¹ Adams, J.N., Bilingualism and the Latin Language (Cambridge 2003)CrossRefGoogle Scholar. More generally on the use of Greek in late Republican Latin writers (a large topic): Kaimio, J., The Romans and the Greek Language (Helsinki 1979) esp. 295–315Google Scholar; Wenskus, O., Emblematischer Codewechsel und Verwandtes in der lateinischen Prosa: zwischen Nähesprache und Distanzsprache (Innsbruck 1998)Google Scholar; Rosén, H., loqui, Latine: Trends and Directions in the Crystallization of Classical Latin (München 1999) 21–30Google Scholar.

² Sedley, D., ‘Lucretius' Use and Avoidance of Greek’, in Adams, J.N. and Mayer, R.G. (eds), Aspects of the Language of Latin Poetry (Oxford 1999) 227–46Google Scholar. For Cicero's use of Greek terms in his philosophical works, see Powell, J.G.F., ‘Cicero's Translations from Greek’, in id. (ed.) Cicero the Philosopher (Oxford 1995) 273–301, at 288–97Google Scholar.

³ Heller, M., ‘Introduction’, in Heller, M. (ed) Codeswitching: Anthropological and Sociolinguistic Perspectives (Berlin, New York and Amsterdam 1988) 1–24, at 1CrossRefGoogle Scholar.

⁴ (n. 1) 18-29.

⁵ Scotton, C.M., ‘Differentiating Borrowing from Code-Switching’, in Ferrara, K., Brown, B., Walters, B. and Baugh, J. (eds), Linguistic Change and Contact: Proceedings of the Sixteenth Annual Conference on New Ways of Analyzing Variation (Austin 1991) 318–25, at 320Google Scholar.

⁶ Wallace-Hadrill, A., ‘To be Roman, Go Greek’, in Austin, M., Harries, J. and Smith, C. (eds), Modus Operandi: Essays in Honour of Geoffrey Rickman (London 1998) 79–91Google Scholar.

⁷ See Gal, S., “The Political Economy of Code Choice’, in Heller, (n. 3) 245–64Google Scholar. More generally, on negotiations of power involved in communication across linguistic boundaries, see Bourdieu, P., “The Economics of Linguistic Exchange’, Social Science Information 16 (1977) 645–68CrossRefGoogle Scholar.

⁸ Modern studies of codeswitching tend to analyse samples of spoken communication rather than literary texts. It is true that the works of Catullus and Caesar are not spontaneous communicative utterances in which the political and social relationship between two language groups has been preserved. To adopt the dichotomy employed in the critical literature, these are ‘artificial’ rather than ‘authentic’ code switches: Callahan, L., Spanish/ English Codeswitching in a Written Corpus (Philadelphia 2004) 107–28CrossRefGoogle Scholar. But the significance of codes witches in these authors' texts (and the context of their reception by readers) nonetheless depends, no less than in oral communication, on the associations of particular linguistic usages in the period, and thus no less bear witness to the sociolinguistic field in which Catullus and Caesar were writing.

⁹ Rossi, A., “The Camp of Pompey: Strategy of Representation in Caesar's Bellum Civile’, Classical Journal 95 (2000) 239–56Google Scholar. at 248.

¹⁰ Barwick, K., Caesars Bellum Civile (Tendenz, Abfassungzeit und Stil) (Berlin 1951) 148–9Google Scholar. Some other exceptions are examined in the pages below. Also on Caesar's incorporation of Greek vocabulary into his commentarii: Eden, P.T., ‘Caesar's Style: Inheritance Versus Intelligence’, Glotta 90 (1962) 74–117, at 94-7Google Scholar; Oldfather, W.A. and Bloom, G., ‘Caesar's Grammatical Theories and His Own Practice’, CJ 22 (1927) 584–602Google Scholar, at 590. Most recently on Caesar's style in the Bellum Civile: Batstone, W. and Damon, C., Caesar's Civil War (Oxford 2006) 143–65Google Scholar.

¹¹ [Caesar], B Air. 19: … sagitaríis ac funditoribus hippotoxotisque compluribus … (‘with archers and slingers and a great many mounted horsemen’).

¹² The passage as a whole consciously imitates and amplifies Caesar's catalogue of Pompey's troops. Labienus' forces are an even more varied ethnic mixture of Numidians, Germans, Gauls, free, slave — in short, men ‘of every kind’ (cuiusquemodi generis). This farrago is mobilised by Labienus to wage a ‘new and unfamiliar kind of battle’ (novo atque inusitato genere proeli). The conflict between Roman and ethnic or moral ‘other’, a theme at the forefront of the texts of the Caesarean continuators, is discussed by Cluett, R., ‘In Caesar's Wake: The Ideology of the Continuators’, PLLS 11 (2003) 118–31Google Scholar. at 121-4.

¹³ Epitome Alexandri [‘Metz Epitome’] 60: Alexander imperai iis, quos hippotoxotas dicunt, sagittas in Porum uti conicerent… (‘Alexander ordered the ones whom they call “hippotoxotae” to shoot arrows at Porus …’). On this text, see Baynham, E., ‘An Introduction to the Metz Epitome: Its Traditions and Value’, Antichthon 29 (1995) 60–77CrossRefGoogle Scholar.

¹⁴ Corpus Glossariorum Latinorum 5.585Google Scholar: Hippotoxoti funditores sunt qui nervum arcus ita componunt ut sit inmedio funde (‘Hippotoxoti are slingers who arrange their bowstring such that it is in the middle of a sling [?]’).

¹⁵ B Afr. 13; also in Tacitus (Ann. 2.16Google Scholar, 13.40) and Vegetius (1.4, 20).

¹⁶ Rossi (n. 9); Tronson, A., ‘Pompey the Barbarian: Caesar's Presentation of “The Other” in Bellum Civile 3’, in Joyal, M. (ed.) In Altum: 75 Years of Classical Studies in Newfoundland (St Johns 2001) 73–104Google Scholar. Lucan 1.392-46. will turn the strategy back on Caesar, emphasising the Gallic element in his catalogue of Caesar's troops to make them seem more foreign: Batinski, E.E., ‘Lucan's Catalogue of Caesar's Troops: Paradox and Convention’, CJ 88 (1992) 19–24Google Scholar.

¹⁷ Hdt.9.80.

¹⁸ Hdt. 9.49. Hippotoxotai were later incorporated into the Athenian army, but service as a hippotoxotes was stigmatised as unworthy of a citizen Athenian (Lys. 15.6).

¹⁹ Facing the ‘great number’ of archers at Dyrrachium, Caesar depicts his soldiers developing a ‘great fear of arrows’ (magnus … timor sagittarum, 3.44Google Scholar). After the battle, the soldiers count out thirty thousand arrows to justify their fears to Caesar, and the centurion Scaeva displays a shield pierced with one hundred and twenty holes (3.53).

²⁰ Timpe, D., ‘Die Bedeutung der Schlacht von Carrhae’, MH 19 (1962) 104–29Google Scholar.

²¹ In cas tris Pompei videre licuit trichilas structas, magnum argenti pondus expositum, recentibus caespitibus tabemacula constrata… multaque praeterea, quae nimiam luxuriam et victoriae fiduciam designarent (3.96).

²² Trichilae / Tricliae appear elsewhere in Latin in Columella's versified tenth book and in the pseudo-Vergilian Copa. Although it seems Greek, its precise etymology is unclear; a connection with triclinium (or trichilinium, in an epigraphically attested form, cf. OLD s.v. triclinium) is possible. For Roman minds beholden to etymologising and wordplay, an etymological enigma invites interpretive audaciousness. Given the context, could we hear in this word an echo of τρι [intensive prefix] + [‘luxury’]?

²³ … in occultis ac reconditis templi… quae Graeci adyta appellant.

²⁴ … genus radicis… quod appellatur chara. Although the discovery of this edible (otherwise unattested) root brought great relief to Caesar's troop at a time of hunger and privation, Caesar leaves the obvious connection with (‘joy’) for the reader to discern.

²⁵ Hoc opus … machinatione navali, phalangis subiectis, ad turrim hostium admovent. Du Pontet brackets machinatione navali as a later interpolation, but the gloss is just as likely to be Caesar's own. Machinado is also, of course, formed from a Greek root, although this word had been assimilated into Latin to a greater extent (cf. TLL s.v. machinatio for uses in Caesar, Sallust, Cicero and Livy). A fuller definition of phalanga — evidently an unfamiliar word — is extant in Nonius (168M). A Greek word used earlier in Caesar's chapter, scutula (that is, ), another kind of roller with which to transport weights, is also used in its technical Greek sense (cf. [Arist]. Mech. 852a16).

²⁶ … etesiis tenebatur, gui navigantibus Alexandria Hunt adversissimi venti. That is, the (Arist. Mete. 361b.36).

²⁷ … ad eum locum qui appellabatur Palaeste omnibus navibus ad unam incolumibus milites exposuit.

²⁸ A selection, from passages detailing Caesar, his troops or allies: ‘Sulmo, a town seven miles from Corfinium’ (1.18); ‘a plain three hundred paces wide’ (1.43); descriptions of the landscape of Massilia and Utica (2.1, 2.14); ‘Buthrotum, a town opposite Corcyra’ (3.16); ‘the camp bordered on a wood and was no further than three hundred paces from the sea (3.66); ‘Aeginium, which is a town opposite Thessaly’ (3.79).

²⁹ Heiler (n. 3) 17.

³⁰ For Cicero's similarly disparaging use of Greek to brand opponents in his oratory, see Laurand, L., Etudes sur le style des discours de Cicéron (Amsterdam 1965) 72-5Google Scholar.

³¹ Lucr. 1.136-45. Cf. Fögen, T., Patrii sermonis egestas: Einstellungen lateinischer Autoren zu ihrer Muttersprache (München 2000)Google Scholar.

³² On Gnipho, see Suet, . De Gram. 7Google Scholar; Quint.1.6.23. On Latinitas, see Kaimio (n. 1) 297-9. The word is first extant at Auct. ad Her. 4.12.17, where it is linked explicitly with the ideal of linguistic purity: Latinitas est, quae sermonem purum conservât, ab omni vitio remotum (‘It is Latinitas that keeps speech pure and separate from every imperfection’). Cf. Caesar's praise of Terence as ‘a lover of pure speech’ (puri sermonis amator, Suet. Vit. Ter. 5).

³³ Hendrickson, G.L., “The De Analogia of Caesar: Its Occasion, Nature and Date, with Additional Fragments’, CPh 1 (1906) 97–120Google Scholar.

³⁴ Cic. Brut. 140: Nam ipsum Latine loqui est illud quidem, ut paulo ante dixi, in magna laude ponendum, sed non tam sua sponte quam quod est a plerisque neglectum; non enim tam praeclarum est scire Latine quam turpe nescire…

³⁵ Cic. Brut. 252: multis litteris et eis quidem reconditis et exquisitis… For the implications of elegans, not ‘elegant’ but ‘careful of thought and style’, see Krostenko, B., Cicero, Catullus and the Language of Social Performance (Chicago and London 2001) 114–23Google Scholar.

³⁶ Brut. 250. ‘Word choice is the origin of eloquence’ (verborum dilectum originem esse eloquentiae) was apparently Caesar's maxim (Cie., Brut. 253)Google Scholar.

³⁷ Caesai, De Analogia fr. 1 [Funaioli], apud Cic. Brutus 253: Ac si, ut cogitata praeclare eloqui possent, non nulli studio et usu elaboraverunt … hunc facilem et cotidianum novisse sermonem num [MS nunc] pro relicto est habendum?

³⁸ Cic., Tuse. 1.15Google Scholar: seis enim me Graece loqui in Latino sermone non plus solere quam in Graeco Latine…

³⁹ De Off. 1.111Google Scholar. In the second century, a character in one of Afranius’ comediae togatae had already expressed an aversion to codeswitching in conversation: Afran., Com. 272–3Google Scholar [apud Non. p. 396]. See Jocelyn, H.D., ‘Code-Switching in the Comoedia Palliata’, in Vogt-Spira, G. and Rommel, B. (eds) Rezeption und Identität: Die kulturelle Auseinandersetzung Roms mit Griechenland als europäisches Paradigma (Stuttgart 1999) 169–95Google Scholar. at 190.

⁴⁰ Suet, . lui. 82Google Scholar.

⁴¹ Julius Caesar III, i, 77Google Scholar.

⁴² It has been argued by Dahlmann, H., ‘Caesars Schrift über die Analogie’, RhM 84 (1935) 258–75Google Scholar. at 264-6, that the De Analogia was originally a more wide-ranging treatise on oratory, but, apart from some general stylistic aphorisms, the surviving fragments and testimonia all deal with individual points of morphology. On the other hand, there is no doubt that the analogist position stems from, and contains within it, a wider conviction that the universe is subject to regular, rational rules: see Varro's, presentation of the controversy at De Ling. Lat. 8–10Google Scholar, especially the arguments from the natural world at 9.23-30, and Caesar, , De AnalogiaCrossRefGoogle Scholar fr. 11 [Funaioli]: nisi omnia consentíant interse, non potest fieri ut nominis simìlitudo sit (“There can be no regularity in words if there is not consistency amongst all things’).

⁴³ On the debate, see Fehling, D., ‘Varro und die grammatische Lehre von der Analogie und der Flexion’, Gioita 35 (1956) 214–70Google Scholar; on its social and political context in Rome, see Sinclair, P., ‘Political Declensions in Latin Grammar and Oratory’, Ramus 23 (1994) 92–109CrossRefGoogle Scholar.

⁴⁴ Varro, , De Ling. Lat. 10.69–71Google Scholar, approaches the same topic from the perspective of the anomalist/analogist debate. On poets' practice in this matter, see Housman, A.E., ‘Greek Nouns in Latin Poetry from Lucretius to Juvenal’, Journal of Philology 31 (1910) 236–66Google Scholar.

⁴⁵ Quintilian (1.5.58-64) says that ‘a lover of the ancient grammarians’ will praise the ‘virtue’ (virtus) of the man who adopts entirely Latin endings, whose actions make ‘stronger’ the Latin language; if a word can be declined a Greek or a Latin way, the person who adopts the Greek ending will escape censure but he will not really be ‘speaking pure Latin’ (non Latine quidem … loquetur, 1.5.64).

⁴⁶ De Analogia fr. 2 [Funaioli], apud Gell. 1.10.4: tamquam scopulum sic fugias inauditum atque insolens verbum. The fact that scopulus is itself chiefly a poetic word, and one derived from Greek , lends the aphorism a certain ironic wit (although the word does appear in Caesar's work — BC 3.27).

⁴⁷ Hall, L.B., ‘Ratio and Romanitas in the Bellum Gallicum’, in Welch, K. and Powell, A. (eds), Julius Caesar as Artful Reporter: The War Commentaries as Political Instruments (Swansea 1998) 11–44Google Scholar, at 28.

⁴⁸ Franto, , Ad M. Aurelium, p. 221Google Scholar: … C. Caesarem atrocissimo bello Gallico cum alia multa militaria, tum edam duos de analogia libros scrupulosissimos scripsisse: inter tela volantia de nominibus declinandis, de verborum aspirationibus et rationibus inter classica et tubas. (‘Gaius Caesar, during the extremely savage Gallic War, wrote both a variety of military texts, and also two extremely pointed books ‘De Analogia’ — amidst the javelins he was writing about declining nouns, amidst horns and trumpets, about the aspiration and regularising of verbs’).

⁴⁹ Quint. 8.1.3: etin Tito Livio … putat inesse Pollio Asinius quandam Patavinitatem (‘and in Titus Livius, Asinius Pollio thought that there was a certain “Patavinitas”’). For an overview of Pollio's career and writings, see Morgan, L., ‘The Autopsy of C. Asinius Pollio’, JRS 90 (2000) 51–69Google Scholar. A possible reconstruction of the chronology of the three men considered here is offered by McDermott, W.C., ‘C. Asinius Pollio, Catullus and C. Julius Caesar’, Anc World 2 (1979) 55–60Google Scholar.

⁵⁰ Suet, . De Gram. 10Google Scholar.

⁵¹ Suet, . lui. 56.4Google Scholar. Of course, Caesar's Bellum Civile was competing with Pollio's own civil war history (celebrated in Hor., Carm. 2.1Google Scholar) as an authoritative picture of events.

⁵² Frag. 5-7 [Funaioli].

⁵³ Sen., Contr. 4Google Scholar. pr. 11 : magnam materiam Pollionis Asinii iocis.

⁵⁴ Sen., Contr. 2.3.19Google Scholar: hoc belle deridebat Asinius Pollio.

⁵⁵ So, at 12.1.22, the criticisms of Cicero by Pollio and his son are described as ‘even hostile in many places’ (etiam inimice pluribus locis).

⁵⁶ The manuscripts read nemo est sinum (O) and nemo sinum (X); mnemosynum is a Renaissance suggestion (by Calphurnius in 1481). Trappes-Lomax, J.M., Catullus: A Textual Reappraisal (Swansea 2007) 63CrossRefGoogle Scholar, argues that the text should actually read mnemosynon: “This is an exotic Greek word making its only appearance in Latin literature; Catullus would not have spoiled the effect by giving it a Latin ending and pretending that it was naturalised in Latin.’ This is certainly possible, although the MS corruption nemo sinum is more easily derivable from the Latinised form mnemosynum (though again, it is hard to know how many stages of corruption the MS have undergone).

⁵⁷ Immerwahr, H.R., ‘Ergon: History as a Monument in Herodotus and Thucydides’, AJPh 81 (1960) 261–90Google Scholar. at 266–75 surveys the use of the word in Herodotus.

⁵⁸ (A.P. 12.68.7)Google Scholar.

⁵⁹ A.P. 5.136.4Google Scholar.

⁶⁰ So, Krostenko (n. 35) 245; Fitzgerald, W., Catullan Provocations: Lyric Poetry andthe Drama of Position (Berkeley 1995) 265Google Scholar n. 20.

⁶¹ On aestimatio as a legal term, see Krostenko (n. 35) 244. On Catullus' substitution of monetary for emotional value in Poems 12 and 13, see Nappa, C., ‘Place Settings: Convivium, Contrast and Persona in Catullus 12 and 13’, AJPh 119 (1998) 385–97Google Scholar. at 394.

⁶² 11.10: Caesaris visens monimenta magni.

⁶³ Scotton, C.M., ‘Code Switching as Indexical of Social Relations’, in Heller, (n. 3) 151–86Google Scholar. at 156.

⁶⁴ So, notably, salsas (4) — a live metaphor at a dinner party.

⁶⁵ Or perhaps ‘defined’ is too strong: as Fitzgerald (n. 60) 96 reminds us, this discourse is all surface; something is in venustas because the Catullan ego says it is. Without any stable referents underpinning it, the use of this vocabulary is itself a kind of performance.

⁶⁶ Neudling, C.L., A Prosopography to Catullus (Oxford 1955) 12Google Scholar, makes the case that Marrucinus is part of his name; for the view that it is merely an insulting nickname, see Clausen, W., ‘Catulliana’, in Horsfall, N. (ed.) Vir Bonus Discendi Peritos: Studies in Celebration of Otto Skutsch's Eightieth Birthday (London 1988) 13–17Google Scholar, at 13. On the Marrucini, see the sources collected in Salmon, E.T., ‘Notes on the Social War’, TAPA 89 (1958) 159–84Google Scholar. at 173-4. Stereotypes of this Oscan-speaking tribe seem to persist doggedly despite the social success of their descendants. Cicero might commend them for their nobilitas and dignitas at Chi. 197, but Strabo 5.4.2 memorialises them as a small, war-like, village-dwelling tribe.

⁶⁷ Note the emphasis Catullus puts on the foreign provenance of the napkins in describing them: they are sudaria Saeteba ex Hiberis (line 14).

⁶⁸ See above n. 56.

⁶⁹ As observed by Clausen (n. 66) 13-14.

⁷⁰ Itaparvae res magnum in utramque partem momentum habuerunt (BC 3.70).

⁷¹ Cf. Uden, J., ‘Impersonating Priapus’, AJPh 128 (2007) 1–23Google Scholar, at 13 on Catullus’ use of pupulus in Carmen 56 certainly to refer to an adult. As in that passage, puer has a condescending and perhaps even derogatory tone: McDermott (n. 49) 55-6.

⁷² Charisius (Grammatici Latim [Keil] 1, 94): ‘One should call them “pugillares” [writing-tablets], both in the masculine gender and always in the plural, as Asinius says against Valerius, since a pugillus contains many tablets stitched together into a series. But that same Catullus in his hendecasyllables more often calls them “pugillaria”, in the neuter [cf. Cat. 42.5]’ (hos pugillares et masculino genere et semper pluraliter dicas, sicut Asinius in Valer<ium>, quia pugillus est qui plures tabellas continet in seriem sutas. Attamen haec pugillaria saepius neutraliter dicit idem Catullus in hendecasyllabis) This passage is discussed in detail in a lecture of Haupt, M. (Wilamowitz-Moellendorff, U. von [ed.], Mauricii Hauptii opuscula, vol. 2 [Leipzig 1876] 67–70Google Scholar), who argues that the reference is to Asinius Pollio and Valerius Catullus. Charisius, who presumably himself added the sentence Attamen … hendecasyllabis in order to illuminate Asinius' remark, must also have believed that the reference was to Catullus, but he may have been mistaken: grammatical commentary on contemporary authors was not common at Rome in that period (Epirota, Q. Caecilius is said at Suet. De Gram. 16Google Scholar to have innovated in this respect in his discussions of Vergil) and no other ancient source identifies Catullus by his nomen alone. It is preferable to see the remark as directed instead at a fellow critic, perhaps the contemporaneous Q. Valerius Soranus, a renowned scholar of (inter alia) word use and etymology. I should also add that the commonly accepted supplement ‘ium lib. I’ of Putschius [Helias van Putschen] for the MS in valer … overshoots the mark, since, as Keil states ad loc, no more than three or four letters could have dropped out, given the size of the MS lacuna.

⁷³ Oksala, T., ‘Zum Gebrauch der griechischen Lehnwörter bei Catull’, Arctos 16 (1982) 99–119Google Scholar.

⁷⁴ The manuscripts have disertas, which makes sense, but the reading is flat and the genitives hard to explain. Thomson, D.F.S., Catullus (Toronto 1997) 240–1Google Scholar, favours diserte pater … but a corruption from pater to puer is unlikely in terms of both paleography and sense.

⁷⁵ Hall (n. 47); Clackson, J. and Horrocks, G., The Blackwell History of the Latin Language (Oxford 2007) 183–228Google Scholar.

⁷⁶ See Henderson, J., Fighting for Rome: Poets and Caesars, History and Civil War (Cambridge 1998) 48–55Google Scholar.

⁷⁷ Caesar calls himself imperator throughout the Bellum Civile, but, for others, imperator is presented as a loaded, even dangerous word: see 2.26, 32 (Curio rejects being saluted as imperator); 3.31 (Scipio styles himself imperator in a bid for authority after some losses); 3.71 (Pompey acclaimed as imperator but wary of using the term in litteris); 3.96 (Pompey tears off his insignia as imperator in flight).