evans did not include draucus in his magisterial Gaulish Personal Names (Oxford, 1967), and it does not feature in any of the modern treatments of the language, e.g. X. Delamarre, Dictionnaire de la langue gauloise (Paris, 2003²). The name Draukos is attested on a Bithynian coin, and Sergent, B., ‘Les premiers celtes d'Anatolie’, REA 90 (1988), 329–58CrossRefGoogle Scholar, at 353, suggested that it may be Galatian and thus akin to the purportedly Continental Celtic names collected by Holder. However, as Sergent himself acknowledges, Δραῦκος is also found in inscriptions from Ormeleis and Lydia (see Robert, L., Noms indigènes dans l'Asie-Mineure gréco-romaine [Paris, 1963], 309–10Google Scholar n. 5 with references), and it could come from any of the languages spoken in Asia Minor, most likely from Greek (where δραυκίον‘necklace’ mentioned above potentially provides a match). Pace Carter, M., ‘A doctor secutorum and the retiarius Draukos from Corinth’, ZPE 126 (1999), 262–8Google Scholar, at 266–8, there is little reason to connect the name Δραῦκος adopted by a pankratiast from Lydian Philadelphia (IG 7.1772, second century c.e.) and a gladiator from Corinth (SEG 49.336, third century c.e.) with Martial's drauci, rather than with other Greek/Asiatic attestations of Δραῦκος. The name Draucus in late Roman inscriptions may well be Greek in origin: CIL 6.17068 (2nd–3rd century C.E., fem. Drauca, the daughter of a certain Thales), CIL 6.1725 (Flavius Olbius Auxentius Draucus, prefect of Rome a.d. 441 and 445) and CIL 6.16954.

²³ See Busch (n. 6) 477 n. 37: ‘Body-Builder’.

²⁴ See Watson (n. 17), 236–9.

²⁵ On the suffix *-iko- see Fruyt, M., Problèmes méthodologiques de dérivation à propos des suffixes latins en -cus (Paris, 1986), 33–87Google Scholar.

²⁶ Pretonic *-o- before a heterosyllabic -- gave -a- by Thurneysen-Havet's Law, see Vine, B., ‘On ‘Thurneysen-Havet's Law’ in Latin and Italic', HS 119 (2006), 211–49Google Scholar.

²⁷ δροόν· ἰσχυρόν. Ἀργεῖοι (Hsch.). One wonders if δροίτη· εἶδος ὀρχήσεως in Hesychius continues *dro-i-tā derived from δροός ‘strong’: in this case the word may have referred to a dance in which young men demonstrated their strength and agility; (for a different approach, see Schwyzer, E., ‘“Hispanisch” dureta’, ZVS 62 [1935], 199–203Google Scholar, at 200, Lawler, L.B., ‘Ladles, tubs, and the Greek dance’, AJPh 71 [1950], 70–2Google Scholar; Beekes, R.S.P., Etymological Dictionary of Greek [Leiden, 2010], 354Google Scholar thinks of a loanword).

²⁸ See Heidermanns, F., Etymologisches Wörterbuch der germanischen Primäradjektive (Berlin, 1993), 603–4CrossRefGoogle Scholar. The double -ww- in Germanic forms is a regular reflex of Proto-Indo-European *-H- (Verschärfung).

²⁹ For a different etymology of Slavic *(sъ)dravъ see Feuvre, C. Le, ‘Vieux russe dobr zdorov, russe moderne zhiv zdorov, avestique druua hauruua et l’étymologie de slave sudravu’, in Petit, D. and Pinault, G. (edd.), Langue poétique indo-européenne (Paris, 2006), 235–52Google Scholar.

³⁰ On the semantics of *doru-, *dre- see Osthoff, H., Etymologische parerga (Leipzig, 1901), 98–180Google Scholar. Other adjectival derivatives of the word for ‘tree’ include *dru-no- (Old Irish dron ‘firm, vigorous’), *dru-mo- (Old English trum ‘strong, stable’) and *druh₂to- (Lithuanian dr tas ‘strong’ and Welsh drud ‘valiant’). Other Celtic reflexes of *druh₂to- ‘strong’ seem to have undergone a semantic change: Old Irish drúth means ‘wanton woman’, and the Romance borrowings mean not only ‘strong’, but also ‘lover, libertine’ (Old Provençal drut, Milanese Italian drudo, Old French dru).

³¹ There is one more possible root-etymology for *droó- > Proto-Italic *draó-: this could be a nomen agentis made from the root *dre- ‘to run’ (Sanskrit drav-; cf. Greek δράω from *dreh₂-). *dro-ó- ‘runner’ (=Skt. dravá- ‘id.’) would belong to the same type as Latin coquus ‘cook’ (coquo) or Greek ἀοιδός ‘singer’ (ἀείδω), τροφός ‘nurse’ (τρέφω). This etymology is hard to dismiss out of hand, but *droó- ‘strong; strong man’ seems to me to offer a better semantic solution, since it would be hard to imagine that *droó- > *draó- ‘runner’ survived for centuries on the periphery of the Latin language as a technical term only to resurface with essentially the same meaning in Martial.

³² Via a noun *drai- ‘firmness, strength’ as the intermediate stage: compare raucus ‘hoarse’ (<*raiko-) made from ravis ‘hoarseness’, itself derived from rauus ‘hoarse’ < Proto-Italic *roó- < Indo-European *h₃roHó- (see Vine [n. 26], 237).